Unlike first language acquisition, second language acquisition [SLA] is a phenomenon neither universally experienced nor uniformly successful. Yet an assumption underlying all of SLA research is that the phenomenon, whether aided by formal instruction or not, is guided by certain principles. The discovery of these principles is the over-riding goal of SLA research.

It is quite evident that the study of phonological issues within second language acquisition [SLA] research has been vitalized in recent years and is attracting wider interest today than it did a decade ago. This can be seen in the major journals in the field, and also in the fact that compensive anthologies solely devoted to the phonological domain of SLA research have begun to appear (Ioup and Weinberger 1987, James and Leather 1987). Since this review is intended to concentrate on works from the last few years (mostly from 1986 on), it is worth pointing out that this period does not in itself mark a radical shift or reorientation towards new research issues. Rather, it is mainly a continuation of topics that originated a few years back, and a consolidation of a field of research which is gradually becoming more established.

The early development of second language acquisition [SLA] research in the area of morphology and syntax can be traced in some excellent publications both exemplifying and evaluating the state of the art in this area (e.g., Anderson 1981, Davies, Criper, and Howatt 1984, Hatch 1983, Wode 1981b). The research questions identified and pursued at that time have continued to occupy researchers. (The ones which are the most important and relevant for the discussion here are as follows: 1 the problem of regularity in interlanguage [IL] morphology and syntax; 2) the contribution of Universal Grammar to SLA, and 3) variability in IL performance. The problems enumerated above will provide a framework for the discussion of the acquisition of morphology and syntax. Some additional comments on other, related issues will be included.

Semantic aspects of learner varieties have only recently gained the interest of second language acquistion [SLA] researchers. This is partly a consequence of some shifts in SLA theory and partly due to the discussion of “form” and “function” and the notions “pragmatics” and “semantics” in linguistic theory. There are two corollaries in recent SLA research: interlanguages should be described on the basis of their use in real communication situations, and the process of learning (strategies and stages in course of acquisition) should be the focus of empirical work. (For a discussion of process and product cf., Dittmar 1984.) As a matter of fact, the present standard paradigm in SLA is the description of learner varieties “outside the classroom” with a focus on “learning without explicit teaching” (also known by the somewhat misleading term undirected learning ) under natural conditions of communication with an emphasis on cross-sectional studies in the seventies (cf., Klein and Dittmar 1979) and on longi- tudinal research in the eighties (cf., Perdue 1982).

Over the past ten years, conversational analysis “CA” has come to wield a significant influence on second language “L2” acquisition and teaching. It originally developed out of a school of sociology called ethnomethodology, develped by Garginkel (1967). Following the work of Garfinkel, Sacks and his colleagues, Schegloff and Jefferson, establised CA, the study of the structural organization of ordinary conversation,. Sack' lecture notes (1964–1972), which comprise appeoximately 2,000 pages, continue to provide a foundation for contemporary CA. In explaining why he developed a framework for CA, Sacks states:

Second language acquisition research has historically been influenced by the field of linguistics, not only in terms of theoretical inisghts, but also in terms of the emphasis placed on varioustopics. Within linguistics, the lexicon has taken on a secondary role in terms of theory construction; within second language acquisition, the rift between studies of the lexicon and theory construction seems to be even greater. While it is no longer possible to say that the lexicon is the ‘neglected component’ of second language acquisition research, it does appear to be the case that most lexical studies are not centrally concerned with the establishment of a theory of the lexicon; rather, the majority deal with descriptive aspects of the lexicon.

When the learner of a non-native language actually uses this language for communication, he draws on the L2 skills acquired so far, and on various other sources of knowledge and skill that are relevant to the communicative task. The more advanced the learner is, the more s/he will be able to rely on his/her L2 skills in a way that is analogous to the L1 speaker's use fo his/her individual linguistic competence. The less advanced learner will have to rely more on his/her metalinguistic knowledge (to the extent that it is available) and on a variety of ways to reduce the gap between, on the one hand, the L2 knowledge and skills available to him/her and, on the other hand, the means necessary to convey the intended message, either by using additional resources (L1, gestures, and so on) or by reducing the message to be conveyed. The term communicative processes [hereafter CP] will be used for the reliance on L2 knowledge and skills, and the term communicative strategies or communication strategies [hereafter CS] for the learner's attempts at bridging the gap between these resources and the message to be conveyed. (There is no clear-cut separation between these two mental activities, of course, as the use of CS encompasses the incorporation of whatever relevant fragments of L2 knowledge and skill happen to be available.)

The role that input plays in second language learning and the interactions between learners and other speakers of the language which provide some of that input are two of the most discussed topics in the contemporary field of second language acquisition [SLA]. Several surveys of the considerable body of literature on input and interaction have appeared recent books on SLA (Hatch 1983a ch. 9, Ellis 1985a ch. 6, Young and Doughty 1987, Chaudron 1988 chs. 3–5) and two collections of papers on input and interaction research have been published (Gass and Madden 1985, Day 1986). This essay will outline the main issues which have emerged form this scholarship and comment on some of the more important theroretical and empirical papers.

The notion that attitudes and motivation would be implicated in second language acquisition is not a new one. As early as 1941, Jordan investigated the relation between attitudes toward a number of school subjects and grades in those subjects, and found the relationships for French to be among the highest. A number of later studies by other researches also showed relationships between attitudes towards learning languages and proficiency in the language (see Gardner 1985 for a review). The first reference to a possible relationship between attitudes toward the other language community and achievement in that language, however, appears to have been made by Arsenian (1945). One of the many relevant questions he raised, for example, was, “In what way do affective factors, such as social prestige, assumed superiority, or—contrariwise—assumed inferiority, or enforcement of a language by a hated nation affect language learning in a child?” (Arsenian 1945:85).

The goals of second language acquisition research, as of first language acquisi-tion research, are to determine exactly what is acquired and precisely how it is acquired. The first concern leads to questions such as the following in the case of second language acquisition [SLA]

•How comparable are the first and second languages of the learners?

•What effect does the perceived closeness or distance of the two languages have on second language acquisition?

•Do learners only acquire linguistic categories and structures of the second language, or do they also acquire a different set of cognitive structures and cultural categories? And finally,

•Do they acquire only linguistic rules, or do they also acquire the sociocultural conventions of language use relevant to the second language?

The second concern leads to the following question:

•What role do cognitive styles, learning strategies, and personality factors—either innate or acquired as a result of socialization in a particular community—play in second language acquisition?

Inverstigating the effects of classroom instruction on second language acquisition is a fascinating, but formidable, task: fascinating, because it addresses fundamental issues concerning the nature of the mental processes underlying SLA and how they are affected by different learning conditions; formidable, because of the complex, multifaceted nature of both SLA and instruction, the difficulty of establishing valid and reliable profiles of each, and the interacting effects of social context and individual variables. In this brief survey, empiricial research on the effects of L2 instruction is outlined in relation to a few basic issues. For more detailed reviews of the literature, the reader is referred to Long (1983; 1988), Pica (1983a), Ellis (1984a; 1985), and Chaudron (1988).

In an influential paper published in the late seventies, H. H. Stern distinguished between learning a language through a process of conscious study and practice (i.e., formally), or through use in the environment (i.e., functionally). According to Stern (1981), this aspect of language behavior can be characterized as a psycholinguistic/pedagogic continuum, or ‘P-scale.’ There is nothing inherently good or bad about activities at either end of the scale, and in organized language teaching one often finds an interplay between formal and functional approaches.

Language testing [LT] research and second language acquisition [SLA] research are often seen as distinct areas of inquiry in applied linguistics. To oversimplify slightly, SLA research takes a longitudinal view, concerning itself primarily with the description and explanation of how second language proficiency develops, while LT research typically observes a “slice of life”, and attempts to arrive at a more or less static description of language proficiency at a given stage of development.

In an cited essay, Hairston (1982) asserts that a revolution is taking place in the teaching of compostition, a revolution she characterizes as involving a basic shift in paradigms with regard to the act of composition. Hairston describes the prevailing view of writing, a view that had its orgins in traditional theories of rhetoric, as one that considers writing a linear process in which writers know what they want to say before they begin to write. The major task of the attends to editing concerns in order to perfect the manuscript. Thus the focus of composition instruction in this paradigm is the product that the writer produces.

Theories and models of seond language acquisition have tended to focus on the role of oral language. While some acknowledgement is given to the role of reading (andwriting) in SLA (e.g., by Krashen 1984; 1985), the relative inattention given to reading in SLA had resulted in a dearth to reading research which is explicitly tied to the most popular theories and models of SLA or which is expressly designed to deal with issues commonly addressed in SLA. Although there is no simple explanation as to why SLAresearch has focused almost exclusively on oral language, to the neglect of written language, it is interesting to speculate about why the “canonical” theories of SLA do not to any significant extent deal with reading and writing, especially when there seems to be broad consensus that language has to be considered from a textual point of view, and when written as well as oral language may be a substantial source of language input. One possibility is that SLA, while in some sense part of the backlash against the structuralism of the audiolingual approach, has nonetheless simply maintained the focus on oral-aural language of audiolingualism (which itself originated as a reaction against earlier grammartranslation models relying heavily on reading and writing).