Are ‘native speakers’ better pronunciation models for our students?

This is an important question.

Not only because the answer will determine how we teach pronunciation, but also because it lies at the very core of the ‘native speaker’ fallacy, or the belief that a ‘native speaker’ is always a better teacher, which is so rampant in our profession.

Our gut feeling might suggest that yes, of course a ‘native speaker’ is the ideal pronunciation model (and by extension a better teacher). This is for example what one teacher said in a FB discussion on this topic:

When I learned German or French, I looked for native speakers, because a huge part of language learning is understanding the accent and intonation, and only a native speaker can provide that.

Discuss! 😉

This is just one example, but this comment is by no means an isolated one. I’ve seen countless similar ones over the years.

There is quite a persistent belief, not only among students, but also language teachers, that a ‘native speaker’ speaks correct, right, natural, original (pick your adjective) pronunciation, while a ‘non-native speaker’ has a bad, incorrect, foreign, intelligible, unintelligible (pick your adjective) pronunciation. Therefore, the former clearly makes a better pronunciation model and teacher.

However, the difference between teaching/learning English and other foreign languages, such as German or French, is fundamental. After all, English has gone global. Call it a lingua franca, an international or a global language, but the fact of the matter is that ‘non-native’ users of the language outnumber ‘native’ ones by probably 5:1.

This means that your average student is much more likely to interact with a variety of speakers from different countries for whom English is not their mother tongue, than with ‘native speakers’.

How then do we as teachers help our students be clearly intelligible in these lingua franca encounters? Which pronunciation model should we teach? That is, which pronunciation model will be the most widely intelligible?

For some of us, our gut feeling might still be telling us that a standard ‘native speaker’ pronunciation model is the best choice. That it is this model that our students should strive for to be more intelligible in international settings.

However, just how accurate is our gut feeling?

The other day, completely by chance, I stumbled across this article by Smith and Rafiqzad, published in TESOL Quarterly, and entitled English for Cross-Cultural Communication: The Question of Intelligibility. The article is interesting for three reasons:

a) it’s almost forty years old, but it seems to have gone pretty much unnoticed

b) it’s the only example I know of such a large-scale study into intelligibility in international contexts

c) it can shed some light on our gut feeling about pronunciation models.

In a nutshell, the authors surveyed 1386 people from 11 countries to check their ratings of intelligibility, which they defined as the “capacity for understanding a word or words when spoken/read in the context of a sentence being spoken/read at natural speed” (p.371). The listeners came from a variety of different disciplines (the authors don’t specify which), but all of them could be described as “educated by a majority of their countrymen” (p.372).

The recordings came from speakers from the US, Malaysia, Japan, Korea, India, Hong Kong, Nepal, The Philippines and Sri Lanka, all of whom spoke an educated variety of English from their country. The speakers were asked to prepare, read and record a short speech which would be appropriate for an educated, but not specialist, audience in their home country.

Intelligibility was rated with a closed test which consisted of the transcript of the recording with words removed from it. The listeners had to complete the gaps with no regard being paid to spelling.

Which speaker do you think came out as the most and the least intelligible?

Discuss! 😉

The researchers made two predictions. The first was that the ‘native speaker’ from the US would be the most easily intelligible across the board. Second, the familiarity with the accent would also increase the intelligibility. In other words, a Malaysian speaker would be more intelligible to a Malaysian listener than a to a Sri Lankan one.

Both hypotheses turned out to be false…

Let’s start with the second assumption. Only in two cases (Korea and Japan) did the listeners find their countrymen more intelligible than all the other speakers. This is surprising as you’d expect that the more familiar you are with the accent, the easier it would be to understand it.

Even more surprisingly, the US ‘native speaker’ (who spoke with a standard General American accent) was consistently among the least intelligible speakers. In fact, on average, the listeners were only able to complete the close test with an accuracy of 55%. The ratings from the highest to lowest are as follows:  Sri Lanka 79%, India 78%, Japan 75%, Malaysia 73%, Nepal 72%, Korea 68%, Philippines 61%, United States 55%, Hong Kong 44%.

Another surprising finding is that the listeners were also very poor at identifying the ‘native speaker’. In nine out of the eleven countries, less than 40% of the listeners identified the ‘native speaker’ correctly.

Side Note: I’m giving a FREE training entitled “How to teach pronunciation with confidence as a ‘non-native speaker'”, where I will share with you all my tips and tricks that will boost your confidence a s a’non-native speaker’ and allow you to successfully teach your next pronunciation class. Register below via email:

Or FB Messenger:


So what does this mean for us in ELT?

First, I think one very important implication is that we need to reconsider the idea that a ‘native speaker’ model of pronunciation is always by definition the most intelligible, and therefore one our students should aim for. While this research was conducted in Asia, it seems clear that standard ‘native-like’ pronunciation doesn’t guarantee intelligibility in international contexts. As the authors themselves put it, “[s]ince native speaker phonology doesn’t appear to be more intelligible than non-native phonology, there seems to be no reason to insist that the performance target in the English classroom be a native speaker” (p.380).

Mind you, I am not saying that any ‘non-native speaker’ is now by default a better model. However, what I am suggesting is that an INTELLIGIBLE speaker, regardless of their accent, place of birth or first language, is a better model.

It is a shame that the researchers did not attempt to analyse the recordings to identify which pronunciation features might have contributed to or reduced intelligibility. However, there is more recent research (Deterding, 2011; Deterding & Mohamad, 2016) conducted in a similar context, focusing on speakers from South East Asia, which seems to confirm Jenkins’ (2000, 2002) Lingua Franca Core proposal. Namely, it turns out that pronunciation features such as word stress, vowel quality, voiced and voiceless , weak forms and features of connected speech are not important for intelligibility. On the other hand, consonants, vowel length, nuclear stress and consonant clusters are crucial for intelligibility.

Second, we’re often told that students prefer ‘native speaker’ teachers. Researchers have also found that students tend to rate ‘native speaker’ speech more favourably (He & Miller, 2011; McKenzie, 2008; Scales, Wennerstrom, Richard, & Wu, 2006; Scheuer, 2008). Nevertheless, it seems that at least the participants in Smith and Rafiqzad (1979) study were not able to identify the ‘native speaker’ correctly. Similar observations were made by Pacek (2005), Scales et al. (2005) and McKenzie (2008). In fact, the latter highlights that only the speakers who WERE identified as ‘native speakers’ were rated more favourably.

As various authors note, it is very likely that students idealise ‘native speakers’ and their pronunciation. So when they say that they prefer ‘native speakers’ or ‘native-like’ pronunciation, it isn’t necessarily any real ‘native speaker’ or any real ‘native speaker’ pronunciation, but rather the imagined and idealised one.

It is very likely because we’ve (or at least certain very powerful institutions) worked very hard over the years in ELT to promote, maintain and spread native speakerism (Phillipson, 1992). We’ve also worked very hard at promoting the idea that ‘native speaker’ pronunciation is more intelligible, more correct, better (pick your adjective). We’ve also entrenched this belief through the use of standard ‘native speaker’ recordings in course books. I’m certainly guilty of the latter two.

Side Note: I’m giving a FREE training entitled “How to teach pronunciation with confidence as a ‘non-native speaker'”, where I will share with you all my tips and tricks that will boost your confidence a s a’non-native speaker’ and allow you to successfully teach your next pronunciation class. Register below via email:

Or FB Messenger:

So what do we do?

It seems to me that we have two options.

We can continue promoting the belief that ‘native speaker’ pronunciation should be the ultimate and only goal all ‘non-native speakers’ (teachers and students alike) should aspire to. It shouldn’t surprise us then, however, if the vast majority of students fail to achieve this goal. It also shouldn’t surprise us if they feel bad about their own pronunciation and about having a foreign accent. Nor should it surprise us if our students continue preferring ‘native speaker’ teachers.

Option 2: we can try to move beyond the ideology of native speakerism towards a more inclusive, international, lingua franca view of the English language which would place emphasis on research findings and on intelligibility in international contexts. This shift in perspective might allow us to better help our students to be more intelligible. It might also raise our students’ confidence when speaking in English by raising their awareness of the fact that they can achieve global intelligibility without having to worry about approximating ‘native-like’ pronunciation and without having to lose their accent. Finally, it might help us further chip away at the ‘native speaker’ fallacy that’s still so widely spread and deeply rooted in ELT.

Which one do you pick?

Discuss! 😉


Deterding, D. (2011). English Language Teaching and the Lingua Franca Core in East Asia.

Deterding, D., & Mohamad, N. R. (2016). The role of vowel quality in ELF misunderstandings. Journal of English as a Lingua Franca, 5.

He, D., & Miller, L. (2011). English teacher preference: the case of China’s non-English-major students. World Englishes, 30(3), 428-443.

Jenkins, J. (2000). The phonology of English as an international language : new models, new norms, new goals. Oxford: Oxford University Press.

Jenkins, J. (2002). A Sociolinguistically Based, Empirically Researched Pronunciation Syllabus for English as an International Language. Applied Linguistics, 23(1), 83-103.

McKenzie, R. M. (2008). The role of variety recognition in Japanese university students’ attitudes towards English speech varieties. Journal of Multilingual & Multicultural Development, 29(2), 139-153.

Pacek, D. (2005). “Personality Not Nationality”: Foreign Students’ Perceptions of a Non-Native Speaker Lecturer of English at a British University. In E. Llurda (Ed.), Non-Native Language Teachers (pp. 243-262). New York: Springer US.

Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford University Press.

Scales, J., Wennerstrom, A., Richard, D., & Wu, S. H. (2006). Language Learners’ Perceptions of Accent. TESOL Quarterly, 40(4), 715-738.

Scheuer, S. (2008). Why Native Speakers Are (Still) Relevant. In K. (ed. and foreword) Dziubalska-Kołaczyk & J. (ed. and foreword) Przedlacka (Eds.), English Pronunciation Models: A Changing Scene (Vols. 1-476 pp., pp. 111-130). Bern, Switzerland: Peter Lang.

Smith, L. E., & Rafiqzad, K. (1979). English for Cross-Cultural Communication: The Question of Intelligibility. TESOL Quarterly, 13(3), 371–380.

Side Note: I’m giving a FREE training entitled “How to teach pronunciation with confidence as a ‘non-native speaker'”, where I will share with you all my tips and tricks that will boost your confidence a s a’non-native speaker’ and allow you to successfully teach your next pronunciation class. Register below via email:

Or FB Messenger:

17 thoughts on “Are ‘native speakers’ better pronunciation models for our students?”

  1. Great article! Very informative and interesting. I couldn’t agree more, as long as the language produced is intellegible… There will be communication! “Neutral English” is good enough. In my view we should also encourage our students to be ambitious to try and stick to the standard pronunciation.

  2. You put forward a good case here, and the results of the 1979 study are fascinating – it would be good to see more studies like that especially in dialogue with the Lingua Franca Core proposals. But as you did repeatedly request discussion, I think there are a few aspects you missed. Firstly, the question of prestige accents, which in certain professions could be as much of an issue for native speakers as non native ones. Secondly, there is the consideration of market forces. The native speaker model ‘brand’ would appear to be popular in all the countries I’ve taught in (UK, Ukraine, Germany, Kazakhstan). This would appear to be unconnected to any considerations of intelligibility. On the contrary, the same students who request a native speaker model may also be frustrated by how native speakers seem to ‘swallow’ the sounds. The third question is the need to teach the vagaries of various English accents, in order to promote aural decoding of them, depending on which speech community the learners wish to become a part of. If the learners are aiming to interact with the “inner core” countries, it might well mean the learners need to be taught at least to understand one or more native accent.

    1. Hi James! Thanks for responding 🙂
      I agree that it would be great to see more similar studies. I’m actually amazed that this one hasn’t been replicated. I might be wrong, but I’m not aware of any similar ones. It would also be good to see similar studies also aiming to establish not just which pronunciation resulted to be more intelligible, but also why.
      I certainly agree that the issue of power and prestige associated with certain accents is very important. However, I do hope we’re past Prof Higgins and Eliza Doolittle’s times, when you were forced to change your accent if you wanted to advance in society 😉
      The market demand is an important question. However, since we’re in the business of education, I think it’s very important we talk to students about this issue. As I mentioned in the article, many students idealise ‘native speaker’ pronunciation. They’re likely to prefer a ‘native speaker’ teacher because they think they will learn better pronunciation, that somehow by being in the presence of a ‘native speaker’ for three hours a week in a general English class, they’ll magically acquire the accent. However, we all know that this is virtually impossible. Pronunciation isn’t like the cold – you don’t catch it. So while I understand that a market demand exists, I think it’s important we educate our learners. Otherwise, they’ll never be informed customers.
      With regards to your third point, I completely agree. For receptive purposes, it’s of course important the learners are given practice and help listening to different ‘native speaker’ models. This, however, shouldn’t mean that they only listen to ‘native speakers’. Even in English-speaking countries, our students are very likely to encounter a huge variety of ‘non-native’ accents.

  3. Scott Thornbury’s wholly sensible conclusion to questions of global intelligibility:

    Successful communication, after all, is contingent on a great many more factors than simply pronunciation – not least the need and willingness of the communicating parties to communicate! In the end, it is not accents that are intelligible, but people – using all their available linguistic, paralinguistic, and metalinguistic skills.

  4. And I’m guessing you, like me, also listen to your interlocutor in any situation and respond respectively. It’s common sense and sensitivity to context that makes for successful and effective communication.

    Also, a question for Marek: Do you regret achieving a native-like accent? Has your native-like accent caused intelligibility problems for you in global communications?

  5. My main questions remain unanswered:

    How far can studying and using ELF models get you?
    How far can studying and using ENL models get you?

    I’m talking about professionally and privately. I’m thinking about work, leisure, travel, reading, watching TV and films, building relationships. How far can each of the currently available models, the Englishes if the world, get you, can benefit you?

    I’d like to hear from ELFers and ENLers in that.

    1. I think the study I reported on in this post provides us with some clues, at least as far as GAm pronunciation is concerned. In general, I’d say that our gut feeling about which pronunciation model might be the most intelligible in lingua franca context might be unreliable, so I would look to research for answers.
      There is no ELF model of pronunciation in the sense that GAm or RP is a model. The Lingua Franca Core (LFC) is a list of pronunciation features which contribute to or do not contribute to intelligibility in international contexts. There is quite a lot of research to back it up. So I would say that it serves as a very good indication of which pronunciation features teachers should focus on in class more to help students become more intelligible.

  6. Lastly, if some students pick up on this ELF approach as a way to avoid learning lots of grammar and lexis and pronunciation items, as a way to make excuses when studying a language becomes hard, when it demands effort and commitment, how can we language providers react. How can we stop ELF becoming English for lazy fuc*ers?

    Please advise.

    1. I don’t think all students should be obliged to learn all the features of a standard ‘native speaker’ model. If they don’t want to sound like a ‘native speaker’, then that’s fine by me. We will focus on the pronunciation features that will help them achieve global intelligibility, but they can keep their L1 accent. If they don’t want to, can’t be bothered, fail to (insert your reason here) to learn the third person ‘s’, it’s fine by me. There are probably much more important things to worry about than the third person ‘s’.
      Effort and commitment don’t have to be connected to the obligation to achieve a complete mastery of the ‘native speaker’ model. Think how demotivating and discouraging for some students it might be to be constantly pushed to achieve a ‘native speaker’ standard, and at the same constantly fall short of it.
      And students should be given choices. I think that’s what ELF provides. A choice. A choice that’s based on research on how English is currently being used, rather than a choice based on the ideology of native speakerism.

  7. And native speaker classrooms also give a choice. You can go for C2, for example, or you can just stop learning English in a formal way. The world of EFL/ESL is not as coercive or conditioned as you think.
    I feel one can be accused of being patronising to students if one assumes the choice they’ve made to study native models is no choice all. You make them seem like automatons.

  8. As a native speaker who is also a EFL teacher, I am fascinated by topics such as these. It seems as though connected speech is one of the main differences between native and non-native English speakers, which can be heard best in songs that are sung by English musicians. Connected speech has nothing to do with intelligence, but could explain why non-native English speakers would have a hard time understanding someone with an authentic American accent. Do you think there is something to be said about learning English from a native speaker, though? When I was learning French, my teacher was American. It was so easy to revert to English, and I wish I had native French speaker who knew no English. I think about how much more of a creative process the learning would have been, and how greater my capacity for French would be:)

    1. Hi Jenny! Thanks for your comment.
      I think there is certainly something to be said about learning from someone who doesn’t speak the students’ mother tongue. However, this doesn’t mean that this person ahs to be a ‘native speaker’ teacher. For example, I teach English in Belgium, and I don’t speak Flemish, which is the first language of my students. Also, you could have the opposite situation. In other words, a ‘native speaker’ of English who speaks the students’ first language (there are plenty of those).
      On the other hand, I think there is something to be said about having classes with a teacher who does speak your language. Using L1 in class, if done judiciously, can facilitate learning I think.
      What do you think?

  9. I and my classmates once talked about this topic in our classroom. It is true that native English speakers can pronounce words much better than non-native Englsih speakers. But we should also notice that Englsih-native speakers from varied districts also have different accents. I mean that people from Chicago may speak English differently from people native to Boston. Before we, English leaners, begin to learn so-called Standard English, we need to make sure whether there is any Standard English or not. When we struggle to reduce our L2 accent, we also need to think whether our accent really effect our communication with others.

  10. Pingback: Profesores nativos: mitos vs realidad – Centro Norteamericano

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.