1. Multi-competence background

In the early 1990s I proposed a perspective called multi-competence, originally defined as ‘the compound state of a mind with two grammars’ (Cook, 1991). This was reworded as ‘the knowledge of two languages in one mind’ (Cook, 2007a), because some people took the term ‘grammar’ in the narrow meaning of syntax rather than in the broad meaning of linguistic competence intended. Multi-competence emphasised the relationships of the languages in the same person’s mind rather than the separate existence of a first language and an interlanguage. Multi-competence is not a model of second language acquisition in the same sense as say Universal Grammar (UG) but is more a way of looking at second language acquisition from the vantage point of the L2 user as a distinct kind of person rather than from that of the native speaker – all of the people described above are doing things no monolingual can do. Multi-competence can therefore have implications for many areas of L2 acquisition and use, even for UG (Cook, t.a.).

The multi-competence perspective was productive in supporting the growing movement to regard the L2 user as a person in their own right rather than as a defective native speaker, in highlighting the reverse transfer effects of the second language on the first, and in showing the distinctive ways of bilingual thought, as summarised in Cook (2007a). The implications for teaching that have emerged concern the role of the first language, the nature of the second language (L2) user’s knowledge of the second language and the goals of language teaching (Cook, 2007b).

So multi-competence was conceived of as Lang₅ rather than as social in the Lang₄ meaning of ‘the possession of a community’ or as societal in the Lang₂ sense of ‘an abstract external entity’ . Others were taking up this point about the restriction to Lang₅ . Kelly Hall et al (2006) made ‘a case for a usage-based view of multi-competence’ (p.220). Brutt-Griffler (2002) proposed a Lang₄ ‘the multi-competence of the community’; rather than assuming communities have a single language, we should start by seeing how languages relate to each other within the community. Jenkins (2006) pointed out the absence of ELF (English as Lingua Franca) from multi-competence, exemplifying a gap in second language acquisition (SLA) research in general.

Another weakness that became apparent was that multi-competence seemed to treat language knowledge as static rather than constantly changing. In terms of the individual, language knowledge is in flux, according to dynamic systems theory (De Bot et al, 2005; Herdina & Jessner, 2002): ‘all language knowledge is socially contingent and dynamic’ (Kelly Hall et al, 2006: 229). In the first language, a person’s language knowledge may be growing, as in children acquiring their first language, or declining, as in first language attrition through injury, ageing or change of circumstances when another language becomes more important to them. The window during which linguistic competence is static may be small and untypical. In the second language, L2 users may be acquiring the second language or may be losing it through attrition: all is change. Individual multi-competence in a sense builds this in through the changing relationship between the two languages in the same mind on the integration continuum (Cook, 2003); the term ‘attrition’ with its negative connotations of loss and damage is no longer appropriate within the system as a whole. The balance and form of the two languages changes over time and shifts dynamically. The relationships between languages in the multilingual community is also continuously shifting, say the growing problems between French-speaking and Dutch-speaking areas of Belgium or the use of Spanish as a ‘niche’ lingua franca in London (Block, 2006), or Italian among migrant workers in German-speaking parts of Switzerland (Schmid, 1994).

2. Communities and language user groups

The social aspects of language have mostly been discussed through the concept of community. Sociolinguistics has come up with many definitions of the speech community (Patrick, 2001), whether as people living in an area such as the Lower East Side of New York (Labov, 1966) or Bergen (Kerswill, 1994), people united by a uniform style of speech (Bloomfield, 1926) or a set of evaluative norms (Labov, 1966), or people who belong to a social network (Gumperz & Levinson, 1996). A perpetual dilemma is whether you first find an existing community defined in, say, geographical terms and then look at the languages they use, or you find a language and look for the community of its speakers: is the sociological group the given or the language?While language is often seen as a shared core value of the community (Smolicz et al, 2003), it is not criterial, as in the case of Jewish communities speaking diverse languages (Myhill, 2003); members of the community are not necessarily fluent in its language, as with Scottish Gaelic (Dorian, 1981). People may be part of a cultural community without speaking its language – how many Irish Americans learn Irish?

The concept of language community can be circular when the number of speakers of a language is taken to be the number of members of a culture, as sometimes happens for Welsh (Laitin, 2000). The core value of a community is, however, almost invariably taken to be a single language; a minority ethnic community is seen as identifying itself with its own language, protecting it and maintaining it as a heritage. ‘An individual’s use of two languages supposes the existence of two different language communities; it does not suppose the existence of a bilingual community’ (Mackey, 1972, 554). This denies the reality of the multilingual communities in the world with more than one language at their core; India for example has a ‘Three language Formula’ 3±1 system in which everyone has to know Hindi and English plus the local state language; if the local state language is neither Hindi nor English, they only need two languages; if they live in a state where Hindi or English is not the state language and they speak another language, they need four languages (Laitin, 2000).

So does a second language confer partial membership of another monolingual community or full membership of a bilingual community – what Brutt-Griffler (2002) terms the ‘multi-competence of the community’? English has a L2 user group of people across the world, whether businessmen, academics or international footballers, for whom the native speaker community is virtually irrelevant. Having two languages may bring people into a different multilingual community. Or it may allow them to belong to a global virtual community in a possible social network unrelated to geographical proximity or to any common language identity in the usual terms.

Let us do the multi-competence reversal of viewing the multilingual community as normal and the monolingual community as an aberration. Most nations in the world regardless of their official positions as monolingual or bilingual actually contain speakers of many different languages. Toronto for instance has 2,746,480 mother tongue English speakers, 58,590 mother tongue French but 2,160,330 speakers of neither English nor French, the official languages of Canada (2006 census, Canada stats, 2007); London has a similar profile with 300 odd languages spoken. At one level people may belong to native speaker communities who talk to fellow-members, say the English speakers, the Italians and the Chinese in Toronto regarded separately. Clearly the overall community in these conurbations is multilingual, with speakers of the non-status language using a second language English to communicate with the English speakers and with members of other communities, Italians with Chinese etc or even another language as a lingua franca, say the use of Italian in Toronto by Vietnamese and Poles in their workplaces (Norton, 2000). Just as the concept of individual multi-competence stressed the L2 user in their own right, so the multi-competence of the community stresses the multilingual community in its own right, not as a collection of people with different L1s but as a community with an integral use of two or more languages. According to Canagarajah (2007), ‘Linguistic diversity is at the heart of multilingual communities. There is constant interaction between language groups, and they overlap, interpenetrate, and mesh in fascinating ways.’

3. The De Swaan Hierarchy

Figure 1. The hierarchy of languages (adapted from De Swaan, 2001)

In this scheme, languages differ in terms of geographical and functional areas – where they are used and why. At the bottom come languages that are peripheral; they are used within a circumscribed territory for the purposes of a local community. This may be a small section within a country, say Gaelic in Scotland, cover a whole country, Japanese in Japan, or extend across countries, Kurdish in Turkey, Iran and Iraq. As the term ‘peripheral’ seems to convey some evaluation, I prefer the term local as more neutral. Next up the hierarchy come central languages used within a geographical area for communication between different groups mostly for education and government, say English in India used by native speakers of many languages for everyday public functions. Above this come supercentral languages that have a wider geographical spread and are used for cross-national communication for a limited range of functions, say Arabic or Latin for religious ceremonies, or Japanese for karate. Finally at the top come hypercentral languages used chiefly by non-native speakers across the globe for a large range of purposes.

4. Groups of language users

Language users can be divided into groups according to the four de Swaan levels.

The first group is people using their first language with each other in the local language geographical territory. This is the sole group to contain only monolingual native speakers of the language. A monolingual Londoner, Berliner or Parisian can get by on their first language alone. The size or location of a speech community does not affect the issue of whether people are native speakers of a local language: you don’t forfeit your native speaker status by belonging to a minority community even if a local variety, say French in Louisiana. Nor indeed do you stop being a native speaker by moving to another language community or by ceasing to use it altogether –  monks under a vow of silence are still native speakers. Group A communities use their native local language for all the possible monolingual language functions of human life. It is not usually questioned whether Group A native speakers are successful, since, apart from those with a disability, they are successful by definition: a native speaker is whatever a native speaker is.

The second group consists of permanent residents using a central second language to communicate with the wider community outside their local language group, in Siegel’s terms a dominant L2 setting (Siegel, 2006). Some local language groups may be indigenous in that they have been established for generations, say the old-established Bengali speakers in London or Gujarati speakers in Brooklyn, or may predate the use of the central language, for instance Aboriginals in Australia. Others may be short-term immigrant communities, like Polish workers in England.

Group B constitutes the classic ‘second language’ situation – a language ‘that becomes another tool of communication alongside the first language … typically acquired in a social environment in which it is actually spoken’ (19). The first language remains alongside the second language for other social and cultural purposes, probably the case most reported in language contact and shift, for example Fishman (1991). The central language may be the language of government and so the means through which speakers of purely local languages fit into the society structurally, in Siegel’s terms an institutional L2 setting (Siegel, 2006). The central language is not, however, used exclusively for communicating with the monolingual Group A but also with fellow-users with different native languages. The Bengali shop owner in Tower Hamlets uses English for filling in his income tax return but also for speaking with Arabic customers, both equally English in nationality, as are most of the speakers of the 300 first languages of London (Baker & Eversley, 2000). The central language enables people to cooperate with each other for everyday living within the same area. The native speakers of anything but the central language have to learn it or be taught it.

While some of the users of a central language are Group A monolingual natives, others are L2 users from many other L1 local languages, say the situation in the states of India. Success for the non-native learner of a central language means the ability to use it for all the necessary purposes of their public lives and their unofficial contacts. In accordance with the multi-competence view of the reverse transfer effects from the L2 to the L1, this may mean that the L1 is to some extent affected by the central language; Cantonese speakers in Newcastle upon Tyne for instance say ‘tiojau’ ('table wine') rather than ‘jau’ ('wine) as used in Hong Kong and ‘bafong’ ('bathroom') rather than ‘saisanfong’ (Li Wei, 1994, 66).

The third group consists of people using a supercentral language across national or linguistic borders for a specific range of functions of language rather than for all functions; the closest correspondence in Siegel’s scheme is coexisting L2 settings (Siegel, 2006). Religions for example have often required believers to use a particular language, whether Hebrew, Arabic, Latin or, occasionally, English. In England for many years civil engineers had to study German because of its importance to their profession. French was once the international language of world diplomacy. Sometimes this has led to specialised international versions of the language for particular functions, for example SeaSpeak (English for Mariners) (Weeks et al, 1988), Simplified Technical English (ASD, 2007), Simple English Wikipedia (2008) or Special English used for broadcasting on the Voice of America (VOA, 2008).

Supercentral languages can be used for wider public or private functions across different countries. Swahili has 770 thousand native speakers mostly in Tanzania but 30 million lingua franca speakers spread across several African countries (Gordon, 2005). While Group C  includes monolingual native speakers, these may be outnumbered by the non-natives; nativeness is not a criterion for admission. Success is measured by the ability to use the supercentral language for specific functions, say landing a plane safely via Air Control English, or by the ability to use it within its international territory, for instance French as part of Francophone culture in Central Africa and France. A particular group is language teachers using their second language for teaching in classrooms – a specialised function of a language and hence falling under this category. In many cases supercentral languages make a token gesture to the native speakers of the country where the language once originated, say France for French or England for English, often showing an historical connection with the colonialising power, say France for Central Africa or Spain for much of Latin America.

The fourth group consists of people using a hypercentral second language, perforce English, globally across all countries and used for all possible second language functions, rather than the limited set in Group C. The usual term for this now is English as Lingua Franca (ELF) – English as a global means of communication between native speakers of other languages. People who speak ELF belong to communities that cross frontiers, united by their common interests – ‘a virtual speech community’ (Canagarajah, 2007, 925). In Graddol’s view, learning English no longer counts as learning another language; it’s an addition to the three Rs necessary for primary school children everywhere (Graddol, 2006). The fact that some people may speak the hypercentral language as native speakers is irrelevant to the use of the language, indeed may be a handicap (House, 2003). The concepts of native speaker or home country are immaterial to these L2 users; the native speaker Group A are only one of the possible types of user, whose local language English may differ significantly from hypercentral ELF. House (2003) distinguishes between languages for identification and languages for communication: no-one is a native speaker of ELF; no-one treats it as their prime identity; they simply use it for communicating with other people like themselves.

The de Swaan analysis treats language users in terms of wider group membership and of language function. De Swaan (2001) sees the acquisition of second languages as typically going up the hierarchy. Speakers of a local language have to learn a central language to function in their own society, say speakers of Ulster Scots learning English in Northern Ireland. Speakers of a central language need to learn a supercentral language to function within their region, say speakers of Ha learning Swahili in Tanzania. Speakers of a supercentral language need the hypercentral language to function globally, true of anybody but a native speaker of English. Some L2 users move in the opposite direction down the hierarchy; an immigrant to a new area may need to acquire the local language, say Finnish in Finland or French in Quebec, seen by Siegel (2006) as a minority language setting. In this case, passing for the monolingual native speakers that make up Group A may be a possible motivation, even if unrealistic for most people. De Swaan (2001) claims that L2 users are the glue that keeps these societies together; without people who can bridge the language gaps they would cease to function. For SLA research and language teaching the lesson is that there are many groups of second language users, not one standard type, and that these differ in the functions that they use the second language for.

The De Swaan hierarchy does not, however, encompass the many groups of L2 users to whom the two dimensions of territory and function hardly apply. Let us try and sketch one or two of these other groups in terms of their membership and language function.

People descended from a particular cultural or ethnic group may want to learn its language, for instance to talk to their grandparents. Mandarin Chinese is now being learnt by 30 million adults round the world (Graddol, 2006); the Confucius Institutes for teaching Chinese around the globe have found that many of their students are ethnic Chinese wanting to learn Mandarin; in the United States 140,000 people are attending heritage Chinese classes (Brecht & Rivers, 2005). Language maintenance classes take place in most British cities, in London ranging from Polish to Greek. Some people are trying to find their roots through language; they don’t want to be L2 users so much as participants in a Group A cultural tradition. The multilingual De Swaan hierarchy seems beside the point; the language is not being learnt for its inter-group value, only as a way of identifying with an ethnic or national group – an identity language. The measure of success is the L2 users’ ability to feel part of their historically related community (Valdés, 2005).

Some L2 users, however, keep their user language within a small social group. People have often joked that the best way of learning a language is to marry someone who speaks it; bilingual married couples feel they are quite capable of passing for native speakers (Piller, 2002). Some parents choose to use a language with their children they will not encounter outside the home, whether George Saunders (1988) using German in Australia or d’Armond Speers using Klingon (d’Armond Speers, 2006). Children, chiefly twins, sometimes create private or secret languages of their own (Thorpe et al, 2000). But unrelated pairs of people can decide to use a second language: Henry VIII wrote love letters to Ann Boleyn and Katherine of Aragon in French (Vatican City, 2006), the language of courtly love. Success for this personal language is perhaps immeasurable. The de Swaan hierarchy of territory and function has nothing to do with these pairs and handfuls of people.

As well as these six reasonably distinct groups, other groupings seem to combine these categories in one way or another. Tourists for example expect to be able to get along with L2 English, regardless of the local language – Japanese tourists using English in Cuba. English for tourism is no longer a matter of English-speaking tourists going to non-English speaking countries or non-English speaking tourists going to English-speaking countries: 74% of tourism through English involves only L2 users (World Tourism Organisation, cited in Graddol (2006)). In other words tourism for English is now a question of the global use of the hypercentral language. But tourists are only one example of the general group of short-term visitors to another country or area with a different language. Others include: athletes going to the Olympic Games, businessmen attending conferences, policemen investigating crimes, pilgrims, British retirees visiting their villas in Spain in so far as they need Spanish at all, the list is endless. All of these have short-term needs for a language, whether for the hypercentral language or for central or supercentral languages. But mostly these short-term visitors need it for very specific purposes to do with their visit and with any associated social functions. It seems irrelevant to the groups A-F since it is not related to territory.

A further group is people gaining an education through a second language. Universities around the globe from Germany to Dubai have started teaching in English, not so much as a lingua franca as a means to an educational end. In reverse, students go to another country to get their higher education, Zaireans to Paris, Greeks to England... Group A native speakers are barely relevant to this (except in so far as they can profit by teaching ‘their’ language). Within this general framework comes the elite bilingualism of children educated in International Schools, whether multilingual, such as the European schools (Baetens Beardsmore, 1993), or taught through a second language, say the French Lycée in London or the Canadian bilingual immersion schools. Success is measured by gaining the appropriate examinations, qualifications, entry to a career, and so on. This education language is probably reserved for movement up the hierarchy to supercentral and hypercentral languages; it is thus distinct from the use of a central Group B language by different local Group As, or the global use of a hypercentral language, though these may be true for some of its speakers; its closest relationship seems to be with the limited functionality of supercentral languages.

There are also people who return to their country of historical origin and need to re‑acquire the first language, or indeed to acquire it for the first time. Examples might be Jews ‘returning’ to Israel having to learn Hebrew even if they were Yiddish speakers (Fishman, 1991), or Puerto Ricans ‘returning’ from the US to Puerto Rico (Clachar, 1997). Others are children of expats going back to the country their family originally came from, say Japanese children returning to Japan (Kanno, 2000); these need to acquire the language of the homeland for practical purposes, often finding it extremely difficult. The difference from Group E heritage users is that the language is for active use as a local language, rather than for gaining cultural identity.

A large group consists of children being taught a second language as part of the school curriculum – the classic foreign language situation whether French in England or Spanish in Japan, called by Siegel (2003) an external L2 setting. They may be Group A native speakers of a language or Group B L2 users of a central language; they are usually learning another language that is higher on the De Swaan hierarchy – say French taught in London secondary schools through English to local children whose native languages may be English, Gujarati and Greek. They could not be fitted into the main groups above since their only need for the L2 is in the classroom, their main goal to get through the hurdles of the educational system. Since this group of non-users will figure in the following discussion, we can call them the CL (Classroom Learner) group.

Nevertheless some language teaching aims purely at effective classroom L2 use. Community Language Learning for example sees language learning in terms of the benefits for the individual learners without reference to the outside world (Curran, 1976). Task-based learning also often seems to see the point of acquiring the language as being the ability to carry out classroom tasks not necessarily related to the world outside: ‘I regard this as desirable but difficult to obtain in practice’ (Skehan, 1998, 96).

Finally it is perhaps obvious that an individual may have multiple memberships in these groups. A professional footballer coming to London needs not just the central language for coping with living there but also the specialised supercentral language of football for interacting with the rest of the team (Kellerman et al, 2005) – a majority of Premiership footballers in England are non-native speakers of English. ‘Each man in his life plays many parts’, simultaneously and consecutively.

5. Language groups and SLA research

Before turning to the implications of this analysis for language teaching, we should see how these groups are handled by SLA research. All of these groups of L2 users have acquired the language in one way or another. Group B central languages come out of untutored acquisition in the home, the street or the workplace by children or adults. Or they may be the product of education in the school classroom for children or specialist classes for adults. A Bengali child in the East End of London may pick up English knocking about with a football in the park or learn English from a teaching assistant. Some SLA research has looked at this type of acquisition from Dulay and Burt (1973) studying grammatical morphemes acquisition among Spanish-speaking children in California to Hannan (2004) doing the same with East End Bengalis. Central language acquisition by local groups has not however usually been distinguished from other types of acquisition, except through the second/foreign language distinction. Research into the acquisition of a central language relating officially to the society and unofficially to other groups within it has seldom been undertaken, except in terms of the development of such groups through passage to a putative Group A native group.

The supercentral language acquired by Group C, the international specialist communicators, may again be learnt in specialist classes for particular needs, Japanese in judo classes, Arabic in the study of the Koran, French for bilingual secretaries. What English do you need to play football professionally (Kellerman et al, 2005)? How much German to function as a migrant worker (Klein & Dittmar, 1979)? SLA research has rarely looked at the specific properties of such acquisition.

Group D, the acquisition of the hypercentral language, is starting to be studied as a specific form of acquisition. Descriptions of the pronunciation of L2 learners in their own terms rather than native terms have been produced by Jenkins (2002); descriptions of the ways L2 users use grammar to each other are coming out of the ELF movement, for example Seidlhofer (2004). But full accounts of the distinctive acquisition of English as a hypercentral language rather than a local language are some way in the future.

What has occupied the vast majority of SLA research are the CL group of learners in classrooms and the group of immigrants to the United States, usually measuring them in terms of either failure to achieve Group A native standards – ‘Relative to native speaker’s linguistic competence, learners’ interlanguage is deficient by definition’ (Kasper & Kellerman, 1997, 5) – or of success at carrying out classroom tasks, not by their ability to function in multilingual situations. The Handbook of Second Language Acquisition (Doughty & Long, 2006) presents 21 or so representative examples of specimen research in shaded boxes spread across its chapters. Three of these give no descriptions of the subjects of the research at all; one is a computer simulation. In the remaining 17, the subjects are L2 learners and all but two groups are adult (one being described as students from two high schools in Britain, odd both because Britain as such is not an educational area and ‘high school’ is a rare designation for schools in England, chiefly used by private schools); they are described as native speakers of English, Arabic etc and as ‘students’ or ‘learners’; in 7 studies a native speaker group features as a control. These pieces of research concern English people acquiring Japanese, French, Korean in English-speaking countries; English being learnt by speakers of Japanese, Arabic, Chinese, Korean, Finnish and Swedish and by four groups of non-native speakers of unspecified L1s; Italian being learnt by English and French native speakers in an unspecified context; French being learnt by Canadians, German by Swedes. None of the situations used by groups B-D appear to be involved (Of course this does not mean that the original research omitted these details, simply that the writers of the summaries did not think them worth mentioning). No details of other languages known by the learners are supplied, yet one in five of the population of the USA speak a language other than English at home (US Census Bureau, 2003), similar figures probably being true of Canada, the UK and most countries.

Nevertheless all the languages being acquired are supercentral or hypercentral; only a few of the learners speak purely local languages: SLA research goes up the hierarchy even if unwittingly. The research is primarily about the CL group; their implied goal is to be as close as possible to Group A native speakers, not L2 users at all. For all our L2 user groups B-G, what counts is the effectiveness of their use within the different contexts rather than the nativeness of their speech; It is only the CL group for which nativeness is relevant, and this only because the teachers, the education system etc and the researchers say so.

SLA research has then been making generalisations about second language acquisition based on the CL group, the only group with no clear goal and no clear community to belong to, and on immigrants considered as supplicants for Group A membership rather than as Group B central language users. Grosjean (1998) showed how careful we must be in specifying the subjects involved in bilingualism research, particularly in terms of language mode. The argument here is that we need to be equally careful in specifying the language groups the learners belong to and want to belong to, rather than treating SLA research as a unified whole. Generalising from the taught CL group is particularly difficult as we cannot isolate the effects of teaching. Interestingly for many of the other groups teaching is not a major concern; it is simply taken for granted that you have to be multilingual in Central Africa or India.

6. Language groups and language teaching

So how can the second language groups B-G be related to second language teaching? We will not deal with Group F, since language teaching is unlikely to be a driving force in this private one-to-one relationship. The sole exception is perhaps the type of teaching known as reciprocal language teaching (Hawkins, 1987), in which pairs of people teach each other their respective languages. Nor will we deal with Group E, which raises a whole host of issues – is Mandarin really a heritage language for an London Cantonese speaker?

Group B learning and teaching of central languages

Historically the teaching of central languages has concerned ethnic minority children and immigrants. Usually these have a local language at home and may be directly taught the language at school or mainstreamed into ordinary classes. They have mostly been measured against the Group A monolingual language (and of course usually found deficient) rather than against its use as a central coordinating language with fellow non-natives as well as with Group A monolinguals. Language teaching needs to recognise Group B L2 use for public functions with Group A officialdom and for social uses with members of Group A and fellow members of Group B. It also needs to take account of the extent to which the overall community is multi-competent or monolingual.

The Adult ESOL Core Curriculum for England (DfES, 2001) provides an example of educational thinking on this topic. The target is four types of learner: settled communities such as Hong Kong, refugees, migrant workers and partners and spouses of learners – all essentially prospective Group B members. It aims at defining ‘in detail the skills, knowledge and understanding that non-native English speakers need in order to demonstrate achievement of the national standards’ (DfES, 2001, 7). In other words, the only goal of ESOL learners is to become part of the Group A rather than part of a Group B: their target behaviour is not multi-competent but monolingual. While the curriculum adapts the delivery of this to the particular needs of ESOL learners, this concerns the means not the target, which is to meet standards appropriate to native speakers, laid down in the master curriculum A Fresh Start (DfEE, 1999). The content of the curriculum is defined in terms of ‘can-do’ statements like ‘can make requests’ and in terms of traditional EFL grammar ‘there is/are + noun (+ prepositional phrase)’, without any distinction between the needs of a monolingual Group A native speaker and a Group B L2 user. Since the learners of ESOL in England are in fact never going to be members of monolingual Group A, what is needed for this and for other situations like it is a curriculum that recognises the particular uses and issues associated with the use of a central language by Group B speakers.

Group C teaching of supercentral languages

One characteristic of supercentral languages is their limited use across borders for a small range of functions. Since the 1960s this specialised functionality has been the concern of many an ESP course, ranging from courses for oilrig-workers to courses for general practitioners; to the consternation of one of my ex-students, her first teaching job was English to a Japanese sex-shop manager.

The other characteristic of the supercentral language is its use for public domains in different countries. The most taught languages in European countries are English (50% of all pupils), followed by German in northern and eastern Europe, French in southern Europe, and Russian in the Baltic and Bulgaria (Eurydice, 2005). Apart from English, the hypercentral language, all of these are supercentral languages with a sphere of influence extending outside their own countries and sometimes outside Europe, say with French and Spanish. Language learning in schools in Europe is overwhelmingly upward in the De Swaan hierarchy towards supercentral languages and the hypercentral language.  

The project set up by the Council of Europe called the Common European Framework aims ‘to facilitate communication and interaction among Europeans of different mother tongues in order to promote European mobility, mutual understanding and co-operation, and overcome prejudice and discrimination’ (Common European Framework, 2008). The Framework achieves this through a comprehensive description of the goals of language teaching, using can-do statements such as ‘I can use simple phrases and sentences to describe where I live and people I know’ and compiling long checklists of different language competences such as socio-cultural competence – ‘Everyday living, e.g.: food and drink, meal times, table manners; public holidays …’. This description is incorporated in the Language Passport (2007), which provides ‘a record of language skills, qualifications and experiences’ through self-rating of six levels of language achievement based on the can-do statements. For example at level A1 (Breakthrough level) Understanding, ‘I can understand familiar words and very basic phrases concerning myself, my family and my immediate concrete surroundings when people speak slowly and clearly’.

While the European Framework insists ‘The language learner becomes plurilingual and develops interculturality’, there is little concession to the actual role and nature of supercentral languages, perhaps because of the EU’s avowed neutral stance between its 23 languages. The overall emphasis in the Framework is on the development of an ability to function like a native speaker. Two goals are ‘Understanding a conversation between native speakers’ and ‘Understanding a native speaker interlocutor’. Native speakers are indeed an important group of users of a supercentral language but the non-native speakers are equally important; the group has a shared need to communicate within a territory. The concept of plurilingual, stressed in the Framework, seems to mean a series of interlocking Group A local languages, not the creation of a multi-competent European society. The role of the supercentral languages of the EU is much more than the creation of cloned native speakers for each other language.

Group D. teaching the hypercentral language

The hypercentral language English is the one that is all things to all people, not confined to a particular territory or a particular function. Its users do not have to take part in a particular society, unlike Group B, or utilise more specific functions in a wider territory, unlike Group C; potentially they use English with anyone anywhere for almost any reason. Group A native speakers of English have no special status, indeed may struggle with some aspects of hypercentral English more than their non-native fellows. Group D speakers retain their own L1 identities while at the same time using an L2 to deal with each other –  language for communication (House, 2003).

While various alternative names have been used for English in this international role, such as World English and English as an international language, the current term seems to be English as Lingua Franca (ELF) (Seidlhofer, 2004). The ELF movement recognises that L2 users are different from native speakers; it does not promote a native speaker standard for teaching. Rather the aim is to teach this particular variety of English for global use for many functions. The emphasis is not on what native speakers do but on establishing and teaching the ELF variety of English. So the phonological syllabus that learners need to use to communicate with each other is described: don’t bother with teaching /D/ (this) and /T/ (thistle) as this causes no communication problems; do take care about making vowels longer before voiced consonants in say ‘bad/bat’, ‘league/leak’, ‘bard/Bart’ (Jenkins, 2002). Similarly the grammar of ELF is in the process of being described through the VOICE (Vienna-Oxford International Corpus of English) project (Seidlhofer, 2004); drop the third person ‘s’ in ‘makes’ and replace infinitive ‘to’ constructions with that-clauses, as in ‘I want that …’.

One meaning of the word ‘language’ is the Lang₂ sense of ‘abstract external entity’ (Cook, 2007c) in Popper’s world 3 of abstract ideas (Popper, 1972), as in ‘the English language’ or ‘the Chinese language’. Lang₂ is something out there in the world of abstractions, like the rules of football, that can be codified in a dictionary and a grammar book. Countries or nations are often linked to Lang₂, as in the aphorism attributed to Weinreich, ‘A language is a dialect that has an army and a navy’: English is the language of England, French the language of France regardless of how many people in these countries have other native languages.

ELF clearly does not fit this conventional Lang₂ framework. It is not an institutionalised object with a set of grammar books and dictionaries; what descriptions exist are descriptions of current usage, not of a notional standard. It is not spoken by a social elite in a particular country but by a heterogeneous international community with nothing in common other than ELF. It is not so much the existing Lang₂ language English used as a Lingua Franca so much as Lingua Franca English, an international variety of its own continually being created for use. Canagarajah (2007) indeed prefers LFE to ELF.

Nor is ELF a question of the choice between local New Englishes. Some varieties of English have indeed become independent of British or American national standards, such as Indian, Singaporean, Australian, and so on; all of these have their own group A local uses for native speakers but also central Group B uses for natives and non-natives; thus Singapore has English as its ‘first’ language with perhaps few native speakers but a large number of Group B central language users from the Tamil, Chinese and Malay communities. Choosing ELF is not identifying with a local variety but with its exact opposite.

So there has been a danger of confusing the two issues. On issue is whether say British English should be taught as a central language in a particular country rather than a local standard, say in India. This is a separate issue from whether to teach ELF as a hypercentral language.. The functions of English hypercentral and central language are different. Indian English is no more or less appropriate as the target of learning than any other language variety. ELF is not speakers of Indian English talking to speakers of Singaporean English, both using their own local English standards. It is instead speakers of Indian English, Canadian English, Japanese, French, or whatever, talking to each other in the global ELF variety: ELF has distinct linguistic differences in phonology, lexis and grammar of its own. Perhaps the only area where the hypercentral language is bound to the old national standards is spelling where the choice worldwide is between British and American styles, with some local variants, e.g. ‘colour’ in Canada and two forms ‘labor’ and ‘labour’ with different uses in Australia (Cook, 2005).

The distinguishing mark of ELF is that it is dynamic on a different time-scale to the changes in other varieties; it is always in the process of instant creation by its users. ‘ELF never achieves a stable or even standardised form’ (Meierkord, 2004, 129). ELF users are mostly capable of smoothing out the difficulties of communicating with each other as they go along; ‘mistakes’, that is to say communication difficulties, are usually remedied by the participants on the spot, though Sampson and Zhao (2003) do provide some hair-raising examples of miscommunication between seamen using ELF. ELF use is characterized by strategies such as Represent, in which the listener echoes the speaker’s remark for purposes such as checking their understanding (House, 2003), and Segmentation, in which the speaker breaks their speech up into smaller chunks that are easier to process (Meierkord, 2004). The characteristics of ELF may be not so much the syntactic patterns and phonemes of the actual product as the processes which give rise to them; another person may produce different forms but use the same overall processes. These processes resemble those involved in other contact situations, such as pidgins, and in the creation of new languages, as in the emergence of Nicaraguan Sign Language (Senghas et al, 2004). ELF is dynamic in the sense that it does not have an established rule-book or standard; it makes it up as it goes along; ELF is a prototype of language as a dynamic system. I have often been asked by teachers what to put in place of the native speaker target for students; my answer used to be that we need a description of what successful L2 users can do, gradually being met by descriptions of ELF. I now feel the answer should be that ELF does not have a final state comparable to that of the Group A native speaker; rather it is a collection of processes that respond to a variety of situations. Teaching is giving people a capability to function, not the knowledge of grammatical rules, lexical items etc. There is no definable end-state in terms of the aspects of language that L2 users know. Hence appropriate tests of ELF success need to be process-based rather than tests of knowledge – measuring students on how well they can communicate to new L2 users rather than testing book mastery of an ELF syllabus of grammatical and lexical items.

This lead to questions about the value of extensive corpus data such as that in VOICE (Seidlhofer, 2004). Collecting and analysing large amounts of people’s speech is an example of Lang₃  –  language considered as set of produced sentences (Cook, 2007c). ELF seems much more a set of processes than a grammar and a vocabulary; it is like describing how people hit the ball in tennis, compared with reporting on their scores. The corpus ‘facts’ of ELF are useful as a basis for working out the processes that produce them but cannot provide a syllabus for language teaching of a conventional kind like the European Framework. Teaching is ideally going to aim at helping people to use these ELF processes rather than any specific grammatical forms or vocabulary. As Canagarajah (2007) says, ‘it is … premature to say if LFE [Lingua Franca English] is teachable like other languages in a product-oriented and formalistic manner.’

The dynamic nature of ELF raises a doubt about the proposal for a phonological syllabus based on language learners’ comprehension problems. Learners are still learning; i.e. it is doubtful whether they have sufficient command of the ELF variety to act as a model as assumed in the learner-based syllabus of Jenkins (2002). Pronunciation difficulties should at best be established from experienced ELF users with different L1s, not from inexperienced classroom learners. And they should be based on the processes necessary to the production of ELF English, such as vowel lengthening, rather than on individual phonemes such as /D/.

7. Conclusions

So language teaching has to be clear whether it is teaching:

-      a local language to people who want to take part in a monolingual local language community, whether Finnish in Finland or Basque in Spain

-      a central language to people who want to take part in a multilingual community where the language is used, say English in London or Delhi

-     a supercentral language to people who want to use it for specialist cross-national uses, say French for diplomacy

-     a hypercentral language to people who want to use it for a range of purposes across the globe.

But most language teaching has not been concerned with these issues. It teaches language – a standard native homeland monolingual variety believed to be good for all purposes and all people. It cannot be assumed that all students of English now need it as a hypercentral language, or that all English children need French as a local language for visiting France, or that all learners of Chinese are heritage learners; and so on. We need to see how language teaching can better prepare people for use of second languages in the diverse situations in which they find themselves in the twenty-first century.

References

ASD Simplified Technical English Maintenance Group (STEMG) (2007), Simplified Technical English. http://www.simplifiedenglish-aecma.org/).

Baetens Beardsmore, H. (1993), The European School Experience in Multilingual Education. Brussels: Vrije Universiteit Brussel/Universite Libré de Bruxelles.

Baker, P. and Eversley, J. (2000), Multilingual Capital, London: Battlebridge.

Block, 2006

Bloomfield, L. (1926), ‘A set of postulates for the science of language’, Language 2, 153‑64.  Reprinted in M. Joos (ed.) (1957), Readings in Linguistics I. Chicago: University of Chicago Press.

Brecht, R.D. & Powers, W.P. (2005), ‘Language needs analysis at the societal level’, in M.H. Long (ed.), Second Language Needs Analysis, Cambridge: Cambridge University Press, pp. 79-104.

Brutt-Griffler, J. (2002), World English: a Study of its Development, Clevedon: Multilingual Matters.

Canada (2007) 2006 census, Canada stats,

Canagarajah, S. (2007), ‘Lingua Franca English, multilingual communities and language acquisition’, Modern Language Journal, 91, 923-939.

Chambers, J., Trudgill, P. and Schilling-Estes, N. (eds.) (2001), The Handbook of Language Variation and Change, Oxford: Blackwell

Clachar, A. (1997), ‘Ethnolinguistic identity and Spanish proficiency in a paradoxical situation: the case of Puerto Rican return migrants’, J.M.M.D, 18, 2, 107-124

Common European Framework of Reference for Languages (2008), Strasburg: Council of Europe. http://www.coe.int/t/dg4/linguistic/CADRE_EN.asp, consulted January 2008

Cook, V.J. (1991), 'The poverty-of-the-stimulus argument and multi-competence', Second Language Research, 7, 2, 103-117

Cook, V.J. (2003), 'The changing L1 in the L2 user's mind ' in V.J. Cook (ed.) Effects of the Second Language on the First, Clevedon: Multilingual Matters, pp. 1-18.

Cook, V.J. (2005), The English writing system, London: Edward Arnold.

Cook, V.J. (2007a), ‘Multi-competence: black hole or wormhole for SLA research?’ in Z.H. Han (ed.), Understanding Second Language Process, Clevedon: Multilingual Matters, pp. 16-26. online

Cook, V.J. (2007b), ‘The goals of English Language Teaching: reproducing native-speakers or promoting multi-competence among second language users?’ in J. Cummins & C. Davison (eds), International Handbook on English Language Teaching, Kluwer, pp. 237-248.

Cook, V.J. (2007c), ‘The nature of the L2 user’, in L. Roberts, A. Gurel, S. Tatar & L. Marti (eds.), EUROSLA Yearbook, 7, 205-220. online

Cook, V.J. (t.a.), ‘Multilingual Universal Grammar as the norm’, in I. Leung (ed.) Third Language Acquisition, Multilingual Matters 000. online

Cook, V.J.  & Newson, M. (2007), Chomsky's Universal Grammar: An Introduction. Third edition Oxford: Blackwell.

Curran, C.A. (1976), Counselling-Learning in Second Languages, Apple River Illinois: Apple River Press.

d’Armond Speers (2006), http://en.wikipedia.org/wiki/Klingon_language).

De Bot, K., Lowie, W. & Verspoor, M. (2005), second language acquisition: an Advanced Resource Book, Routledge.

de Swaan, A. (2001), Words of the World: The Global Language System, Cambridge: Polity Press.

Department for Education and Skills (DfES) (2001), The Adult ESOL Core Curriculum. London: DfES.

Department for Education and Employment (DfEE) (1999), A Fresh Start – Improving Literacy and Numeracy, London: DfEE.

Dorian, N.C. (1981), Language death: the life cycle of a Scottish Gaelic dialect, Philadelphia: Philadelphia University Press.

Doughty, C.J. & Long, M.H. (eds.) (2006), The Handbook of second language acquisition, Oxford: Blackwell.

Dulay, H. & Burt, M. (1973), ‘Should we teach children syntax?’, Language Learning 3, 245-57.

Eurydice (2005). Key Data on Teaching Languages at School in Europe. Eurydice European Unit (http://www.eurydice.org).

Fishman, J.A. (1991), Reversing Language Shift, Clevedon: Multilingual Matters.

Gordon, R.G., Jr. (ed.) (2005), Ethnologue: Languages of the World, Fifteenth edition. Dallas, Tex.: SIL International. Online version: http://www.ethnologue.com/.

Graddol, D. (2006), English Next. London: the British Council. http://www.britishcouncil.org/files/documents/learning-research-english-next.pdf

Grosjean, F. (1998), ‘Studying bilinguals: Methodological and conceptual issues’, Bilingualism: Language and Cognition, 1 (2), 131—149.

Gumperz, J.J. & Levinson, S.C. (1996), ‘Introduction: linguistic relativity re‑examined’, in J.J. Gumperz & Levinson, S.C. (eds.), Rethinking Linguistic Relativity, Cambridge: Cambridge University Press, pp. 1-18.

Hannan, M. (2004), A study of the development of the English verbal morphemes in the grammar of 4-9 year old Bengali-speaking children in the London borough of Tower Hamlets. Ph.D., University of Essex.

Hawkins, E. (1987), Modern Languages in the Curriculum. 2^nd edition, Cambridge: Cambridge University Press.

Herdina, P. & Jessner, U. (2002), A Dynamic Model of Multilingualism: Changing the Psycholinguistic Perspective, Clevedon: Multilingual Matters.

House, J. (2003), ‘English as a lingua franca: a threat to multilingualism?’, Journal of Sociolinguistics, 7, 556-578.

Jenkins, J. (2002), ‘A sociolinguistically based, empirically researched pronunciation syllabus for English an international language’. Applied Linguistics 23, 1, 83-103.

Jenkins, J. (2006), ‘Points of view and blind spots: ELF and SLA’, International Journal of Applied Linguistics, 16, 2, 137-162.

Kanno, Y. (2000), 'Bilingualism and identity: The stories of Japanese returnees', International Journal of Bilingual Education and Bilingualism, 3, 1, 1-18.

Kasper, G. & Kellerman, E. (eds) (1997), Communication strategies: psycholinguistic and sociolinguistic perspectives. London: Longman.

Kellerman, E., Koonen, H. & van der Haagen, M. (2005), ‘“Feet speak louder than words”: a preliminary analysis of language provisions for professional footballers in the Netherlands’, in M.H. Long (ed.), Second Language Needs Analysis, Cambridge: Cambridge University Press, pp. 200-224.

Kelly Hall, J., Cheng, A. and Carlson, M.T. (2006), ‘Reconceptualising multi-competence as a theory of language knowledge’, Applied Linguistics, 27, 2, 220-240.

Kerswill, P. (1994), Dialects converging: Klein, 1986, Rural speech in urban Norway. Oxford: Oxford University Press.

Klein, W. (1986), Second Language Acquisition. Cambridge: Cambridge University Press.

Klein, W. & Dittmar, N. (1979), Developing grammars: the acquisition of German syntax by foreign workers. Berlin: Springer-Verlag.

Labov, W. (1966), The Social Stratification of English in New York City, Washington DC: Center for Applied Linguistics.

Laitin, D.D. 2000. ‘What is a language community?’, American Journal of Political Science, 44, 1, 142-145.

Language Passport, The (2007), Strasburg: Council of Europe/London: National Centre for Languages.

Li Wei (1994), Three generations, two languages, one family: language choice and language shift in a Chinese community in Britain. Clevedon, Avon: Multilingual Matters.

Mackey, W.F. (1972), ‘The description of bilingualism’, in Fishman, J.A. (ed.): Readings inthe Sociology of Language. The Hague: Mouton, 554-584.

Meierkord, C. (2004), ‘Syntactic variation in interaction across international Englishes’, English World-Wide, 25, 109-132.

Myhill, J. (2003), ‘The native speaker, identity and the authenticity hierarchy’, Language Sciences, 25, 1, 77-97.

Norton, B. (2000), Identity in Language Learning, London: Longman.

Patrick, P. (2001), ‘The speech community’, in J.K. Chambers, P. Trudgill & N. Schilling-Estes (eds.), Handbook of Language Variation and Change. Oxford: Blackwell.

Pienemann, M. (1998), Language Processing and Second-Language Development: Processability Theory, Amsterdam: John Benjamins.

Piller, I. (2002), ‘Passing for a native speaker: identity and success in second language learning’, Journal of Sociolinguistics, 6, 2, 179-206.

Popper, K.R. (1972), Objective Knowledge. Oxford: Clarendon Press.

Sampson, H. and Zhao, M. (2003), ‘Multilingual crews: communication and the operation of ships’, World Englishes, 22, 1, 31-43.

Schmid, S. (1994), L'italiano degli spagnoli. Interlingue di immigranti nella Svizzera tedesca. Milano: Franco Angeli.

Seidlhofer, B. (2004), ‘Research perspectives on teaching English as a lingua franca’, Annual Review of Applied Linguistics, 24, 209-239

Senghas, A, Kita, S. & Asli Özyürek, A. (2004), ‘Children creating core properties of language: evidence from an emerging sign language in Nicaragua’, Science 17 September Vol. 305. no. 5691, 1779–1782.

Siegel, J. (2006), ‘Social Context’. In Kroll, J. & De Groot, A. (eds), Handbook of Bilingualism. Oxford: Oxford University Press, pp. 178-223.

Simple English Wikipedia (2008), http://simple.wikipedia.org/wiki/Wikipedia:How_to_write_Simple_English_articles.

Skehan, P. (1998), A Cognitive Approach to Language Learning, Oxford: Oxford University Press.

Smolicz, J.J. & Secombe, M. (2003), ‘Assimilation or pluralism? Changing policies for minority languages education in Australia’, Language Policy, 2, 3-25.

Thorpe, K., Greenwood, R., Eivers, A. & Rutter, M. (2001), ‘Prevalence and developmental course of “secret language”’, International Journal of Language and Communication Disorders, 36, 1, 43-62.

US Census Bureau, (2003), http://www.census.gov/Press-Release/www/releases/ archives/ census_2000/001406.html.

Valdés, G. (2005), ‘Bilingualism, heritage language learners and SLA research: opportunities lost or seized?’, Modern Language Journal, 89, iii, 410-426.

Vatican City (2006), http://www.ibiblio.org/expo/vatican.exhibit/exhibit/a‑vatican_lib/
City_recovers.html)

Voice of America (VOA) (2008) ‘Special English 

Weeks F., Glover A., Johnson E., and Strevens, P. (1988), Seaspeak Training Manual: Essential English for International Maritime Use. Oxford: Pergamon.

Wierzbicka, A. (1996), Semantics: Primes and Universals. Oxford: Oxford University Press.