This paper attempts to explore some of the implications of viewing language learning as a process that deals with a product that is not static. Language, we can increasingly see, is not a set of definable and prearranged tokens but a process of linguistic and cultural negotiation of meaning. Students have to develop their innate ability to use language for their own purposes in interaction with the purposes of other language users. Communicative syllabuses have been discussed partly through a recognition of this need, but a discussion of syllabus organisation is not enough, for the methodological implications are much more important. If languages are not, in principle, discrete, language teaching will not be served well if it is based on the assumption that we can specify the learner's product exactly. All syllabuses tend to concentrate on the product, or content. This paper argues that it is time to move away from syllabus definition, which has been adequately discussed for the current state of classroom practice.

A communicative approach to language learning necessarily entails a commitment to reality. While it is quite possible to teach language as a system in a completely self-contained way, once we begin to try to teach language for use we are inevitably involved in the complexity of human interaction. We focus not on what learners know about the language system but on what they can do with it in the real world beyond the classroom.

It is generally accepted that the aim of second or foreign language teaching is communicative competence; after a sequence of courses, students should be able to use the language (approximately and in certain fields of communication) as native speakers do. But how, for that matter, do native speakers use their language? An exhaustive answer to this question would certainly demand more than one paper. Attention will therefore be drawn to only two points that are important for the problem under discussion: correctness and communicative value.

The development of the communicative approach to language teaching and language learning, started only a few years back with the pilot work done in several centers for research in applied linguistics and by the language project group of the Council of Europe, has now reached a stage where it can safely be said that the pace has been set, and that a return to a strict linguistic or structural approach is not likely to occur. This does not mean, though, that all is for the best in the best of all possible language teaching worlds: all the pedagogical issues raised by the new approach have not been solved, nor thoroughly thought out yet, and our current practice, still in its exploratory stage, is largely based on intuitions and rules of thumb.

During most of the present decade the foreign-language teaching profession in the United States has devoted prodigious efforts to the creation and publication of classroom exercises to develop the learners' ability to communicate. While the resulting diversity of activities bears testimony to the ingenuity of foreign language educators, it also manifests the lack of a coherent system for specifying objectives and creating or selecting appropriate activities for practice. That ubiquitous term “communicative competence” has lacked an operational definition. The concept of functional syllabus design, founded on the identification of the sociolinguistic factors that comprise communication events, suggests some means to fill that void. Once the learners' communication needs have been ascertained, inventories of functions, notions, and keys should be valuable tools for selecting appropriate linguistic exponents and generally determining course content.

In a communicative approach to language learning, the learner soon has to cope with a fundamental aspect of pragmatics: the argumentative and polemic function of language. From a pragmatic perspective, language is a tool we use not so much for describing or representing the world but for changing it, not for exchanging feelings of empathy, but for saying what we like and what we don't like, not just for expressing quiet satisfaction but for saying what is wrong, not for saying what we already know but for asking about what we want to know, not for explaining what we already have but for complaining about what we still lack and want to have: and for getting things done which would not be done if we did not ask for them and if we did not say why and how they have to be done.

As in most other countries, the interest in communicative language teaching has been considerable in Canada. This interest has influenced curriculum development, materials production, experimental teaching, and approaches to language testing; and a good deal of experience has been reported. As everywhere else, there is in Canada today a need to clarify what is meant by communicative language teaching (CLT). In paper appearing in the proceedings of the 1978 AILA Congress in Montreal, the present author has attempted to distinguish between the psychopedagogic and the sociolinguistic concepts under which CLT has been understood (Stern, 1978). As will be seen, the distinction underlies this discussion.

Language is not simply a formal system of sounds, words, and syntactical structures; language also reaches into the domain of human interaction, which for its own part follows certain rules. Every native speaker assimilates individual social experiences characteristic of his own culture. These experiences inhere in statements that obtain their communicative significance through interpretation: “Die Bedeutung eines sprachlichen Zeichens kennen heisst wis en, wie es verwendet werden kann, d.h. wie man mit ihm handeln kann, welche Regeln für seinen Gebrauch gelten” (Heringer, 1977). Each society accumulates rules according to which concrete statements are interpreted abstractly and which are valid among communicating partners through common usage.

Most current foreign language teachers and those about to enter this profession have themselves learned the foreign language by structural methods. The curricula under which they studied are not likely to be exchanged for communicative curricula, as such decisions are not made at the classroom level. Because of these conditions, my paper will concentrate on pragmatic problems of a communicative methodology – problems that arise in attempting to answer the question: does the functional-notional approach, without being either a curriculum or a method, and while operating under the constraints of the present school system, help develop a framework and a set of principles that will help teachers make foreign language learning more communicative than it is at the moment?

As Kelly (1969) reminds us, the development or an art – and foreign language (FL) teaching is assuredly an art rather than a science – is cyclical. Viewed from a short-term perspective, say, the last two decades, FL teaching doctrine appears to have taken a dialectical course: the audiolingual method may be viewed as an antithetical reaction to grammar-translation-reading (GTR). In turn, the so-called cognitive-code approach of the late 1960s constitutes a rejection of overemphasis on outward form and of the behavioral model of learning. Finally, stress on communicative competence developed from a broadening of the narrow focus on linguistic structure shared by both structural linguists and generative-transformational grammarians.