This fifth volume of the Annual Review of Applied Linguistics (ARAL) returns essentially to the broad review of Applied Linguistics with which the series opened in 1980. It is the conception of the Editorial Directors that the broad field of Applied Linguistics changes rapidly enough to justify a more general survey every four or five years. It has become the practice of the Editorial Directors to publish more targeted volumes in the intervening years; thus, while the first volume of ARAL attempted a broad coverage (i.e., Bilingualism, Pidginization and Creolization, Structural-Cognitive Pedagogical Approaches, Notional/Functional Approaches, Computer-Assisted Instruction, Psycholinguistlcs, Error Analysis, Language Testing, Sign Language, and Sociolinguistics), volumes II, III, and IV were more narrowly focused-i.e., respectively on Language and Language-in- Education Policy, Written Discourse, and Literacy. This volume attempts again a broad coverage of the whole field.

Research within the domain of applied linguistics is at the moment lively and vigorous. Exciting work is being conducted in many settings by scholars trained in diverse areas of specialization. Certainly the inclusion of separate strands in (at least) psycholinguistics and sociolinguistics is now obligatory in most graduate programs in linguistics, bilingual education, foreign language education, or English as a second or foreign language, whereas the subjects were just beginning to be introduced two decades ago. The vitality of the Linguistic Society of America (LSA), the American Council on the Teaching of Foreign Languages (ACTFL), Teachers of English to Speakers of Other Languages (TESOL) and other organizations bodes well for the strength of our discipline. Annual or semi-annual meetings of each of these groups are well attended and offer broad, stimulating programs for participants. These organizations directly or indirectly sponsor a range of scholarly publications widely distributed to a broad membership.

This survey is primarily an update of the review published in ARAL I, (Kaplan 1981:2-24). The research on various aspects of bilingualism during the last five years shows three main characteristics. First, there is questioning of some basic concepts which are still considered, as it were, sacrosanct by researchers for the description and analysis of bilingualism--individual and societal--and for actual fieldwork (cf., section 2 below). In such questioning--however mute at present--several theoretical and methodological sacred cows are under attack. Second, one notices a shift in the areas of research towards experimentation, with more precise methodology and techniques, to answer questions concerning the bilingual's brain (cf., sections 5 and 6). Third, there are insightful breakthroughs in crosscultural and cross-linguistic research with serious applied orientation (cf., sections 10 and 11) and, equally important, there is concern for a social commitment in such research. True, in this intense research activity there are very few questions asked which may be considered as breaking new ground. However, the newness lies in the answers which are provided to old questions. In these answers, we notice many fresh insights gained through a wealth of cross-cultural data, through the results obtained from longitudinal studies (e.g., Hakuta and Diaz in press, Rosier and Holm 1980), through the experiments conducted with highly refined and sophisticated tools and techniques of data collection and data analysis, and through increasing understanding of the bilinguals' and monolinguals' neuropsychological processes.

Language can be acquired by means of the writing system, directly, without the medium of either speech or sign. Deaf children can acquire written language through an association of written forms with environ- mental objects and events, just as hearing children acquire language through an association of speech sounds with environmental experiences. This article considers in detail the rationale which underlies using written language as a native or first language for children who are severely or profoundly hearing-impaired. A perspective in terms of historical and current ideas concerning such theory, including the views of Alexander Graham Bell, is provided.

In 1980, in this journal, Joshua Fishman presented the major theoretical issue in sociolinguistics as being the link between microsociolinguistic and macrosociolinguistic processes. In 1984 that is still the case, although the issue is receiving more explicit attention than it did four years ago.

There are two branches of sociolinguistics which approach this issue in different ways. These two branches are interactionist and variationist sociolinguistics. Interactionist sociolinguistics is principally interested in what language use can tell us about social processes, and therefore a central concern is the social meaning of language use. Variationist sociolinguistics is interested in accounting for linguistic variation and change, at least partly as a product of the social distribution of language varieties. It is, therefore, less concerned with meaning as process, and more concerned with the interaction of linguistic and social systems; in this view the significance of language is mainly symbolic. In this review, I will discuss the contributions of these branches to the problem of the relationship between microsociolinguistics and macrosociolinguistics, as well as the theoretical problems peculiar to each branch.

In a retrospective survey, it is helpful to use the wisdom of hindsight to impose neat categories and descriptive labels on the movement of professional ideas. The past is thus made to make much more sense than when it was the present. In terms of language pedagogy, the years surveyed in ARAL I (the late seventies) might be characterized as the era of the functional/notional syllabus. If one is seeking a similar global label to characterize the period since then, one might well speak of the return of methodology

In 1980 Frank Otto wrote the last entry for this series on computer aided instruction (CAI). A great deal has happened in the field since then. This brief paper will discuss several trends and issues that have recently become clear and attempt to describe the state of the field at present. The reader should be forewarned, however, that the situation is changing very rapidly, sometimes daily, and that what is true as of this writing may not be true tomorrow. The entire field of computers is changing very rapidly.

An essentially interdisciplinary activity, like many areas of applied linguistics, the most immediately contingent area to stylistics remains that of literary studies, although recent years have witnessed extension into other domains such as lexicography and teaching English as a Foreign Language. This survey is divided into five main sections but, given the interrelatedness of the areas, there will be inevitable overlaps as well as potential cases for sub-categorization. The sections are: 1) Linguistic stylistics 2) Literary stylistics 3) Style and discourse A) Pedagogical stylistics 5) Stylistics and the foreign language learner.

In the relatively recent past, a whole issue of ARAL (Kaplan, et al. 1983) was devoted to text analysis. In a way, an article of the sort offered here may be premature in the sense that it appears only two years after the more thorough review volume. There are two reasons why it seems appropriate to look at this area again so quickly. First, the area of text analysis is an extremely active one, and no major bibliographies have appeared since 1983. Second, the approach taken here is somewhat different from that in ARAL III; this review has no direct concern either with pedagogical applications or with contrastive forms of text. There is, of necessity, some overlap in bibliography between this article and the earlier volume, but the great majority of sources cited did not appear there. This particular area is in a state of very rapid growth, moving toward the evolution of a new paradigm which may radically modify the field of applied linguistics by providing a new set of questions and a new way of answering them. The rate and direction of change seem to justify another brief sortie into text analysis, and thereby to try to disambiguate some notions that have been available for greater or lesser periods of time.

There is no doubt that lexicography is an exciting specialization, full of promise, and capable of radically new developments. A review of lexicography in the 1980's cannot hope to convey a comprehensive picture of everything that is going on, but I shall try to pick out a few representative trends and issues. There are very few precedents to follow (the survey of Quemada (1972) covered more ground, while concentrating on historical lexicography in Western Europe to the late 1960's), but it seems that I can at least try to do three things: