2.1 The USA: Chomsky

The next phase of my career was teaching in a private language school in the UK in the early 1970s to finance my Ph.D. literary research. I still had no teaching qualification and no idea what structuralism was, nor did I have any inkling of developments in linguistics or applied linguistics. At that time, unknown to me, in the academic world there were two incompatible schools of thought about language, the gulf between which is well represented by the vast expanse of the Atlantic Ocean.

In America, Noam Chomsky had laid out his theory of transformational grammar and deep structure (Chomsky, Reference Chomsky1957, Reference Chomsky1965). Chomsky was reacting against the kind of structuralism which treated language as a surface phenomenon which could be described purely on the basis of observed behaviour. For Chomsky, the most significant fact was the native speaker's ability to distinguish the grammatical sentences of their language from the non-grammatical ones. Beneath the surface of sentences lay basic, universal structures which, via a series of transformations, ended up with the forms we read and hear. This is a grossly simplified version of what was an elaborate and evolving theory, but, significantly, it lay claim to universality: we are all endowed with the same, basic, universal grammar. As little children, we match that universal template to what we hear around us and build a model of the target grammar which forms our competence in our mother tongue (Chomsky, Reference Chomsky1964). The notion of universal grammar is not new. In the history of European linguistic theory, it goes back to classical antiquity, is seen in the theories of the Medieval modistae, appears again in the seventeenth and eighteenth centuries and is resurrected by Chomsky (for further details, see McCarthy, Reference McCarthyin press, passim). It is a grammar of the mind; it looks inward to the cognitive world. I knew nothing of all this and only came to read Chomsky a decade later. I still struggle to see what relevance his thinking could have for my teaching. To be fair to Chomsky, he never claimed that his ideas were directly applicable to language teaching.

However, Chomskian theory had a profound influence on the emergence of second language acquisition (SLA) as a sub-field in applied linguistics, which, it must be said, led down some scholastic blind alleys and, to me, proved very little. The so-called morpheme studies of the 1970s were based on experimental procedures which claimed to show that there were ‘universal’ orders of acquisition of features of grammar which took their course independently of the sequence of teaching. These features included (for English) articles, present- and past-tense endings, plural endings and so on (e.g. Bailey, Madden, & Krashen, Reference Bailey, Madden and Krashen1974, among others). No doubt many teachers uncritically followed these prescriptions of teaching sequences, which would have amounted to no more than a self-fulfilling prophecy. The morpheme studies have been gradually picked apart and shown to be less valid and less universal than first thought (see most recently Murakami & Alexopoulou, Reference Murakami and Alexopoulou2016). All my experience told me that the most influential factors in determining difficulties encountered by learners of English were L1/L2 (first language/second language) contrasts and the learner's degree of willingness to engage with the L2 and to become language-aware, and that I could help my students most by being sensitive to the characteristics of their L1s and their language learning backgrounds, in so far as I could reliably inform myself of them. I was simply never convinced by empiricist SLA experiments and claims of universality. My experience told me that my students were individuals with individual language, educational, cultural and social-class backgrounds and different motivations for learning.

2.2 Britain: The neo-Firthians

On my side of the Atlantic, in the 1960s and 1970s, language study was also going through a period of remodelling, under the influence of the British linguist J. R. Firth. Firth's approach, one which he instilled in his disciples, including Michael Halliday and my mentor of the 1980s, John Sinclair, was the opposite of Chomsky's; it looked outward, from language forms to the context of utterance and the semiotics of society (Firth, Reference Firth1951/1957). What later became known as systemic-functional linguistics, elaborated by Halliday in a series of seminal papers, described interlocking systems at different levels, explaining relationships from the phonetics and orthography of texts to the grammatical morphemes and lexical forms and how they collocate and colligate in the larger units of phrases and clauses and relate to the context of situation and society (Halliday, Reference Halliday1961, Reference Halliday1978).

I did not discover Halliday and neo-Firthian linguistics until I went to work at the University of Birmingham in the early 1980s; in the 1970s I was as ignorant of developments at home as I was of those on the other side of the Atlantic. But then I did not need to know anything about either school of thought. I was a native-speaking EFL teacher and, as far as I could tell, was doing a decent job. My task in the early 1970s consisted of taking apart in class extended model texts which I found in newspapers and magazines, bringing to the surface ‘language points’ which were then explained on the board and discussed. I didn't learn this way of working from books or journals; I learnt it all through staff-room advice from more experienced teachers and seeing what worked with my students in class. We read the texts, I explained key grammar and vocabulary and any tricky pronunciation, the students made notes, answered comprehension questions and did exercises. Slotted in between these sessions were the dreaded hours of ‘conversation class’, where subjects such as world peace, capital punishment and the advantages and disadvantages of houses versus apartments were to be discussed. I sensed even then that this was an artificial exercise where I floated ideas and the students either said they agreed or disagreed, hardly anyone's definition of a ‘conversation’. But again, I do not casually dismiss the activity; it worked in its own way, and generated vocabulary and grammar which it was my job to support as we went along.

3.1 Ignorance was bliss (seriously)

Most of my students at that time (1970–1975) were well-off, middle-class youngsters preparing for Cambridge exams (First Certificate and Proficiency), predominantly from Europe plus some from other continents, in mixed L1 groups. The job was more demanding than at the Berlitz school, but I still got by on remarkably little technical knowledge of English and relied on my native-speaker instincts as regards grammaticality. Chomsky would have been proud of me. I, and many others like me, untrained, inexpert and often directionless in career terms, typically from a modern foreign languages background, supplied labour to the English language teaching (ELT) industry outside of the universities. From 1970 to the middle of the decade, ‘method’ played no part in my daily work, nor did I ever hear it mentioned in the staff room. I had outgrown Berlitz structuralism and saw my job as helping my students to mine texts, digging out gold nuggets of grammar and vocabulary and giving the best learning experience I could offer. To this day, I believe that you can be a good language teacher with little knowledge of linguistics, educational theory, fashionable methodologies or the latest technologies so long as you are conscientious, motivated, committed and moderately knowledgeable about the language you are teaching and the language(s) of your students. My subsequent identity as academic, course book writer, language researcher, English for academic purposes (EAP) teacher and teacher trainer is mere icing on the cake of what those early experiences gave me. Academic reading, training courses and acquaintance with the latest theories are all worth pursuing, so long as you maintain a critical stance and don't get swept away by it all. Good teaching and learning succeed on the basis of positive interpersonal relations and mutual respect, without which the best theories, methods and materials are doomed to failure.

3.2 (Re-)education and training: the (re-)birth of method

In the mid-1970s, a friend spent a year at Leeds University UK doing a course in teaching English as a foreign language (TEFL) (which at that time I thought was a sort of non-stick coating for frying pans). He spoke highly of it. I became curious enough to make enquiries and subsequently to sign up. Armed with a Ph.D. in literature, I went back to university for a year as a mature student.

The course at Leeds, one of the few in existence at that time for EFL/ESL teachers, was transformative. Our course tutor was Tony Cowie (1931–2015), a scholar of the old school, a lexicographer who worked on several excellent learners’ dictionaries for Oxford University Press, and an inspiring teacher. Little by little, he gently and undogmatically unpicked our view of English teaching and helped us realise that the biggest task facing any learner was building a large vocabulary; there were just so many words and they worked in such complex ways. He introduced us to the notion of collocation, which Firth, Halliday and Sinclair had developed in the previous decades (Firth, Reference Firth1951/1957; Halliday, Reference Halliday, Bazell, Catford, Halliday and Robins1966; Sinclair, Reference Sinclair, Bazell, Catford, Halliday and Robins1966). At the same time, because the course led to a postgraduate diploma in Education and, ultimately, qualified teacher status in the English state school system, we also followed general teaching courses in the university's School of Education. Those I greatly disliked and have almost entirely forgotten. But I never forgot the lessons about collocation. Collocation, an arcane concept in 1975, is now part of the stock-in-trade of vocabulary teaching. We owe it to scholars such as Firth, Halliday and Sinclair that our raw material, the language, could be systematically described and made plain without recourse to classical models inherited from Latin or dependence on the ‘ELT canon’ which publishers, materials writers, curriculum designers and other stakeholders have followed largely uncritically for decades but which has little empirical underpinning or validation in research (Burton, Reference Burton2019).

At Leeds, we also had courses in English for special purposes (ESP), a relatively new concept, and we had opportunities to run practical tutorials for the university's ‘overseas students’, as they were called then. We also had a course in phonetics. Remember phonetics? All in all, I began to appreciate that I was pursuing my career and being (re-)educated at a time of great ferment in the ELT world, for it was also the years when the communicative movement burst on to the scene, a swathe of radical ideas put forward by figures such as David Wilkins (Wilkins, Reference Wilkins1976) and Chris Brumfit and Keith Johnson (expanded upon in Brumfit & Johnson, Reference Brumfit and Johnson1979). Now language was no longer words, phrases and constructions woven to make model texts for dismemberment in class, but a repertoire of strategies for performing communicative acts (requesting, apologising, directing, etc.) and for realising semantic notions (e.g. relations in time, location). The new gospel of the notional-functional approach was upon us.

4.1 Bandwagons and blinkers

Our profession as English teachers suffers from a problem that doubtless affects many other professions: a tendency to jump on bandwagons and head off, blinkered, at full speed without a backward glance and with insufficient self-criticism. Such was the case with communicative language teaching (CLT) in its early days, and such will probably be the case with every new fad and fashion that hits us. For example, we are currently offered the new approach of gamification, listed by the British Council as one of ‘Ten trends and innovations in English language teaching for 2018’ (British Council, 2018), where game-like elements are introduced into language teaching/learning processes. Games are loved by millions of people. However, the assumption that what people enjoy must be good for language learning remains unratified. Without a hint of modesty, the website of the Duolingo programme, which rests heavily on the approach, claims that, ‘A study has shown that 34 h of Duolingo are equal to 1 university semester of language courses.’ (Duolingo, 2020). I'm not sure I would want to send a youngster to that university. Gamification may indeed prove to be a worthy approach once enough evidence is marshalled, but things that are fashionable, fun and motivating do not necessarily equate to a good learning environment or rigorous, effective materials and verifiable and replicable learning outcomes.

In the mid-1970s my career took me to Sweden. As well as general English for adults, I taught ESP courses for medical and pharmaceutical personnel, ball-bearing manufacturers, lorry designers and people in various other industries. My world was now very different from that of the Berlitz class of ’66. My employers were enlightened aficionados of CLT but wisely let us teachers find a balance between the new wave and a good foundation in grammar, vocabulary and pronunciation. The book repository at the institute where I worked was stocked with the best-sellers of the 1970s, including Kernel lessons Plus (O'Neill, Reference O'Neill1973), Strategies (Abbs, Ayton, & Freebairn, Reference Abbs, Ayton and Freebairn1975), Functions of English (Jones, Reference Jones1977), Streamline English (Hartley & Viney, Reference Hartley and Viney1978), which exposed us to new trends in materials and methods, alongside solid, useful grammars such as Leech and Svartvik's communicative grammar (Leech & Svartvik, Reference Leech and Svartvik1975). Household names, but just names on books for me and my colleagues in those days.

4.2 CLT boot camp

Now and again, we would be visited by publisher-sponsored teacher trainers, some of whom were CLT zealots who usually only succeeded in putting our backs up. It was annoying to be told that we were old hat and superannuated if we didn't throw grammar out of the window and focus instead on getting people to open the window and pass the salt, then apologise for spilling it. All this despite the fact that Wilkins, one of the foremost figures of the communicative movement, never advocated abandoning grammar (Wilkins, Reference Wilkins, Brumfit and Johnson1979). For those adopting the strong version of CLT, this was typical of the cherry-picking which often occurs with elaborate theories: grab the fun bits and ditch the more challenging aspects, in this case how to balance fluency and accuracy. My colleagues and I adhered to the common-sense instinct that a healthy mix of CLT and more traditional core language elements was the best approach. What worked, worked. The extreme, strong view of CLT never appealed to me. It was frequently promulgated by self-appointed experts who delivered teacher-training workshops lacking rigour and depth but with a sense of confident authority matching that of a sergeant in a military boot camp. Behind CLT lay speech act theory, the complexities of form-functional relationships and the far-from-simple notion of communicative competence (Canale & Swain, Reference Canale and Swain1980). So often, workshops and training sessions simply preached that CLT was a silver bullet, with little in the way of critical thinking, and we were often made to feel inadequate and in need of re-education if we disagreed.

Leech and Svartvik, in the second edition of their 1975 A Communicative Grammar of English published in 1994, claimed that the communicative approach had become ‘an efficient and popular method for learning foreign languages’ (Leech & Svartvik, Reference Leech and Svartvik1994: xi). ‘Efficient’? On what evidence? I only later discovered, when working in Asia, that my reservations concerning efficiency at that time were shared by many teachers outside of the echo-chambers of Europe and America. There is ample evidence that local pedagogical cultures, along with practical day-to-day constraints, meant that many parts of the world carried on as before, sometimes paying lip-service to CLT but maintaining allegiance to traditional and local pedagogical values (Battistella, Reference Battistella and Wheeler1999; Yu, Reference Yu2001), and why shouldn't they?

5.1 The 1980s: discourse analysis and corpus linguistics

When I returned to the UK in the early 1980s, I became a full-time academic at the University of Birmingham and worked alongside of some great names in applied linguistics and ELT who generously shared their expertise on a daily basis: John Sinclair (lexis, corpus linguistics, discourse analysis), David Brazil (intonation), Malcolm Coulthard (discourse analysis), Tim Johns (ESP/EAP and early computer-assisted language learning (CALL)), Tony Dudley-Evans (ESP/EAP), Michael Hoey (written text analysis). Sinclair encouraged me to read Halliday and become conversant with systemic-functional linguistics. This I did, and my outlook on language was transformed; for me, Halliday's work to this day remains the most coherent, appealing and usable account of English. But the main reason I felt at home at Birmingham was that everyone seemed to take it for granted that at the centre of the applied linguistics and language teaching enterprise was the study of language. Sinclair's enthusiasm for language was infectious. During that time, the notion of methodology faded further from my consciousness. The CLT juggernaut rolled on outside of my academic world and was moving towards task-based learning and the burgeoning lexical approach. I lay myself open to justified criticism here: taken in ‘off the street’ as it were, by Sinclair, I had the luxury of researching English, not just teaching it, a privilege accorded to few in our profession.

The 1980s saw the beginning of my long partnership with the late Ron Carter. We were both disciples of Sinclair's approach, which was firmly embedded in the outward-looking, neo-Firthian tradition of studying language in context. That period saw the growth of corpus linguistics and, thanks to Sinclair and his team of lexicographers, the ground-breaking COBUILD dictionary (COBUILD, 1987). Its explanations of meaning were based on corpus evidence, which represented a revolution in dictionary-making, something hard to imagine now in an era when no self-respecting ELT publisher would launch a dictionary not based on a corpus. But there was something about over-keen corpus linguists that would get people's backs up in a similar way to how CLT zealots got my back up in the late 1970s.

5.2 The impetus for spoken corpora in ELT

By 1990, when I moved to the University of Nottingham to work more closely with Ron Carter, we were keen to build a corpus of everyday spoken English, for at that time, most corpus investigation was carried out on written texts. This was understandable, as small, portable audio recording devices had yet to be widely available outside the world of James Bond, and the collection and transcription of spoken data was prohibitively time-consuming and expensive. But I was keen to be involved in spoken corpus building, prompted by my past and thoughts of the future. From the past I recalled the stilted spoken sentences of the Berlitz structuralist method and the later, inadequate conversation classes I had presided over. I gained from those two different experiences a desire to know more about the spoken language and possibly to develop materials for the teaching of it which would be innovative, useful and evidence based. Also, in my brushes with discourse analysis and conversation analysis, it struck me that scholars typically analysed individual transcripts and were missing the bigger picture. What we needed was a bird's eye view of everyday speaking. After all, the amount of writing humans engage in day by day is dwarfed by the amount of speaking. Furthermore, it was becoming clear that the world was on the move and that the well-written, formal letter or an essay about world peace would soon be things of the past in an international setting where people were more and more talking to one another face to face, on the phone and via video conferences.

5.3 Which English, whose English?

When Ron Carter and I were building the CANCODE spoken corpus in the early 1990s, we little knew what controversy it would generate. We wanted recordings made in as natural and wide a selection of everyday situations as possible involving speakers from a broad range of ages, social class and educational and geographical backgrounds. We both felt that we could do better than the unnatural, stiff and precious middle-class dialogues which peppered courses dominated by models derived from writing and south-east British English. We ended up with five million words transcribed from recordings of native speakers made around Britain and Ireland. We were proud that we had captured ordinary people of all social classes at home, at work, in the shops, while driving, while studying, while eating in restaurants, etc. (see McCarthy, Reference McCarthy1998 for an account of the structure and content of the corpus). Our initial research proved promising: we began to unearth patterns of spoken grammar that had been underplayed or ignored in existing reference and pedagogical grammars and materials and we put out two papers reporting our provisional findings (Carter & McCarthy, Reference Carter and McCarthy1995; McCarthy & Carter, Reference McCarthy and Carter1995). But we had stepped into a minefield.

Critics of corpus linguistics had already begun marshalling their arguments in the early 1990s, partly agitated by John Sinclair's stance as regards the primacy of corpus evidence (Sinclair, Reference Sinclair1991) and Carter and I were on the receiving end from 1995 onwards. The critics were of two kinds: the first were practising teachers at conferences who expressed anything from indifference to outright hostility towards the idea that spoken corpus evidence might make a positive contribution to materials. On one famous occasion, an audience member said they didn't want to hear about the grammar of Ron Carter's and Mike McCarthy's ‘semi-literate friends’. The other phalanx of critics consisted of academics who challenged the relevance of a predominantly British dataset in a world where English had already branched off into ‘Englishes’, the robust local varieties that had evolved in places such as India, Africa and parts of Asia (see, for example, the papers in Collins, Reference Collins2015). The notion of ‘inner circle’ countries (e.g. the UK, the USA, Australia), ‘outer circle’ ones (e.g. countries in Africa and Asia where English traditionally enjoyed a high status) and the ‘expanding circle’ of countries where English was assuming ever-greater importance in international communication (Kachru, Reference Kachru, Quirk and Widdowson1985) was a powerful counter to our data collected in a quirky corner of the world which was becoming less and less important on the global scene. English was now an international language rather than the language of its imperial motherland; it had become the world's lingua franca that prompted many to re-arrange the initials EFL to ELF and to call for a pedagogy sensitive to the forms used by English as a lingua franca (ELF) users. Our British English could not just sit out there as a neutral object for exploitation and consumption: it came with ideological baggage (Pennycook, Reference Pennycook, González and Melis2012). We were rightly reminded by persuasive statistics that we were living in a world where native speakers of English were in a minority (Graddol, Reference Graddol1997), and a somewhat irrelevant minority at that.

Other powerful arguments were also held up to us: the kind of spoken language we had in our corpus (everyday, context-embedded conversations) was inextricably tied to its contexts. A corpus wrenches the data out of its contexts, so it is no longer real or authentic (Widdowson, Reference Widdowson2000). Meanwhile, Owen (Reference Owen1996) argued that intuitive prescription in language pedagogy did not require corpus evidence in order to be valid and Cook (Reference Cook1998) held that the ways grammarians organise the language in their grammars also constituted facts about the language. Given what I had by then learnt from my positive encounter with Halliday's systemic-functional grammar, I could hardly disagree with Owen and Cook. But Cook went further, and claimed that ‘immense damage’ could be done by those taking a hard line on the use of corpora, adding that a good deal of attested language was ‘inarticulate, impoverished, and inexpressive’ (Cook, Reference Cook1998, p. 61) and that it was a mistake to equate ‘real’ English with native-speaker English. Critics were pointing the finger at a strong version of corpus linguistics which seemed to be heading in a blinkered fashion down the wrong road just as had happened with over-zealous CLT.

6.1 The rebirth of methodological hard liners?

There were two sides to the spoken corpus debate of the 1990s, and some of the arguments on both sides are still relevant. Firstly, I can only agree that any new trend in language teaching, if embraced in its strongest form, is likely to be less than effective in delivering good language education. This is because when extreme approaches are adopted, the approach takes over and nothing must get in its way: the ideology becomes more important than the human participants it is meant to serve. This happened with extreme versions of structuralism, CLT, Chomskian SLA and corpus linguistics during the twentieth century, and I experienced all of them in one way or another. Having said that, in relation specifically to corpora, I often wondered what the ‘immense damage’ which Cook feared would look like, the spectre of students sounding ‘inarticulate, impoverished and inexpressive’ (and who would be the judge of that? – the middle-class intelligentsia, I would guess), though Cook, it must be said, was more generous to the position Carter and I had adopted than he was to others whom he named.

One of the criticisms levelled against corpus hard-liners was related to their belief in dissolving the conventional boundaries between grammar and vocabulary and focussing instead on ‘lexicogrammar’ (a term originally coined by Halliday) and formulaic ‘chunks’ rather than the slot-and-filler approach of my 1960s Berlitz book. The lexical approach advocated by Sinclair and Renouf (Reference Sinclair, Renouf, Carter and McCarthy1988) and Willis (Reference Willis1990) was based on the idea that the most frequent words and chunks were the fundamental fabric of communication rather than ‘structures’ and atomic vocabulary items, and so they are, as corpora have demonstrated. Unfortunately, this view also often harboured the assumption that what the data revealed would be indisputably important and relevant to learners anywhere and everywhere, and early, radically different materials based on the approach failed to convince teachers and did not achieve their commercial promise.

6.2 The human value of spoken corpora

Carter and I never saw ourselves as hard-line corpus linguists, and the EFL/EAP teaching I did at Birmingham and Nottingham was an admixture of the well-tried text-mining, vocabulary building and attention to grammatical conventions that had served me for a decade or more (though, for sure, I had abandoned ‘the pencil’, ‘the book’ and ‘the table’). However, what I added was evidence from our spoken corpus, some of which had been gathered in university lectures, tutorials and seminars, which formed the basis of advice which I felt was useful to my students. My teaching was corpus-informed, not corpus-driven. Behind this lay the fact that we were less interested in the language items which realised communicative acts than in how people communicated. In EAP and in general, everyday spoken English, the corpus confirmed how speakers and listeners worked hard to create satisfactory encounters, whether socialising, buying stamps in a post office, participating in a university seminar or chatting to one another while driving or having their hair cut (for examples, see Carter & McCarthy, Reference Carter and McCarthy1997). For me, it was less important that the speakers in the corpus were British or Irish than that they were human beings trying to rub along together as best they could. As Carter and I supervised Ph.D. students who were building spoken corpora in other languages (in my case, speakers of Turkish and Burmese) for comparison with our English data, we began to understand better the commonalities that underlie human communion: a ‘universal grammar’ but not one of deep structure and well-formed sentences, rather a grammar of interaction, whose goal was building and consolidating relationships to get things done. Subsequently working with co-researchers analysing Spanish and Russian data and comparing results with English confirmed even more strongly that this was the gift corpora could bring to the table (see for example, Amador Moreno, McCarthy, & O'Keeffe, Reference Amador Moreno, McCarthy and O'Keeffe2013; Malyuga & McCarthy, Reference Malyuga and McCarthy2018). Steadily amassing insights into how conversations worked enabled me and my co-authors to put together the commercially successful Touchstone and Viewpoint course book series published from 2005 to 2014, based on north American spoken data and learner corpora. For the course series, we constructed a conversational syllabus on a set of principles which we condensed from our observations of the data. Our rationale for the syllabus and its corpus-informed grounding may be found in McCarten (Reference McCarten, O'Keeffe and McCarthy2010).

6.3 The dethroning of the native speaker

But what of the arguments about the (ir)relevance of our British and Irish data, conversations from two small islands, in a world where scholars were telling us English no longer belonged to us? There are persuasive arguments for the view that English is in the hands of its global users, and that it is they who will shape it to their needs in future. The facts were already apparent in the 1990s, but even more so now in 2020 when billions of words of English are jabbed into phones and tablets every day around the world to come to rest on websites, emails and social media platforms. A number of online sources currently offer counts of which languages are most used on the internet, and, at the time of writing, all agree that English is number one. And in international business, academic exchange and tourism, English reigns supreme.

So English hardly belongs to its native speakers. But do languages ‘belong’ to anyone? Is the whole debate on ‘ownership’ flawed? Does the population of native users really think, or has it ever thought, that it ‘owns’ English? It has certainly never struck me that way. English is ‘my’ language, but I don't feel I own it any more than I own ‘my’ brother or a train when I say, ‘my train leaves at 4pm.’ Over the years, I have used various languages in places where I have worked (Spain, Sweden, Malaysia) and at times I have wished that ‘my’ Swedish/Spanish was better. There is a slipperiness in the notion of ownership. It is wiser to think of English as a communicative tool which carries with it the culture(s) of its users, and is there for anyone to use, crucially in ways that they wish or may aspire to. No-one owns it, neither I as a native user nor the multitudes of ELF users in all their variety; I am still to be convinced that there is a global ELF ‘core’ (see the discussion in Prodromou, Reference Prodromou2007) any more than there is a global native user core. For this reason, Carter and I were pleased when critics of our spoken corpus went off and gathered their own data reflective of their own realities, contexts and needs. We now have a wealth of spoken corpora of different varieties of English, native and non-native, ELF, specialist and general, learner and expert-user, and an ever-expanding catalogue of corpora in other languages, too many to list here, all easy to track down on the web and many now openly accessible for all to explore. Let's hope that diversity and equality of worth is respected and that zealots don't start to insist that some types of spoken corpora are more equal than others.

6.4 Ownership and access

For English, the problem is not one of ownership but of access. When I began as an EFL teacher, access to English (outside of creaking national school systems) was primarily the privilege of relatively well-off middle-class students at university or those who could travel or go to evening classes. The massive world-wide expansion of university education and increased global mobility have enabled more and more people to have access to English and generated a need or desire to use it, though the developing world has often trailed behind in terms of human and physical resources. But there is no doubt that in the present century the rapid increase in the movements of people around the world, whether through war, persecution or economic deprivation, has precipitated a new level of urgency in the provision of English for survival among displaced peoples, a situation investigated in an insightful study by Harrington (Reference Harrington2018). It is therefore no longer a question of who owns English, or even a question of how we might define English in its international role, or whether everyone should jump on yet another bandwagon and ‘teach ELF’. The fragmentation and plurality of English is its strength and everyone, native-speakers, expert-non-native-speakers, learners, internet users, displaced persons, gamers, business people, social media friends and followers, etc. can use it in any way they wish rather than ‘own’ it. And this very dissolution of a sense of ownership might perchance bring with it the demise of the middle-class value system which has for too long dominated ELT as reflected in the stultifying orthodoxy of its materials. On this last point, I am not over-optimistic, with the latest EFL coursebooks still largely portraying a world of middle-class mores and comfortable mobility via woke and worthy dialogues and pictures of happy, shiny, beautiful people with perfect teeth, e.g. see the sample materials online for English File fourth edition (English File, 2020) and for Empower Starter (Empower, 2020), which are presumably offered as representative of the materials. I do not say this as a world-weary and cynical old fogey: I sympathise with creatively minded authors who are blocked from portraying anything but a particular version of reality by risk-averse publishers. I speak from experience; my own co-authored materials have been subject to such constraints and I've had to live with them.