A Reply to A. Holliday’s “Why we should stop using native-non-native speaker labels”


1  In the domain of English language teaching, there is just about universal agreement that discrimination against non-native speaker teachers must stop. Those who fight to end such discrimination have my full support.

2  In the domain of SLA research, native speakers of language X are people for whom language X is the language they learnt through primary socialization in early childhood, as a first language.

3  To paraphrase Long (2007, 2015), the psychological reality of native speakerness is easily demonstrated by the fact that we know one, and who isn’t one, when we meet them, often on the basis of just a few utterances. When monolingual speakers are presented with recorded stretches of speech by a large pool of NSs and NNSs and asked to say which are which, the judges are always very good at distinguishing them, with inter-rater reliability typically above .9. How do they do this, and why is there so much agreement if there is no such thing as a NS?

4  For the last 60 years, the term “native speaker” has been used in the literature concerning studies of language learning, and one of the most studied phenomenon of all is the failure of the vast majority of post adolescent L2 learners to achieve what Birdsong (2009) refers to as “native like attainment”.

On the prevailing view of ultimate attainment in second language acquisition, native competence cannot be achieved by post pubertal learners. There are few exceptions to this generalization (Birdsong 1992).

5  Claims concerning the relative abilities of native speakers and learners of the target language are not disconfirmed by individual cases. The claims all accept the psychological reality of native speakerness.

6  The specific claim that very few post adolescent L2 learners attain native like proficiency is supported by a great deal of empirical evidence (see, e.g., reviews by Long 2007, Harley and Wang 1995; Hyltenstam and Abrahamsson 2003).

7  When trying to explain why most L2 learners don’t attain native competence, scholars have investigated various “sensitive periods”. It’s widely accepted that there are multiple sensitive periods for different domains of second language learning  – pronunciation, morphology and syntax, lexis and collocation (see Long, 2007, Problems in SLA, Chapter 3 for a review of sensitive periods).

To the issue then

Adrian Holliday, Professor of Applied Linguistics & Intercultural Education at Canterbury Christ Church University, has just published a post on his blog: Why we should stop using native-non-native speaker labels  in response to queries about his claim that the terms native speaker and non-native speaker are neo-racist. He addresses the questions: “What does ‘neo-racist’ mean?” and “Are there no occasion (sic) when these labels can be used?”.

He starts with his own subjective impressions of what ‘native speaker’ means to him and then says

In academia the established use of ‘native speaker’ as a sociolinguistic category comes from particular paradigmatic discourses of science and is not fixed beyond critical scrutiny.

I’ve no idea what the phrase “particular paradigmatic discourses of science” refers to, but I’m sure we can all agree that the use of ‘native speaker’ as a sociological category is not fixed beyond critical scrutiny. Holliday seems to be saying that quantitative research based on testing hypotheses with empirical evidence, as carried out by many scholars trying to understand the  psychological process of SLA is part of a “mistaken paradigm”. Since in SLA research there isn’t, and never has been, any general theory of SLA with paradigm status, and since I’m sure that in the field of sociolinguistics and cultural education they’re even further away from any such theory, talk of paradigms, like talk of “imagined objective ‘science’”, and problems that reside in differences being evoked “regardless of the words that are being used”, and labels referring to things that “do not actually exist at all”, belongs to the giddy world of post modern sociology where words mean what their authors choose them to mean “neither more nor less”, as Humpty Dumpty triumphantly concludes.

Whatever the term ‘native speaker’ might be used for in sociolinguistics, in psycholinguistics ‘native speaker’ refers to real people, as I’ve explained above, and nothing that Holliday says challenges this fact. So we’re left with the charge that when we refer to people as ‘non-native speakers’, we imply that they are “culturally deficient”, which amounts to “deep and unrecognised racism”.  We “define, confine and reduce” this group of people and refer to their culture in a way that evokes “images of deficiency or superiority – divisive associations with competence, knowledge and race – who can, who can’t, and what sort of people they are”.

In my opinion this is so badly written as to be almost incoherent, but perhaps it expresses exactly what Holliday means to say. Whatever it means, it’s difficult to counter something like neo-racism if it’s “unrecognised”, and if any attempt we make to use other terms just pushes the labelling “even further into a normalised, reified discourse, where we are even less likely to reflect on their meaning, and where a technicalisation of the labels somehow makes them more legitimate”. Still, since Holliday confidently asserts that “the native-non-native speaker labels” refer to something “that does not actually exist”, it should be easy enough for sociolinguists (and those involved in intercultural education too, I suppose) to stop using them. Meanwhile, back in the real world,  it’s a different story.

Long (2007) argues that the issue of age differences is fundamental for SLA theory construction. If the evidence from sensitive periods shows that adults are inferior learners because they are qualitatively different from children, then this could provide an explanation for the failure of the vast majority of post adolescent L2 learners to achieve Birdsong’s “native like attainment”. If we want to propose the same theory for child and adult language acquisition, then we’ll have to account for the differences in outcome some other way; for example, by claiming that the same knowledge and abilities produce inferior results due to different initial states in L1 acquisition and L2 acquisition. Either way, the importance of the existence (or not) of sensitive periods for those scholars trying to explain the psychological process of SLA indicates that native speakerness will continue to be used as a measure of the proficiency of adult L2 learners.


Harley, B. & Wang, W. (1997). “The critical period hypothesis: Where are we now?”. In A. M. B. de Groot & J. F. Kroll (Eds.), Tutorials in Bilingualism: Psycholinguistic perspectives (pp. 19–51). Mahwah, NJ: Lawrence Erlbaum Associates.

Hyltenstam, K. & Abrahamsson, N. (2003). “Maturational constraints in SLA”. In C.J. Doughty & M.H. Long (Eds.), Handbook of Second Language Acquisition. Oxford: Blackwell.

Long, M. (2007) Problems in SLA. London, Erlbaum.

Long, M. (2015) SLA and Task-based Language Teaching. Oxford, Wiley.


7 thoughts on “A Reply to A. Holliday’s “Why we should stop using native-non-native speaker labels”

  1. hi Geoff

    psycholinguistics, or more generally the cognitive tradition, in using a measure (in this case level of proficiency) is mostly in contexts of the relative performance of various levels; i think this issue is missed in a lot of the talk of whether to use or not use “native speaker” in academic talk i.e. a misconstruing of relative performance for absolute performance

    in your example of how people know very quickly whether a speaker is a “native” speaker or not this depends on the perceivers previous experience & consequent schema of what is a “native” speaker; i.e. someone who has never heard French will find it difficult to judge the “nativeness” of a speaker (all other things equal). so did the studies quoted by Long control for previous experience of a language?


    Liked by 1 person

    • I’m sure there’s a Derwing and Munro paper that claims that people can identify native speakers and non-native speakers by their pronunciation even in languages that the listeners don’t speak.


      • Thanks for this. Probably Munro, M. J., Derwing, T. M. & Burgess, C. (2010). Detection of nonnative speaker status from content-masked speech. Speech Communication, 52 (7-8), 626-637. I haven’t read it, but I will.


    • Thanks for the link, Julia. I liked (and ‘Liked’!) the post. Good to see CELFS TEACHING AND LEARNING NETWORK flying the Bristol Uni flag. I have some very fond memories of visiting Bristol uni.


  2. ELT bites (http://www.eltresearchbites.com/201702-2383/) reported on a recent Applied Linguistics forum post by Jean-Marc Dewaele doi:10.1093/applin/amw055 where he suggests a) replacing NS by L1 user and NNS by L2 user. This move from learner/bilingual to L2 user follows Cook, V. J. and D. Singleton. 2014. Key Topics in Second Language Acquisition. Multilingual Matters. Dewaele then suggests b) replacing L2 user by Lx user, since L2 is currently a cover term for L3, L4 etc.

    I think L1 user and L2 user are good terms, since as you point out they are meaningful in second language research, and they are more neutral and accurate that the terms native and non-native which imply monolingual competences which are not typical all over the world. I’m not so sure about Lx, since the x would have to mean “anything more than 1,” but I appreciate the intent.

    And as you say at the start of your post, Geoff, there is a difference between talking about speakers’ competence and about language teaching. Perhaps the term “user” is helpful here as a more neutral alternative.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s