BOOK NOTICES 777 'deceptive fluency' of second-language learners who have progressed beyond the initial stage can be applied verbatim to the skills of many bilingual students in the U.S., even at the university level: 'They are unable to sustain a piece of narrative or descriptive writing, or if they do so, lack flexibility and variety in the syntactical forms and in the vocabulary they have at their disposal. Such limitations may pass relatively ignored in speech; in writing they are at once apparent.' Such students reach even our best universities, where the cruel task of freshman composition programs is to waken them to the truth about their lack of competence in written English, and to try to provide them with the linguistic skills they will need in a demanding academic setting. Anyone engaged in this endeavor (or troubled by the need for such efforts) will respond sympathetically to D's account. D has drawn on American research and experience, particularly in her section on bilingual education—the provision of instruction in the lower grades in the home-language of the students as well as in English. She is quick to point out differences in the two nations' problems : Britain is faced with a large number of recent immigrants, while the U.S. has significant numbers of native-born speakers oflanguages other than English. (The number of Celtic speakers in Britain cannot, e.g., be compared with the number of Spanish speakers in the U.S.) She comes to the interesting conclusion, however, that Britain can learn much more from the American experience than it can from its partners in the European Economic Community, who have also had to face the problem of students speaking foreign languages in their schools. Thus West Germany and France encourage schools to provide mother-tongue instruction to the children of foreign workers because it is assumed that these children will return to their homelands. Britain, however, must accept (even if unwillingly) that the large numbers of non-native speakers of English in its schools are likely to spend the rest of their lives in Britain, as British citizens. The Gujerati speaker from India and the Greek or Turkish speaker from Cyprus is thus an immigrant rather than a migrant, and so comparable to the Mexican newcomer in Los Angeles or the Hong Kong arrival in San Francisco. D's up-to-date and concise bibliography (some items as late as 1976) provides a useful guide to relevant publications, particularly those in lesser-known British journals that may well be unfamiliar to some American investigators. All in all, this little publication provides a valuable view of a familiar problem from an unfamiliar viewpoint. [Phyllis Brooks, Berkeley.] Linguistic theory: what can it say about reading? Edited by Roger W. Shuy. Newark, DE: International Reading Association, 1977. Pp. x, 185. This collection of articles, written by linguists, is meant to provide an introduction toresearchinlinguistictheory forprofessionals in the field of reading, and to suggest areas of common interest. The volume has five sections, each containing two articles : in most cases the first article is a survey of the theoretical field, suggesting applications to research and teaching on reading, while the second follows up with specific proposals for classroom work, research proposals, or applications of the work outlined in the first article. The contributors are Don Larkin, with Yetta Goodman & Jennifer Greene, on grammar; Walt Wolfram, with A. F. Vaughn-Cooke, on phonology; R. W. Shuy, with Barbara Horvath , on sociolinguistics; Bruce Fraser, with Peg Griffin, on pragmatics; and Joel Sherzer, with Ray McDermott, on the ethnography of speaking. The survey articles offer uneven coverage of their respective subfields. Thus Wolfram's essay is devoted exclusively to a sketch of the received view of generative phonology (despite its title: 'Generative phonology: a basic model for reading'). Fraser's article on pragmatics is a presentation of his analysis of conversational competence—which, while interesting in itself, is hardly a survey of pragmatics . The other survey articles tend to fall between these two approaches. This disparity, probably unavoidable to some extent, will surely affect an uninitiated reader's perspective on the value of these surveys. A different sort of problem...