The matrix of language: Contemporary linguistic anthropology Ed. by Donald Brenneis and Ronald H. S. Macaulay (review)

BOOK NOTICES 907 lar approach and formalism. This book stands well on its own as a way of teaching formal approaches to language in general, if one is willing to overlook the fact that GPSG is notnecessarily the mostpopular or widely studied syntactic formalism. B weaves smoothly between the introductions of formal principles and notation conventions and their application, typically at a rate of one or two new principles per chapter. This leads him quickly from a description ofphrase structure rules (from approximately first principles) into powerful and arcane notations by Ch. 9 or so. Each addition to the formal toolbox is well-motivated and well-accompanied by example sentences, usually in English, that the principles -as-yet-explained cannot easily describe. This focus on English is jointly one of the strengths and one of the weaknesses of this book; because the examples are primarily drawn from English, it is easy for English-speaking students to understand and follow , but it also gives the impression that GPSG can only describe English syntax. The formal notation is perhaps more detailed and difficult to follow than necessary; a particular problem is B's decision to use Gazdar's integers to represent subcategorization frames. Although these integers are historically accurate, B does not provide any reference tables to remind the student what (e.g.) SUB CAT 2 means. More broadly, this lack of reference tables and appendices is the primary weakness of this book. Without this material the book runs a risk of being nearly incomprehensible after the course finishes. Despite the careful, gradual buildup to a detailed formal analysis of English, B presents no complete grammar or capsule summary of the principles for a harried linguist to flip through. Even when reading the book, it can be disconcerting to find an unfamiliar acronym that one vaguely recalls from two chapters ago and no handy reference chart in the backto summarize what, exactly, 'Foot Feature Principle' means. After six months away from this book, the final chapters might be nearly opaque. These flaws are compensated for by a couple of well-written and all-too-short chapters at the back of the book answering the most important questions for a student—How does GPSG relate to other syntactic theories? and Why GPSG in particular? The wellknown computational advantages of GPSG are briefly discussed along with an overview of some of the proposed solutions to the equally well-known weaknesses. Specialists may be disappointed to find less help than they had wished about computational solutions to their particular problem. Still, B does a good job of introducing linguistics students to the needs of computing, introducing the computer science students to the problems of language formalisms , and building a much-needed bridge across the gap between these two camps. [Patrick Juola, Oxford University.] The matrix of language: Contemporary linguistic anthropology. Ed. by Donald Brenneis and Ronald H. S. Macaulay . Boulder, CO: Westview, 1996. Pp. 338. This book provides a detailed view of the discourse -centered approach cuoently dominant in linguistic anthropology. Although inevitably many worthy pieces were omitted—most noticeable is the near absence of material on African-American linguistic practices, and sexuality is not addressed at all—the collection is generally balanced and rich with interconnections. Shirley Brice Heath opens Section 1, on language socialization, with a comparison of literacy across class and race lines in the southern United States. She stresses the cognitive and socializing aspects of narrative, as do Elinor Ochs, Ruth Smith, and Carolyn Taylor in their chapter on joint problem -solving in the dinnertime naoatives of European American families. Steven Feld and Bambi Schieffelin provide a non-Western perspective and move from narrative to other discourse genres in their description ofKaluli metalinguistic knowledge and language socialization. Section 2, on gender, is the weakest, emphasizing the much-criticized comparative approach over recent alternatives. Daniel Maltz and Ruth Borker's widely cited essay on female and male 'cultures' is balanced by Elinor (Ochs) Keenan's early analysis of the cultural specificity of gendered language use. Penelope Eckert uses data from her study of white suburban teenagers to argue that gender variation is the result of power not prestige. Despite its importance , the...