In the last decade, language ideology has become an important topic of research in linguistic anthropology. This fine collection significantly elaborates on work that was first presented in a "Language Ideology" session at the 1991 annual meeting of the American Anthropological Association and that was published in preliminary form in 1992 in a special issue of the journal Pragmatics. In subsequent years, at conferences in Santa Fe, Chicago, Atlanta, and Washington DC, many of the same writers developed their respective themes further. The volume under review seeks to incorporate the thoughts presented at these many different meetings and conferences.

One important source of the concept’s appeal lies in how language ideology systematically links form with function. As Woolard points out in her useful introduction, authors vary in the relative emphases they place on form or function. One approach to the topic draws from Michael Silverstein's metapragmatics and examines implicit and explicit commentary on and signaling about language-in-use. Another approach focuses on languages in contact and the resulting ideologies of "purism" and "standardization." A third line of inquiry examines the influence that linguistic theories and social movements have had on each other. In most of the essays under review, the social function of language is the clear focus of interest, as the implicit audience for this volume is sociocultural anthropologists and linguistic anthropologists who are interested in the social functions of language use.

The concept of language ideology, however, also has much to contribute to the study of linguistic form--metapragmatic and otherwise. Studies of language change, for instance, have tended to focus on mutations in phonological, morphological, and syntactic shapes but have paid relatively little attention to the social functions that those shifts might play. But some structural changes are difficult to explain without bringing the concept of language ideology to bear. A stunning example is the disappearance of the once-familiar thou in English. Its use by the egalitarian Quakers to address everyone, justified on rationalist grounds as numerically precise (although sometimes honorifically problematic), provoked a backlash in 17th-century England to the point that by 1700 the shift to you was completed. Without linguistic ideology, this change is hard to explain, given the practical difficulties it must have created for reference as well as the generally conservative nature of pronominal systems. One of the great benefits of the concept of linguistic ideology is that it links language structure with the social and moral interests of the speakers.

The essays in this volume deal in a variety of ways with the social and moral interests of speakers. The first group of essays examines the scope and force of dominant conceptions of language in selected African, Mexicano, New Guinean, and Arizona Tewa societies. Judith Irvine's discussion of honorifics among Zulu, Wolof, and ChiBemba (and also Javanese) considers the hypothesis that the presence and form of honorifics can be explained with reference to features of social structure--social stratification and the presence of court systems, for example. Examining the historical and sociological data, Irvine finds this explanation insufficient and argues that grammatical honorifics in these very different societies are all accompanied by ideologies about the importance of flattened affect, conventionality, and the avoidance of engagement with the concrete or sensory.

Jane Hill finds that the speakers of Mexicano (Nahuatl) who are most nostalgic for the past are the ones who are, ironically, the most hispanicized; the ones most likely to dispute the nostalgic vision of a past in which respectful people spoke Mexicano are those who speak it now. But where Irvine seems to suggest a sort of analytical autonomy to the notion of language ideology, Hill seems to imply that Mexicano language ideologies might be linked to the political economy of disenfranchisement experienced by Mexicano speakers over the last century or more.

Don Kulick finds that in the tiny Papua New Guinea village of Gapun, despite villagers' insistence that they value their local language, Taiap, it is rapidly disappearing because of local ideologies that associate it with anger, femininity, vulgarity, and self-display. Here is a striking example of the multiple layerings of ideology.

Kroskrity's piece introduces the useful concept of "strict compartmentalization"--an ideological proscription by Arizona Tewa speakers against the mixing of sociolinguistic varieties. "Just as ceremonial practitioners can neither mix linguistic codes nor use them outside of their circumscribed contexts of use, so--ideally--Tewa people should observe comparable compartmentalization of their various languages and linguistic levels in their everyday speech" (pp. 109-110). For example, the mixing of Tewa with either English or Hopi is explicitly devalued.

A second group of essays focuses on the operations of language ideologies in institutions of power. Elizabeth Mertz audiotaped the first semester of contract law classes in eight different law schools. Although she does not attempt any quantitative generalizations (despite the existence of what must be a very large corpus of data), she nonetheless notes the way in which professors use language to force students to adopt role playing in which there are "right" and "wrong" answers, and in which hesitation, uncertainty, and silence are not permitted. The professors’ tactic socializes students for life in a legal system that constructs only "winners" and "losers" and in which people are translated into roles (plaintiff or defendant) and their actions into legal categories (tort, breach of contract). Debra Spitulnik examines the ideological presumptions guiding the state allocation of radio air time among Zambia's multiple languages and discovers that beliefs about demographic predominance, urban provenience, and linguistic distinctiveness are subtly interwoven with official ideologies of egalitarian pluralism. Jan Blommaert and Jef Vershueren's essay on the role of language in European nationalist ideologies makes the intriguing observation, based on the authors’ reading of German, Belgian, and French newspapers, that in Europe, multilingualism is acceptable and praiseworthy as a feature of an individual but is a more dangerous and corrosive force when part of a nation or society. They argue that the media they surveyed overwhelmingly assume an ideology of one nation, one culture, and one language. Their sample was small, however, and their conclusion might have been stronger had they reviewed, for the sake of contrast, some of Europe’s minority-language newspapers.

A third group of essays focuses on the collision of multiple ideologies of language in particular contexts. Charles Briggs argues in his essay on the Warao of Venezuela that there is no single ideology of language that encompasses the discourse of the senior male curers, women, younger men, and older men lacking the status of curers. He shows how, for example, the discourse of the curers, in particular, devalues and deflects the speech of challengers as a form of gossip. James Collins provides a vivid account of how his commitment to professional linguistics nearly ran afoul of his Tolowa friends' efforts to produce their own grammar as a tool for preserving--indeed reviving--their native language. Joseph Errington uses the context of an extraordinary Javanese Language Congress held in 1991 to reflect on Indonesian ideologies of language development. On the one hand, all participants supported the egalitarian, nationalist sentiments that gave rise to the national language; on the other hand, much concern was expressed over the decline of (relatively esoteric) courtly and refined politeness and literary forms. Like Errington, Schieffelin and Rachelle Charlier Doucet discuss elite commitment to rare but cherished language usages; in the case of Haitian kreyol, conflict arises over what "real Haitian creole" is and how to write it down: Is it a dialect of French? Or is it an autonomous language of the masses, and if so, which one (e.g., rek, swa, gwo?). Susan Gal concludes with a particularly clear and concise summary and an intriguing proposal outlining some general features--iconization, fractal recursivity, and erasure--by which forms and functions interlock in language ideologies to produce linguistic and social change.

Although many of these important essays have appeared in print elsewhere, their publication together in a volume in the distinguished Oxford Studies in Anthropological Linguistics makes it more likely they will reach the wider audience they deserve.