In my presentation I argue, not for the first time, the need to rethink orthodox ideas about the relationship between applied linguistics and English language teaching that have been promoted in the past and still prevail. I will do this by taking up issues in questions that I invited colleagues to send me in advance; issues that I have discussed elsewhere, most recently in Widdowson (Reference Widdowson, Wright, Harvey and Simpson2019, Reference Widdowson2020). Therefore, in many ways, this talk is a reformulation of views I have expressed in the past, but which, I would claim, have a direct relevance to the present.

To quote the novelist L. P. Hartley in The Go-Between, ‘the past is a foreign country; they do things differently there’. Its maps are no longer a reliable guide to the present, its roads and landmarks are out of date. And now, more than ever, the vast and unpredictable networks of interaction brought about by digitalization and globalization call into question established concepts and beliefs about almost everything. So, I want to suggest that both the field of applied linguistics in relation to English teaching, and the present orthodox way that English language teaching (ELT) is conceived stand in urgent need of fundamental reappraisal, because they have both been mapped out in ways that are no longer relevant to present-day realities. The issues that I see as related to this general theme, and arising from the particular questions colleagues have posed, are as follows:

1. 1. The definition of applied linguistics: disciplinarity and problem-solving.

2. 2. Applied linguistics and second language acquisition (SLA): academic research and the real world.

3. 3. The nature of SLA: the concept of competence.

4. 4. The relationship between learning and teaching.

5. 5. Conformity, competence and communicative capability.

6. 6. The diverse foreignness of second languages.

7. 7. Specific purposes: ELT and English for specific purposes (ESP).

8. 8. The objectives of learning: training vs. education.

So, to begin with, how has applied linguistics been mapped out as a field of inquiry? I think there is a general agreement that the essential aim of applied linguistics is to bring linguistic expertise of various kinds to bear on problems about language, which people experience in the real world. So, in principle, it is an enquiry into the relationship between expertise and experience directed at practical outcomes.

But it also, of course, aspires to be recognized as an academic discipline with the same prestigious status as any other; it has its journals, its departments, its professors, and so on. And this has an effect on how these problems are actually formulated. What happens, I think, is that they tend to get defined in disciplinary terms to make them more amenable to inquiry in a disciplinary context. Disciplinary research uses data to provide empirical support to theory, and theory is necessarily a general abstraction from actuality. So, of its nature, it is remote from the actual and local problems that are experienced in practice. This, I think, raises the question: How can applied linguistics be both disciplinary and problem-solving and so deal with abstractions and actuality at the same time?

In fact, applied linguistics is not only conceived as a disciplinary, but an inter-disciplinary field of activity. Here, for example, is a statement from an editorial in the journal Applied Linguistics. Although this is dated 2000, it expresses what is still, I think, a prevailing view:

The assumption here is that the ability to engage with real world issues is unilaterally conditional on disciplinarity: that the more interdisciplinary applied linguistics becomes, the greater its practical relevance. I would want to argue exactly the opposite. Here, I make reference to the question put by Dr. Massoud Yaghoubi-Notash of Tabriz University:

• • How can applied linguistics go interdisciplinary? (Is this an intra-theoretical way out of crisis?) If we fail to do so, why not go beyond theories to define a new paradigm, (an extra-theoretical solution), while not losing sight of local elements?

I would argue that going interdisciplinary, even going disciplinary without being ‘inter’, is always likely to result in losing sight of the actual local problems that applied linguistics claims to be able to handle: that we do indeed need to ‘go beyond theories’. This brings me to questions raised by Professor James P. Lantolf of Pennsylvania State University about the relationship between SLA and the practice of language teaching:

• • What do you make of the gap between SLA theory/research and language teaching practice? Can these be unified in some way, or should there be a separate and independent theory formulated for teaching practice?

• • Do you think SLA research has a sufficient body of accumulated knowledge to inform language teaching practice?

• • A related question would be the following: What do you consider to be the accepted/agreed-upon findings of SLA research?

Here, I have to confess that I am sceptical about the practical relevance of SLA research. SLA is generally accepted as squarely located within applied linguistics; indeed, in some quarters the two are assumed to be synonymous. But the focus of SLA is on theory and empirical research. It is essentially an academic inquiry, and as such, as I hinted earlier, the problems that it addresses are formulated in ways that serve its disciplinary agenda. Therefore, SLA sets out to discover underlying generalities of, to cite a well-known book (Lightbown & Spada, Reference Lightbown and Spada2013), ‘How languages are learned’ – whatever the language, whoever the learners.

There is of course a massive amount of SLA research that has accumulated in quest of an explanation of how languages in general are learned and I am not sure what ‘accepted/agreed-upon findings’ have emerged that reveal this explanation. That is something best known to the researchers themselves. The point I want to make is that whatever the findings, they can only reveal that at some abstract level there are generalities in the acquisition of a second language, whatever the second language might be. But no matter how revealing such findings might be, they are, from a language teaching perspective, only of academic interest unless they can be acted upon in practice.

What teachers are concerned with in their classrooms is not SLA in general but with the actualities of how to get particular groups of learners to learn a particular language, which has a local, specific ‘secondness’ or foreignness in relation to their own first languages. Languages are enormously different, in themselves, and in relation to other languages. What SLA is ultimately concerned with is the academic problem of how to eliminate what is incidental, or circumstantial, the intervening variables that get in the way of finding the underlying essential commonalities that define the language acquisition process in general. But of course, it is just such variables that teachers have to cope with. The academic problem of disciplinary inquiry and the practical problem of pedagogy do not correspond.

And here is the gap between theory and practice that I think James Lantolf refers to: a gap that is created as a necessary condition of any academic study; the gap between expertise in linguistics and the experience of language. If applied linguistics is really to engage with real world problems, and not simply be a kind of self-generating academic area of inquiry, then it follows, as far as I can see, that SLA is not applied linguistics at all; it is an example of what I have referred to as ‘linguistics applied’, which brings me to the questions posed by Professor Brian Tomlinson of Anaheim University:

• • What contributions do you think applied linguistics has made to the improvement of language teaching?

• • What do you think applied linguists need to do to make a greater contribution to the improvement of language teaching?

With regard to the first question, as far as the so-called applied linguistics – or as I would say, the linguistics applied of SLA – is concerned, it will be obvious from what I said that I think it has made little, if any, contribution to improvements in language teaching. The findings of its research, presented to SLA insiders at conferences and published in learned journals are generally inaccessible to most language teachers in the real world, and even if they are made aware of these findings, the likelihood of their knowing how they can be acted upon so as to effectively inform their actual language teaching practice seems to me to be remote. There is little evidence, it seems to me, that the initiating of vast numbers of students into the expert mysteries of SLA research in university courses all over the world makes them ‘better prepared for the principled handling of the real world issues’ they have to cope with in the context of their classrooms. Of course, others might well challenge this view and it is, I suppose, the purpose of a webinar of this kind to provoke such a challenge. What then of Brian Tomlinson's second question?

My answer, predictable from what I have been saying already, is that I think the first thing applied linguists need to do is to stop giving primacy to disciplinarity, and focus instead on the real world problems they claim to deal with. In my view, they need to separate the process of second language acquisition as actually experienced from the academic field represented by the acronym SLA. We need instead, I think, to reconsider what the process of learning a particular other, second or foreign language actually entails. So, the central question is: what then does it mean to acquire a second language? Now the orthodox answer is that it means achieving competence in another language.

Competence, for me, is a very problematic concept. The term has become a kind of catch-all term which is all too readily attached to all kinds of phenomena. We have competences of every conceivable kind: linguistic competence, sociolinguistic competence, communicative and cultural competence, discourse and pragmatic competence, and so on. But what concept does the term actually refer to? The first point to make is that it is a disciplinary abstraction conceived as what native speakers in a particular community know of their language and the conventions of its use. It is based on the simplifying assumption that there is a well-defined category of native speakers whose languages and communities are stable and self-enclosed entities. And in SLA, it is this competence that learners are said to acquire, and their success in doing so is measured by how far they approximate to it along an interlanguage scale. Thus, learners are categorized in terms of relative success in acquiring the competence of native speakers.

Even when the concept is extended from linguistic to communicative competence, it is still a disciplinary abstraction. Hymes’ familiar formulation is routinely cited as giving the necessary disciplinary authority for the pedagogic practice of communicative language teaching. For Hymes, the extent to which one is communicatively competent is how far, given a sample of language, we can judge it to be relatively possible in conformity with encoding rules, feasible in this sense of being processible, appropriate to context and attested as having been actually performed. This judgment can, of course, only be made in reference to an established norm, as Hymes (Reference Hymes, Pride and Holmes1972) himself makes clear:

The definition of communicative competence is based on the construct of a distinct community whose normal (which is to say ideal) members have complete knowledge of the encoded features of its language and the conventions of its use. Obviously, otherwise they would not be able to make judgements as to how far a sample of a language conforms to this common norm. And it is this disciplinary construct of competence, linguistic or communicative, that is carried over into the pedagogic domain, the assumption being that teaching learners how to communicate in another language is a matter of getting them to replicate the ways normal members of native speaking communities do it. So communicative language teaching, based as it is on Hymes’ definition of communicative competence, is not concerned with how the potential for meaning making in language can be exploited, with how English or any other language actually functions as a communicative resource, but only with the normative stereotypical form that communication would take in a particular community of idealized native speakers.

Competence, then, whether linguistic or communicative, is an abstract construct defined in reference to the similar abstract representations of languages and communities as distinct self-enclosed entities. Of course, these abstractions have their validity in the discipline of linguistics: they are convenient and necessary abstractions, without which there would be no disciplinary research at all. But how convenient are they for applied linguistics, which, since its concern is with real world problems, has as its primary purpose to be useful?

And validity and usefulness are by no means the same thing. You can have valid theories, which have little if any usefulness and you can certainly have invalid theories which are put to very effective use. The key question here, it seems to me is, useful for what and for whom? Whose problems is applied linguistics seeking to address? Who formulates these problems? The kind of theoretical abstractions that disciplines devise can be very useful in dealing with institutional problems of a sociopolitical kind where it is very convenient to typify individuals and put them into different social or ethnic or religious categories. This is a very useful thing to do; it makes individuals easier to manage and control, and of course, if you can claim disciplinary authority for your categorization, so much the better.

In the case of foreign or second language teaching, the construct of native speaker competence is very useful in that it solves the problem of what is to be taught, and what therefore is to be tested. This is very convenient for course designers, policy makers and publishers, because it gives them something definite to prescribe for teachers to teach, well documented by standard works of reference, and it is, of course, especially convenient in the case of English in that it sustains the highly profitable ELT industry, the promotion of ELT goods and services which it is assumed native speakers are somehow uniquely qualified to provide. Therefore, the disciplinary construct of competence solves a lot of problems, and I suppose that in this respect, one might say that it bears out the applied linguistic claim that disciplinarity provides the means for handling real world issues. But again, we will need to ask, whose problems? Whose reality are we talking about? The competence construct conveniently solves the pedagogic problem of what teachers are to teach and the institutional problem of what is to be tested. But what about the learners? How does this relate to the problems that learners encounter?

It is interesting to note, in passing, that when language pedagogy is referred to, the sequence is almost always teaching and learning and not the other way round: ‘Communicative Language teaching and learning’, ‘Task-based teaching and learning’, and so on. This sequence would seem to imply that there is a necessary, unilateral dependency of learning on teaching. And I would argue that this implication is indeed borne out in the way English, or any other language as a subject on the school curriculum, is conceived.

In the orthodox conception of the foreign or second language subject, learning is taken to involve conformity to what is prescribed as competence; what is learned is given credit only to the extent that it corresponds with what has been taught. If learning does not correspond with what has been taught, it does not count as learning. On the contrary, it is seen as a failure to learn and thus language assessment is based not on what has been learned, but what has been taught and how far the teaching has been successful in shaping the learning. The record of success is, to say the least, not impressive. Despite the teachers’ best efforts and all the various methods and approaches that have been proposed over the years, learners still persistently fail to conform. So why is it that learners so often are not persuaded to do what they are taught to do? What is their problem? It clearly hasn't been solved, or indeed even addressed, by an orthodox competence-based way of thinking which is focused on the teaching objective, and institutional measurement of achievement.

I would suggest that the main problem for learners is essentially one that is pedagogically imposed upon them, because their learning is actually impeded by the very teaching that is meant to promote it. The second, other, or foreign language that they are confronted with is an abstract construct dissociated from their own experience and this dissociation is emphasized by the customary practice of monolingual teaching. This, I think, has the alienating effect of making the foreign language even more foreign, because it doesn't correspond to the learners’ own experience of language. Attempts by learners to reduce the foreignness by relating it to their own language is seen as interference in the approved process of competence acquisition. These so-called errors may be tolerated as interim efforts to conform, but ultimately these learning non-conformities need to be corrected if the objective of teaching is to be achieved.

But in producing these nonconformities, widely supposed to be the defective characteristics of learner language, learners are actually doing quite naturally what all language users do, and what learners have themselves done as users of their own language: drawing on and adapting whatever linguistic resources are available to them to get their meaning across. Learning, in this respect, is a natural process whereby learners seek to link the foreign language to the one they are familiar with by appropriating it as an additional resource in their communicative repertoire. In this way, they seek to familiarize the foreign language, and so authenticate it by making it a reality for themselves.

Meanwhile, the teacher is striving, very often in vain, to get them to replicate the ‘ideality’ as we might call it, of native speaker users, thereby denying this natural learning-in-using process by getting learners to conform to an artificial construct. In requiring them to conform to how native speakers supposedly communicate so as to replicate their competence, the teacher in effect inhibits learners from learning how to use the language as a communicative resource and so develop a general lingual capability. So, what we get is a glaring disparity between natural, multilingual learning and the enforced imposition of monolingual teaching, which effectively reduces learners to what I call teachees. No wonder learners have a problem.

In sum, if applied linguistics is to engage with learners’ problems rather than define them from a teaching perspective with reference to a disciplinary linguistic construct of competence, then we need to consider how a foreign or other second language looks from the perspective of the learners. But we have to be careful of convenient categorization, and here I return to the point I made earlier about second languages being individually different and to what I said about SLA being based on the simplifying assumption that at some level of abstraction all learners learn in essentially the same way, and second languages are all alike in being ‘second’. The fact is that if we consider the real world, and that, after all, is what we are supposed to be doing in applied linguistics, we do not and cannot actually know how languages are learned because all languages and all learners are ultimately different, and we cannot account for the variables.

A second language is foreign, but the nature of its foreignness crucially depends not only on how it relates to familiar first languages, but also on how the second language itself is perceived. I am not only referring to language distance, the extent to which the code of the two languages are different, but more significantly to how a learner sees the role and status of this second language in relation to his or her own. Foreign is defined in terms of what is familiar and this relationship is always going to be different from one language to another. It is obvious, for example, that German is not foreign for speakers of Russian in the same way as it is foreign for Dutch or English speakers. English is not foreign for Iranians in the same way as it is foreign for the French, or the Indians or the Chinese. Foreignness is essentially locally defined.

Therefore, attitudes to the otherness of a particular foreign language will depend on various factors which have to do, in Bourdieu's terms, with economic, cultural, social, symbolic capital. In the case of English, the dominant factor is likely to be taken to be its economic value. The question that arises is how far this can be dissociated from its cultural and social capital as a native language and from its symbolic capital as the language of colonial oppression, past and present, and so serve as a relatively ‘neutral’ means of communicating. The foreign language may be closely identified with the culture of its primary community speakers. This might lend the language a symbolic prestige, or on the contrary, give rise to prejudice against it. Just as individuals variably accommodate to each other, so learners variably accommodate to the second languages they are learning, converging where they can identify with it in some way, diverging where they cannot.

So, how learners conceive of the foreignness of their L2 has an obvious effect on their motivation, and it also regulates what features of the language they are disposed to acquire. If they seek to appropriate a second language by reference to the experience of their own language, as I have argued they do, they would be naturally inclined to focus their attention on those features which seem to have most communicative value. But many of the formal features that the learners are required to learn have little, if any, communicative value: they are conventions of native speaker usage, which like received pronunciation, are deemed to represent the norms of what is socially correct, linguistic comportment. Since learners obviously have no idea what these norms are, for them correctness has no obvious purpose especially if they encounter users of the language who can get by very effectively without it – users they interact with on social media, for example, or celebrities they admire and wish to emulate. These users are likely to be much more real and therefore more influential as role models than an ideal model of correctness which is remote from their experience. It is hardly surprising that teachers have such a hard time trying to impose this ideal model upon them. What and how learners process another language depends on their conception of its foreignness, which necessarily means that the process will be locally variable and context-dependent. This, of course, runs counter to the teacher-imposed, competence-based approach to language teaching that at present prevails. And this brings me to the questions that Professor M. Reza Ataei of Kharazmi University has raised about ESP:

• • Given the highly accountable nature of ESP education and its claims for effectiveness and efficiency, do you predict that ESP principles and practice will apply to all kinds of ELT in the future? If so, what are the promises and the threats?

• • How can ESP/English for Academic Purposes (EAP) education resolve the clash between strict adherence to target language use domain as the major source of information in needs analysis and the critical approach to education?

To my mind, the orthodox approach to ELT that I have been discussing is essentially the teaching of ESP. Its purpose is to teach the English that is specific to its native speakers. Its ‘effectiveness and efficiency’ are measured by how far the learning conforms to this teaching, not how far the learning itself can be put to effective and efficient communicative use: what purposes learners will actually need English for when they find themselves users of the language in the real world beyond the classroom is not considered. One might say that the orthodox approach to ELT is specific, but to no valid purpose.

But then we need to ask what a valid purpose would be, what objective courses in ELT should be designed to achieve. At present, the objective is, as I have said, to get learners to meet the prescribed competence requirement of assessment at the end of the course. What tests measure is how successful the teaching has been. But although the course is the end of teaching, it is not the end of learning. Indeed, one can say that the whole point of teaching is to provide the motivation and momentum for learning to continue and extend beyond the course and beyond the test. In other words, language teaching, like the teaching of any kind, has to have a projective purpose, if it is to have any real value. So, its essential objective has to be not the conformity to competence, but the development of learning as an investment in the ‘capability’ for further learning through using the language in the context of real world communication. In other words, the teaching objective should not be to train learners in the competence needed to pass the exam, but to educate them in capability, which can be used and further developed beyond the exam. Here, I come to the question put by Professor Ken Hyland of the University of East Anglia:

• • Some 40 years ago (I think in your Learning Purpose and Language Use in 1983), you said words to the effect that EAP is essentially a training exercise, as it fails to prepare students for unpredictable assignments and encourages unimaginative and formulaic essays. You contrasted this, I seem to remember, with education which involves assisting learners to understand and cope with a wider range of needs. What changes do you think have occurred in specific language teaching over the intervening years, and have you changed your mind on this point?

As will be clear from what I have been saying, I would still regard the distinction between ‘training’ and ‘education’ as crucial. I am not familiar with what changes have occurred in EAP study and it may well be that over the years it has adopted a more educational perspective. The only point I would make is that the more specific it is, the more training it becomes, and the less likely to develop capability and so ‘to prepare students for unpredictable assignments’ – assignments in the sense of communicative demands made upon them in real world contexts that they will need to ‘understand and cope with’.

Time to come to a close. In this presentation, I have expressed my view of the relationship between applied linguistics and English language teaching by reacting to questions that colleagues have raised. But I have yet to consider the most wide-ranging of the questions sent to me, that raised by the organizer and moderator of this seminar, Dr. Seyed Yasin Yazdi-Amirkhiz of Teheran University of Medical Sciences. This is a particularly appropriate question to end with since it gives me the opportunity to recapitulate the points that I have made earlier and so brings this presentation to a summary conclusion:

• • Looking retrospectively, would you consider revising or revisiting any of your thoughts and contributions to applied linguistics and language teaching? For instance, 1. Applied linguistics vs. Linguistics applied. 2. Competence vs. Capacity. 3. English in training and education. 4. Your conceptualization of EGP vs. ESP. 5. Authenticity of teaching materials in ESP. 6. Language audits (Present situation analysis vs. target situation analysis).

With regard to the distinction between applied linguistics and linguistics applied, this was first proposed 40 or so years ago, but it will be clear from what I have said in this talk that, for me, it is fundamental. With linguistics applied (LA), the direction of dependency is from discipline to problem, so the language problems that people actually experience tend to be defined with reference to expertise in linguistics. But applied linguistics (AL) in my view must have a problem-to-discipline dependency, whereby problems as actually experienced are first identified and then disciplinary expertise called upon to the extent that it contributes to their clarification or solution. Although I have only touched upon language audits and assessment, it will be clear that the LA/AL distinction is relevant in that it raises the crucial issue of what is analysed and who does the analysis. To the extent that audits are focused on what has been taught as competence, they fail to capture what is actually learned. In my view, there needs to be more focus on how to give credit to the development of capability as an essential investment for further using and learning of the language. ‘Capacity’, I should add, was an earlier term I used to refer to what I now call ‘capability’, being a more appropriate term.

The teaching of English that is generally referred to as EGP – English for General Purposes – is really ESP in that it has the specific purpose of getting learners to acquire native speaker competence and is in effect essentially a training activity, designed to prepare learners to meet the predictable requirements of assessment. In contrast, teaching which focuses on the development of capability is educational in orientation in that it provides for the continuing adaptive use of language resources to serve the general unpredictable communicative purposes that learners as users will need to achieve beyond the confines of assessment. In so doing, the language they know is authenticated. It may well not conform to the so-called authentic language of native speaker usage that they have been taught to aspire to, but for their communicative purposes as users, there is no reason why it should. The promotion of authenticity as uniquely a property of native speaker usage has the negative effect of preventing learners from authenticating the language as a natural means of communication. Once rid of such pedagogic constraint, they are free as users to exercise their authenticating capability.