The native speaker teacher. Theoretical considerations and practical implications

1 Introduction

In this paper, we will look at the concept of the native speaker (NS), which is central to so many studies into language both theoretical (e.g. Chomsky’s concept of Universal Grammar) and applied (i.e. in language teaching and assessment). We will firstly (Section 2) look at the use of the concept of NS as a model within language teaching and assessment. In Section 3, we will discuss the concerns of those who oppose the centrality of the native speaker in linguistic theory, in language teaching and assessment. Finally, in Section 4, we will look at the role that NS teachers may play in language teaching, and at whether, as conventional wisdom would have it, learners really do acquire a language more effectively when taught by a NS of the target language.

In our discussion, we will draw on examples mainly from English, which is not to say that the considerations that we make have no relevance to other languages. We can hardly discuss a language like English without taking into account its international role as a lingua franca. In such contexts, where the number of non-native speakers (NNSs) exceeds that of NSs, and where most interactions are between NNSs using the language largely for instrumental purposes (e.g. Graddol 2007 or Kachru 1985), the relevance of the NS/NNS dichotomy is less than that found in a more “confined” language where most speakers are NSs, and most interaction is NS–NS.

2 The native speaker as a model in language teaching and assessment

We do not have space here to go into the details of the demonstrably very different processes of first and second language acquisition (FLA and SLA respectively).^[1] We shall take it as a given that in FLA, so-called learners acquire their first language (L1) instinctively due to some innate mechanism, which Chomsky (1965, 1968 originally called the Language Acquisition Device, but which he later subsumed into his concept of universal grammar (1981). We shall also assume the truth of the critical period hypothesis (Chomsky 1965; Lenneberg 1967): the posited existence of an optimal age range for FLA, roughly the period up to puberty and declining afterwards.^[2]

Perhaps the most important, and most obvious, difference between FLA and SLA lies in their outcomes. FLA is typically successful and almost all learners reach the highest levels of proficiency^[3] – excepting those who suffer from some kind of neurological condition such as aphasia that may impede language acquisition. By contrast, SLA does not result in all learners reaching a high level of proficiency. Indeed, SLA learners, even those who have studied together, may attain a wide range of different levels. Only a small minority reach the advanced levels, which may be loosely compared to that of the NS, something that Selinker (1972) put down to fossilization. The existence of various levels of competence for SLA learners, as manifested in scales like those of the Common European Framework of Reference for Languages: learning, teaching, assessment (CEFR) – Council of Europe (2001) provides the need for certification to enable learners to demonstrate the level that they have achieved in the L2 in question: almost an obligation for L2 learners but rarely, if ever, required for L1s.^[4]

It is also interesting to note that the position of the NS as regards scales such as the CEFR is unclear. It may seem reasonable to assume that their natural place lies at the top especially because the NS was typically used as the model in traditional second and foreign language teaching. The implication then is that a NNS may progress up the levels (e.g. A1, A2, B1 etc. on the CEFR) and eventually, if they have the aptitude, opportunity, and make the effort reach “native speaker level” (somewhere above C2 presumably), in effect becoming a NS.^[5] In our experience, this constitutes a popular misconception, and not only among non-experts. For example, we have seen numerous calls for applications for teaching assistant contracts that have listed “native speaker” as one of the requisites. This has then been interpreted by the selection committee to include applicants who declare themselves to be NSs purely by dint of possessing certification at C1 or C2 level,^[6] In such calls, discerning who is and who is not a NS is, in practice, often tackled firstly by self-certification, and then at the interview stage, if there is one.

In fact, even if such a thing were possible, or justified, there are still major problems with using the NS as a model. One of these, as discussed at length by Christiansen (2018a), is the fact that, with almost any language, other than those spoken by small compact communities, NSs will speak many different varieties, both regional and social (e.g. class, age, professional), and thus no single immutable model will exist. This is particularly obvious with international languages like English, Spanish or Arabic because, due to historical diasporas, notable differences exist even at the highly idealised level of national standards. Using a NS model as the basis for a scale is thus like using a piece of elastic to measure something.

In practice, it is extremely rare for NNSs to reach a level where they may be taken by a NS to be a genuine NS (something that may be explained also by such things as acculturation).^[7] This points to the fact that there are fundamental differences between the two kinds of speaker.

Furthermore, NSs are typically far from perfect or error-free users of their L1 (least of all the standard version, which is often the target variety of a language course within formal education). It is for this reason that Chomsky (1965) famously distinguishes between competence (knowledge, potential ability) and performance (actual use of language).

What is more, some NNSs may actually be better than some NSs at certain things which they have learnt how to do in the process of SLA (especially when they are studying the language for further study in formal education); as the CEFR states (Council of Europe 2001: 249), describing the ALTE (Association of Language Testers in Europe) framework:

ALTE Level 5 (Good User): the capacity to deal with material which is academic or cognitively demanding, and to use language to good effect, at a level of performance which may in certain respects be more advanced than that of an average native speaker.

To assume that, on any unified scale applicable to both native and NNSs, the NS would automatically represent the higher levels is therefore simplistic.

It was partly due to these considerations that led the CEFR to adopt the approach of defining levels, not by comparison to an idealised NS, but by so-called “Can Do” statements designed to measure observable performance rather than abstract competence. Indeed, the fact that the NS is not the intended model of the CEFR is made abundantly clear:

Level C2, whilst it has been termed ‘Mastery’, is not intended to imply native-speaker or near native-speaker competence. What is intended is to characterise the degree of precision, appropriateness and ease with the language which typifies the speech of those who have been highly successful learners. Descriptors calibrated here include: convey finer shades of meaning precisely by using, with reasonable accuracy, a wide range of modification devices; has a good command of idiomatic expressions and colloquialisms with awareness of connotative level of meaning; backtrack and restructure around a difficulty so smoothly the interlocutor is hardly aware of it.

3 Opposition to the centrality of the native speaker in linguistic theory and language teaching and assessment

Most scholars within theoretical linguistics have focussed, like Chomsky, exclusively on the production and behaviour of NSs when describing languages: a fact that Coulmas (1981: 5), cited in Seidlhofer (2011: 32) laments:

He [the NS] is the one who can legitimately supply data, and his language is what grammatical analyses are meant to account for. Thus, nativeness is the only universally accepted criterion for authenticity.

However, in specific areas of linguistics such as some aspects of social linguistics, including studies into Creoles, and the growing field of English as a Lingua Franca (ELF), this refusal to consider the NNS has been opposed. In applied linguistics (above all the teaching of English to Speakers of Other languages), the nativeness principle (for the use of NSs as models for NNS learners – see Levis 2005)^[8] has been widely criticised particularly within studies into ELF, a specific domain of language use where the majority of the participants are NNS, not NS. Indeed, Christiansen (2017) shows that, in many contexts of English language teaching, adherence to the nativeness principle does not run as deep as one might have thought, at least not in the case of English.

Through analysis of the reactions of L2 learners of English to different statements, Christiansen (2017) highlights that, although most respondents agree with the idea of wanting to speak like a NS of English,^[9] the majority of the other opinions that they favour are more consistent with an “ELF-oriented” approach to English language learning: i.e. their priorities are in fact not assimilating themselves with the NS community but rather in being able to communicate effectively with speakers from around the world (not just NS of English). Furthermore, it also emerged that learners were not enthusiastic about hiding their origins or identity while speaking English – something that contradicts their stated desire to speak like a NS. In summary, Christiansen states (2017: 75):

What appears to have happened in this survey is that most respondents consciously pay lip service to the nativeness principle while unconsciously setting themselves goals and harbouring attitudes and that are more coherent in an ELF-oriented mind-set.

In two further studies based on subjects’ assessments of recordings of NSs of English and NNSs (Christiansen 2018b, 2019), it emerges that learners are not drawn only to admire (and presumably thus to emulate) the English of NSs but that also of celebrities and famous people whom they may look up to (whether a NS of English or not), or of people like themselves with whom they can feel affinity (again regardless of whether these are a NS of English or not).

The idea that the distinction between NS of English and NNS is not so relevant in the context of ELF leads to Graddol’s (2007: 110) reworking of Kachru’s famous “three circles of English” model (1985). The latter divided English speakers into three groups: an inner circle of L1 speakers; an outer circle of second language speakers; and an expanding circle of foreign language speakers. By contrast, at the centre of his circle, Graddol places not NSs but “highly proficient users”, with progressively less proficient ones moving out from the centre in a succession of concentric circles. Graddol’s representation is in tune with Kachru’s changing views on the usefulness of the NS/NNS distinction (2005: 210):

[…] it is obvious that the cross-cultural and localized functions of Englishes have now made the dichotomy of native versus non-native theoretically and functionally questionable.

There are other reasons for making less of the fact of whether a speaker is NS or not. Focussing only on the NS, and ignoring the contribution of NNSs, gives not only an incomplete picture of the kind of discourse taking place in specifically ELF contexts but may lead to researchers missing some important trends that can eventually affect the evolution of the English language in general (Seidlhofer 2005: 339–340):

Despite being welcomed by some and deplored by others, it cannot be denied that English functions as a global lingua franca. However, what has so far tended to be denied is that, as a consequence of its international use, English is being shaped at least as much by its non-native speakers as by its native speakers. This has led to a somewhat paradoxical situation: on the one hand, for the majority of its users, English is a foreign language, and the vast majority of verbal exchanges in English do not involve any native speakers of the language at all. On the other hand, there is still a tendency for native speakers to be regarded as custodians over what is acceptable usage.

The contribution that NNSs make to the evolution of languages is often overlooked. Recently, scholars have identified the way that such phenomena as macroacquisition or social second language acquisition ^[10] can be seen to leave a lasting mark on languages. McWhorter (2007) calls these phenomena non-native acquisition and makes the interesting general observation that it is precisely the languages that have spread beyond their original speech communities – such as English, Mandarin Chinese, Persian (Farsi), Colloquial Arabic, and Malay – that have become relatively more grammatically simple over time. This simplicity originates in the fact that the NNS learners fail to grasp all the intricacies of the target language, and yet their unorthodox way of speaking nonetheless contributes to the input to which successive generations (including infants subject to FLA) are exposed. Christiansen (2021) posits that this was a major factor in the evolution of English already at its origins within the British/Irish Isles, long before it became a global language spoken all over the world.

Finally, beyond these observations on the misguidedness of any approach in general linguistics that concentrates on the NS alone, other scholars are critical of the very existence of the concept of NS, questioning its validity and the way that it is defined. Cook (1999: 185–186) claims:

This core meaning of native speaker is often supplemented by detailing the characteristics that native speakers share apart from their birth. Stern (1983) lists: (i) subconscious knowledge of rules, (ii) intuitive grasp of meanings, (iii) ability to communicate within social settings, (iv) range of language skills, and (v) creativity of language use. The Encyclopedic Dictionary of Applied Linguistics (Johnson and Johnson 1998) adds (vi) identification with a language community. Davies (1996) adds (vii) ability to produce fluent discourse, (viii) knowledge of differences between their own speech and that of the “standard” form of the language, and (ix) ability “to interpret and translate into the L1 of which she or he is a native speaker.”

Some of these characteristics are in a sense obvious: native speakers are not necessarily aware of their knowledge in a formal sense (i and ii), but nor could they explain how they ride a bicycle. Some are debatable: many native speakers are unaware how their speech differs from the status form (viii), shown for example in the growing use of non-standard between you and I for between you and me even in professional speakers such as newsreaders. Many native speakers are far from fluent in speech (vii), some having to communicate via alternative means, such as Stephen Hawking and Helen Keller. Some native speakers function poorly in social settings (iii). In the Chomskyan sense of creativity, any novel sentence uttered or comprehended is creative (v); a computer can create ‘new’ sentences, for instance the speech program that answers telephone directory enquiries with every possible telephone number. In a general literary sense, creativity belongs to a small percentage of native speakers, such as poets, rap singers and so on. The ability to interpret (ix) is only possessed by native speakers with a second language and not necessarily by all of them. Native speakers are free to disassociate themselves completely from their L1 community politically or socially (vi) without giving up their native speaker status, whether Karl Marx in London, James Joyce in Zurich or Albert Einstein in Princeton.

These characteristics are then not only variable but also in a sense accidental; lack of any of them would not disqualify a person from being a native speaker. A monk sworn to silence is still a native speaker. Many are also shared by non-native speakers almost regardless of their level of proficiency in the language: non-native speakers show a rapidly developing awareness of gender-linked pronunciation (Adamson and Regan 1991) and of the status of regional accents (Dailey-O’Cain 1998); what level of L2 English did it take for Marcel Duchamps to create ‘surrealistic aphorisms’ such as My niece is cold because my knees are cold (Sanquillet and Peterson (1978), p. 111)?

Some of Cook’s (1978) criticism seems a little strained, even disingenuous. In a Chomskyan sense, the facet of a syntactic system to produce, via a limited set of rules (or principles if one likes),^[11] an infinite number of structures (also through embedding and recurrence) is a constituent characteristic of language. While it is true that artificial intelligence and computer programs can now replicate this ability pretty convincingly, it is an ongoing philosophical question whether what they produce can be considered authentic language, at least in the human sense, but one which unfortunately we do not have space for here.^[12] Furthermore, someone like Stephen Hawkings, by means of his computer, or Helen Keller using a sign language, is still a “speaker” (i.e. language user), even if they may not be considered “fluent” in a conventional sense. For the same reason, a Trappist monk can still be classed as a NS, because speaker, in this sense, indicates that someone has the competence to process and use the language (e.g. to listen to mass, to pray silently), not necessarily to communicate with it exclusively via their so-called vocal organs.

Those points aside, Cook does succeed in showing that definitions of NS, though satisfactory when applied to general or typical cases, are often lacking when it comes to individual ones. The fact that there are “exceptions” or cases where neither the definition for NS nor NNS perfectly applies is indication that the relationship between the two is not one of discrete categories but of states on the same continuum: a situation of “more or less” rather than “either … or”. This makes perfect sense when one remembers that one is speaking about individual human beings; even if language acquisition is to a large degree biological and genetically determined, precise time scales may vary (as they do say for puberty) and environmental factors (diet, health, social context during childhood) may also influence outcomes (as they do such things as height and weight reached in adulthood). There is ample possibility for the neat division between FLA and SLA to become blurred and confused. In specific cases, due to the life experiences of the individual in question, the process of acquisition of a given language may display features of both, and thus be hard to classify.

This is also something which we have had the opportunity to see in calls for applications for posts of NS teaching assistants at schools and universities, where, unlike in the example of self-certification we mentioned in Section 2, some attempt is made to have applicants demonstrate that they are NS rather than just declare so by means of a vague declaration. For example, among other things, one might, at the most simplistic level, try to solve this problem by asking applicants to prove that they are a national of a country where the target language is the official L1, de jure or de facto.^[13] Such a view is ethno-nationalistic and assumes wrongly that countries are rigorously monolingual, or that citizens of a given country are inevitably born there. Because of this, it would not only fail to exclude NNS who are naturalized citizens, but also fail to include NSs, who for whatever reason, are nationals of a country where their L1 is not the official language (e.g. a refugee or asylum seeker who becomes a naturalized citizen of their new home, and renounces the nationality of the country that they have fled from).^[14] It would also be difficult to apply in the case of nationals of countries with multiple official languages at a regional and state level (e.g. Belgium, Eire, India, South Africa, or Switzerland).

Alternatively, adopting a relatively more sophisticated solution, a call may require that an applicant must declare (and thus provide the proof of the same were there to be some legal challenge) that they have attended school for the entire period of their scholastic career in a country where the language is spoken. While such a definition would probably be successful in excluding a NNS, it may also exclude some applicants who are legitimately NS, for example those who are bilingual with parents of different L1s, having acquired both of these at home in the family environment during the critical period of their language development. Furthermore, other NS candidates may not have completed all their education in a country of their L1; they may have spent periods in other countries where other languages were used. They may indeed have lived permanently in a country where another language is spoken but attended schools which used their L1 as a medium of instruction e.g. at the “La scuola d’Italia Guglielmo Marconi” in New York (for a NS of Italian). In either of these cases, the candidate may constitute a genuine NS, but be excluded by the call.

Indeed, in such complicated or borderline cases, perhaps the best approach is to ask the simple question: “if such and such a language is not the applicant’s L1 then which one is?” It being a given, we assume, that (except in extreme and highly unusual cases regarding brain damage or complete childhood isolation) everybody must be a NS of at least one language. Consequently, any definition that allows for the possibility that a person has none is irreparably flawed.

4 The role of native speaker teachers in language teaching

As we have shown, the NS no longer formally serves as a model in any language teaching and assessment that is in line with the CEFR, or similar systems, which must account for the majority of courses not only in Europe, but also around the world. One might also assume that a NNS teacher, i.e. someone who has learnt the target language by means of SLA, would be the person in the best position to guide their learners along the same path, not least if they share the learner’s L1. Such a teacher should be aware of the kind of language transfer that may assist or impede their learners in this process.

The main advantage the NS offered in traditional contexts, especially when technology was far more primitive than that which is available today both to university language centres/schools and to individual learners, was that such teachers provided a living model of the target language, above all as regards pronunciation, up-to-date authentic usage and idioms. Nowadays of course, internet provides infinite possibilities to practice the language, and expose oneself to a whole range of different models, so the need for such NS specimens is diminished.

Notwithstanding this, the figure of the NS teacher remains a much sought after commodity even though various scholars and practitioners have questioned the real value of such educators. Graddol goes so far as to label this phenomenon “the native speaker problem” (2007: 14):

Native speakers of English have enthusiastically promoted the learning of their language abroad. By the end of the 20^th century, less effort seemed to be required, as learning English became seen no longer as an option but as an urgent economic need. NSs were regarded as the gold standard; as final arbiters of quality and authority.

In the new, rapidly emerging climate, NSs may increasingly be identified as part of the problem rather than the source of a solution. They may be seen as bringing with them cultural baggage in which learners wanting to use English primarily as an international language are not interested; or as ‘gold plating’ the teaching process, making it more expensive and difficult to train teachers and equip classrooms. Native speaker accents may seem too remote from the people that learners expect to communicate with; and as teachers, native speakers may not possess some of the skills required by bilingual speakers, such as those of translation and interpreting.

The fact that NSs are regarded as “gold plating” is something which, as Rinvolucri (2001: 41–42) points out, private language schools have made capital out of by promising to provide NS teachers using the so-called direct method (where the learner’s L1 is effectively banished from the learning process), which free state education systems can rarely do:

When we look back over the past 25 years, it is clear that a lot of money has been made by many schools and a little money has been made by many teachers through sustaining and propagating the view that the native speaker of English who does not know the students’ language is the teacher the students will learn most and best from.

Partly, through the efforts of teaching practitioners like Rinvolucri (2001), as well as scholars like Graddol (2007), Selinker (1972) or Corder (1967), the direct method has been largely discredited, at least among experts,^[15] and the positive aspects of language transfer from the learner’s L1 have been recognised, rather than being viewed as unwelcome linguistic “interference”: the major cause behind the learners’ errors.^[16] Furthermore, studies into what is now more positively described as translanguaging (See García 2009; García and Li 2014; Li 2018) have highlighted the fact that, very much in tune with Chomsky’s concept of universal grammar (and the idea that, at their core, all languages share the same common features and principles), plurilingual^[17] language users (who are perhaps the majority of people on the planet)^[18] do not separate their linguistic repertoires into separate hermetic containers (one for each language), but rather treat the sum of their linguistic competence as a common resource that may be searched and selected from when communicating and expressing themselves. Following this reasoning, language learners are encouraged to use their substantial linguistic knowledge of their L1(s), and of any other L2s, when acquiring a foreign or second language.

It is also interesting to consider that, while there has been in many countries (or rather markets in the commercial sense) enthusiastic promotion of the NS teacher as the deluxe product, this practice has thankfully never really extended to other areas related to language teaching such as: testing (e.g. international certification); materials production (e.g. authoring of course books); or academia either (e.g. all branches of linguistics). In these areas (in particular those relating to English, no doubt by dint of its international status and polycentric nature),^[19] the dichotomy between NS and NNS is clearly not so relevant. Numerous NNSs occupy important and high profile positions in such fields. This state of affairs, though rarely even alluded to, is yet another indication of the fact that whether someone is a NS or not is no guarantee that their contribution will, on some vague level, be worth more than that of a NNS in areas related to specific languages or the teaching or testing of them.

To return to our original discussion, the kind of teacher that Graddol (2007: 14) and Rinvolucri (2001: 41–42) decry is not in fact the NS English teacher per se, but more specifically the monolingual NS of English teacher using the outdated direct method (but of course if the teacher has no other way of communicating than English then they are obliged to use such an approach).^[20] However, it should be remembered that by no means does this apply to all NS English teachers; to speak of “the native speaker problem” is misleading. In fact, many such teachers are plurilingual (a common route into language teaching is after all the study of languages), and a substantial proportion, particularly those working outside of English-speaking countries, will be teaching students’ mostly of the same L1, which they will probably have some knowledge of as a resident of that locality. One should then speak rather of “the monolingual teacher problem”.^[21]

Another, equally important issue highlighted in the Graddol (2007: 14) quote above is the complicated area of culture, in particular the traditional idea that learning a language involves also learning the associated culture. As countries adopt English for the strictly instrumental goal of international communication, traditional nation-specific culture becomes less relevant (e.g. Thanksgiving in the USA and Canada, Vegemite in Australia, or red phone boxes in the UK). Such an emphasis on foreign culture may even be unwelcome if the country in question wants to protect its own identity and values against creeping globalisation. At an extreme, totalitarian regimes have always lived in fear of the outside influences to which language learning may expose learners (e.g. liberal democracy, equality of the sexes, religious tolerance, or LGBTQ+ rights). A NS teacher may, from this perspective, be seen as a threat especially if they do not understand the often-invisible lines that must not be crossed in the classroom.

In the case of other languages, those which are less international so-to-speak – those confined to specific geographical regions and ethnic communities (for example Yoruba, Finnish, Maltese or to a degree Hindi or even varieties of Chinese) – language and culture may still be very closely linked. Consequently, NS teachers may also possess not only valuable linguistic competence but also what the CEFR terms sociolinguistic competence (Council of Europe 2001: 118–122) and socio-cultural competence (Council of Europe 2001: 220).

It is however difficult to apply such concepts in the case of ELF, which in reality consists not in a specific and fixed variety of English as do, for example, Cockney, Glaswegian, or Jamaican Patwa. Instead it is manifested as a set of transitory variations which are more fluid and ad hoc, and subject to change on a case by case basis depending on such factors as: the context of situation; the use to which the language is being put; and who the participants in the speech event are.^[22] With ELF, being a NS of English would not necessarily equip one to deal competently with the specific kind of discourse required; the average NS of English, familiar mainly with NS-to-NS discourse, may lack the necessary communicative skills (e.g. accommodation strategies)^[23] and, if monolingual, the ability to translanguage.

Furthermore, Graddol (2007: 114) argues in effect that the very presence of a NS of English might in some way inhibit the rest of the participants (presumably, because they suffer some kind of inferiority complex due to their self-perceived lower linguistic level).

Global English is often compared to Latin, a rare historical parallel to English in the way that it flourished as an international language after the decline of the empire which introduced it. The use of Latin was helped by the demise of its native speakers when it became a shared international resource. In organisations where English has become the corporate language, meetings sometimes go more smoothly when no native speakers are present. Globally, the same kind of thing may be happening, on a larger scale.

This is not just because non-native speakers are intimidated by the presence of a native speaker. Increasingly, the problem may be that few native speakers belong to the community of practice which is developing amongst lingua franca users. Their presence hinders communication.

Such an observation is merely anecdotal but no doubt based on Graddol’s wide experience as a linguist and observer of language education and language policy around the world. Following this logic, it might be predicted that learners may find being taught by a NS more off-putting than by a NNS, especially someone of their own linguacultural background. To our knowledge, there is little convincing research on this point, which is not surprising when one considers the myriad variables (e.g. attitudes and background of the individual learner; attitudes and personality of the teacher in question; teaching context, atmosphere in class, teaching methodology used, and so on) that would have to be isolated, measured, and weighed up before one could attempt to draw any conclusions.

5 Conclusions

It is perhaps paradoxical that, despite the attention given to the nativeness principle by its proponents and opponents, the issue of whether a teacher is a NS or NNS is not as important as people typically think. As Graddol (2007) and Rinvolucri (2001) point out, the benefits of NS teachers have been subject to a degree of exaggeration by those selling language courses. Furthermore, the policy of promoting NS teachers as something distinct and special can backfire on NS teachers themselves. While they may look down on their generally less linguistically competent NNS colleagues, they may themselves be the object of some resentment on the part of the latter (who often may be higher up in the institutional hierarchy).^[24] The latter may see them as qualified only because of their innate nativeness (a gift that they have been born with so to speak, not worked for). The NNS may ignore the NSs’ other qualifications (which may in any case be of a kind unfamiliar to them) or professional expertise, and in such a way regard them as little more than “native informants”, not equals, essentially providing the service of living model, which, as we note above in Section 4, is rapidly being offered also by technology and artificial intelligence.^[25]

In fact, what helps learners learn most is arguably not a NS teacher per se, nor indeed a linguistically highly-competent teacher per se, but rather a good all-round teacher. This means someone who has a high level of linguistic competence (ideally but not necessarily NS), but who also has a whole set of characteristics among which: being experienced and well-trained in their craft; having been a successful language learner themselves and thus being able to empathise with the students in their efforts; and having a rich linguistic repertoire and familiarity with different cultures, which enables them to draw comparisons between the lingua-cultural system that they are teaching and other languages/cultures, ideally, but not necessarily, those of their learners.

An analogy we can make with the policy of only employing NS teachers is of an imaginary basketball team that selects players solely according to their height in line with the popular belief that taller players have an advantage in the game. On this basis, a team might simplistically set, for example, a lower limit of 6 ft 9 (approx. 2.10 m)^[26] for any aspiring player, and adopt a policy of excluding, without even a trial, any applicant who is shorter, regardless of whether they may be faster, nimbler, better at passing, more accurate at shooting etc. than a taller applicant.

The analogy is imperfect however because, while height, like being NS, is not something that an individual can do anything to change once they have reached adulthood (at least not at the time of writing this), it is an objective criteria which can certainly be measured precisely.^[27] By contrast, the statuses of NS and NNS are more difficult to define in individual cases, e.g. where a person’s background and life story is out of the ordinary in some way, and harder still to prove (see Section 3). There is then often a degree of inherent unfairness when excluding an applicant for not being a NS, which would not be present if a basketball team were to exclude applicants purely because of their height because, just or not, the latter is a criterion that can at least in theory be applied to everyone in the same way.

However, it is too much to ask for the NS/NNS dichotomy to be abandoned completely in language teaching however much some may wish for such a thing (Section 3). This is because there does indeed exist a basis for a distinction between NSs and NNSs, at least as broad categories, in both language acquisition and in language competence (see Section 2). Nonetheless, as we have seen in this paper, just because relevant differences can be shown to exist between NSs and NSSs, this is not justification for the conclusion that one of these two categories is inherently better than the other, any more than it is to decide that a basketball player who is tall is necessarily better than one who is short.

Being a good language teacher requires a whole range of skills and qualities; no two good language teachers are good in exactly the same way. To prioritise only one of the many things which may make a language teacher competent is simplistic and misguided. Most of all, it does injustice not only to all language teachers (whether NS or NNS), but also to learners.