The Native Speaker Standard

I wrote the following for a class on Second Language Acquisition. These are some of my thoughts and research on the use of the native speaker as the final state of second language acquisition. Apologies in advanced for the unpolished nature of the paper. I had a lot going on when I wrote it.


Native Speaker as a Final State of SLA

“Large numbers of people will learn English as a foreign language in the 21st century and they will need teachers, dictionaries and grammar books. But will they continue to look towards the native speaker for authoritative norms of usage?” (Graddol, 1999, p. 68)

This paper investigates the concept of “native speaker” as a final state of second language acquisition. It defines the terms “native speaker” and “final state” and suggests that the native speaker standard of second language acquisition (SLA) is problematic. The paper hopes to show that alternatives to these concepts are needed and to point to possible directions to seek those alternatives.

“Final state” refers to Chomsky’s levels of acquisition: initial state, interlanguage, and final state (Saville-Troike, 2012, p. 52). In SLA, the initial state is the first language (L1), the intermediate state of “interlanguage” is the dynamic grammars of the learner during the process of acquisition, and the final state is the full adult grammar of the language (Saville-Troike, 2012, pp. 52-55). The term “final state” assumes that there is a definite line between acquiring and knowing in both L1 and L2. One might reasonably ask where this line rests, but such a question is outside of the bounds of this paper.

It is notoriously difficult to define “native speaker” in applied linguistics. However, it is also clear that the “native speaker” has been used as a benchmark, a norm, for students attempting to acquire a second language (Saville-Troike, 2012, p. 188). Most textbooks feature pictures of monolingual native speakers and the listening portions of course books also include conversations between native speakers using “standard” accents, to a point where the nonnative speaker students feel “othered” by their class materials (Hatoss and Saito, 2011, p. 119). But what is this elusive native speaker? What does the native speaker have that the L2 user needs to acquire? Can an L2 user ever become a native speaker? And if not, why is the native speaker standard used as a final state for acquisition? What are the alternatives? Are we even asking the right questions?

The concept of “native speaker” is a tricky one. Alan Davies (2003) attempts to describe the native speaker in his book, The Native Speaker: Myth and Reality. After over 200 pages of exploration, he finally concludes that it is impossible to arrive at a good definition. According to Davies (2003, p. 208), “to be a native speaker means not being a nonnative speaker.” Davies follows this definition with a statement that the distinction between native speaker and nonnative speaker is one of confidence and identity (p. 213). Meanwhile, Vivian Cook takes a much more strict interpretation of “native speaker”: Cook’s native speaker is “a monolingual person who still speaks the language learnt in childhood” (Cook, 1999, p. 187). Cook contrasts this with an L2 user = someone who uses an L2. In Cook’s view, a nonnative speaker (L2 user) can never become a native speaker. Claire Kramsch (1997, p. 363) puts forth a third view which suggests that “native speakers are made rather than born.”

Chomsky holds an honoured place in the history of defining a native speaker: according to him, a native speaker is an idealised fictional person who knows his language perfectly (Kachru, 1994, p. 796). It is this view that has been most dominant in SLA studies and that has played a powerful role in setting the standards for L2 acquisition. Following Chomsky, Dobson (2001, p. 65) identifies a native speaker as one who “is equipped with ultimately correct linguistic knowledge and communicative competence, which are normally devoid of diversity.” The problem, according to Kachru, is that “[r]esearchers compare the performance of any native speaker(s), or judgment of the investigator(s), with learner performance, and generalizations made from such comparisons about learner competence are treated as legitimate evidence for SLA hypotheses” (1994, p. 796). This leads not to a standard so much as a chaotic mass of subjective judgements. In agreement with Kachru and in light of the absence of an agreed-upon definition of “native speaker”, how can such a concept be used as a model of the final state of SLA?

It seems to be a common assumption that the goal of second language acquisition is to get as close as possible to a native speaker (Saville-Troike, 2012, pp. 188-189). While most English-language research about SLA regards English acquisition, the assumption is also prevalent in acquisition of other languages, including French and Spanish. Drewelow and Theobold (2007) did a study which showed that native French speakers care far less about the degree of perfection of nonnative French speakers than do the nonnative teachers of French (p. 503). Fioramonte and Thompson (2013) found that some nonnative teachers of Spanish believed phonological factors in their acquisition were directly linked to respect they’d receive from their students (p. 573). Kramsch (1997, p. 362) summed up the inertia evident in SLA research neatly, pointing to the prevailing attitude of “Where would teachers and learners take their models from if there was no such thing as a native speaker?” Hence the native speaker standard has remained persistent and problematic.

The L2 user’s attitude is one of the many reasons for the persistence of the native speaker model. According to Davies (2003, p. 197), “The native speaker is a fine myth: we need it as a model, a goal, almost an inspiration.” Indeed L2 users seem to gravitate towards this goal. Anne Pakir pinpoints the attitude involved when she writes, “standard language is an idea in the mind rather than a reality” (1997, p. 172). L2 users are aiming for an unattainable fiction because, according to Cook (1999), that is what they have been told is the goal. Cook claims that the attitude of the students is a product of the system (p. 196) and feels strongly that “this acceptance of the native speaker model does not mean these attitudes are right.” Saito and Hatoss (2011, pp. 108-109) seem to support this. They found that Japanese EFL students “strive to ‘perfect’ their English to be like that of the native speaker” and at the same time develop negative perspectives on non-native varieties of English (including their own ‘failure’). Their perception that English is the property of native speakers (p. 110) may have enhanced their feeling of failure in language acquisition. The students learned through the observation that teaching materials all focus on the native speaker standard (p. 119).

That said, it seems in some places the attitudes of L2 users are changing. Braj Kachru (1996, p. 135) argues for Englishes to replace a single standard English because “Englishes” symbolises the distinct identities of users of English. It stresses “we-ness” rather than native vs. non-native dichotomy. Mesthrie (2010, p. 600) adds, “I place greater weight on evidence from language shift that shows that features from a variety that was once a non-native English can easily stabilise and pass as those of a native variety. Furthermore, pragmatic issues of fluency and confidence of control are also important.” The native varieties he refers to include Indian English and Singaporean English, although others could be added. Confidence of control cannot be over-stressed. A confident language user would laugh at anyone who suggested they were speaking in a non-standard way. The idea of confidence recalls Davies’ point that it is confidence and identity that distinguish a native speaker.

Textbooks, it seems, are having trouble keeping up with the changing needs of L2 users. According to Kachru (1996, p. 139), “[i]nteractions involving English in non-Western countries are mostly carried on by nonnative users with other nonnative users not, as one would suppose, by nonnative users with native users.” International languages and lingua francas are not new concepts , but have become a topic of study in SLA recently since English has taken on that cloak (Sridhar, 1994, p. 800). Kachru claims that native speaker input has become unnecessary (1996, p. 149). Perhaps someday textbooks will be written by and for L2 users. Already classroom teachers are trying out alternatives, including Virtual Exchange Projects in which monolingual EFL classrooms connect to similar classrooms in other countries by means of available technology with the goal of introducing their multilingual students to other multilingual students in order to promote communication.

One reason why the native speaker standard is inappropriate for L2 users is that the “native speaker” is often seen as being monolingual, while an L2 user is by definition multilingual. According to Kramsch and Whiteside (2007, p. 911), “There are also increasing doubts as to whether any monolingual speaker can be upheld as the norm for L2 learners who are, by definition, aspiring bilinguals.” There is always an internal dialogue between L1 and L2. This dialogue is what enables other versions of the standard language. It is the attitude of the users towards the language that determines the prestige of the language they use. If they see themselves as failed native speakers, their motivation will be aversely affected. If they see their language skills in terms of their accomplishments, their relationship with the language will be positive.

Often theories of SLA forget to focus on the learner as a human. Deters and Swain (2007) found that the language learners’ goals and needs have an impact on their level of acquisition. Learners are influenced by their attitude towards the target language and towards the target community. In their study, one learner who interacted primarily with native speakers based on the belief that it’s the way to learn the language achieved a lower proficiency than another learner who interacted primarily with other L2 users (p. 823). Deters and Swain conclude that learners construct their own identities and have their own goals, and these should not be discounted by SLA researchers (p. 831). These goals and needs often do not lead to a native-like competence. A case in point is the case study done by Ulla Connor (1999). Her subjects were more interested in lexis than syntax and tended to vary their L2 use depending on their relative relationships to their interlocutors. Connor found that “native and non- native speakers of English are more concerned about mutual intelligibility than about keeping to some ‘standard’ form of English such as standard American or received British English” (p. 128). The goal for these L2 users was communication with other L2 users, in whatever way worked, which included being sensitive to the perceived L2 level (compared to themselves) of their interlocutors.

Sridhar (1994) recommends a “reality check” for SLA research. He points out that more L2 users need the L2 to complete tasks than to integrate into a native speaker community (p. 800). According to Sridhar, more L2 learning occurs in foreign language environments than in second language environments. The needs of such learners differ, as do their goals. Sridhar therefore argues that the native speaker model is inappropriate for two reasons: “native speaker norms are a distraction when the primary interlocutors are nonnative or nonstandard speakers … and the Chomskyan notion, because it idealizes away variation, performance, and especially bilingualism, is even less suitable to SLA than it is to linguistics” (Sridhar, 1994, p. 801). Kramsch (1997, p. 367) agrees that current SLA research fails to “exploit the linguistic diversity that students bring to language learning.” Indeed when L2 education is focused on reaching unreasonable standards, creativity is often discarded.

That the native speaker model is unfair to learners is evident from Cook’s scathing article, “Going beyond the native speaker in language teaching” (1999). Basing his argument on the premise that a monolingual native speaker is not comparable to a multilingual L2 user, Cook presented a different type of language user: a “multicompetent” one. Cook’s term “multicompetence” refers to the “compound state of a mind with two languages” (p. 190). The term is meant to be free from exogenous judgement. Cook argues that multicompetence is intrinsically more complex than monolingualism, involving code-switching and even changes in cognition (p. 191). Knowing an L2 changes their relationship with their L1 in a way that has no equivalent in monolingualism (p. 194). An L2 user is so much more than the failed native speaker presented by much SLA research. Even a person who learns an L2 for just a year before abandoning has acquired something of use, rather than failed to acquire a native speaker’s language ability (p. 204). Unfortunately the students (and their nonnative speaking teachers) don’t always agree, as Saito and Hatoss (2011) and Drewelow and Theobald (2007) show. Cook concedes in his response to comments on his article (2000, p. 331), that the “goal of L2 use has to be adapted to the wishes and aspirations of our actual students now, even if we as teachers hope that future students will be less rigid in their views of what being a successful L2 learner actually means.”

For many reasons identified above, the native speaker model is both persistent and problematic. It would be ideal to move away from such a model, and Cook (1999, pp. 197-203) offers some suggestions for first steps: setting L2-user appropriate goals, including in-classroom teaching in which the students decide what they want to learn; using materials that include L2 user role plays and roles which present situations in which L2 users actually take part; using methods of teaching that recognise and validate the students’ L1; and basing teaching on L2 users’ descriptions. Cook argues that students should encounter skilled L2 use, rather than just native speaker use (p. 198). Perhaps even experience with unskilled L2 use would benefit L2 users, since their communication will likely be with real, not hypothetical, people.

On a political level, to effectively remove the native speaker standard, Pakir (1997, p. 176) calls for attitudes to locally defined standards to be accepted and supported by professionals, local institutions, and the general population in order to be successful, reinforcing the point that L2 users are their own worst enemy in terms of acceptable standards. How can a standard be accepted from without if it is not yet accepted from within?

Thompson and Fioramonte (2013, p. 577) suggest that an appropriate goal for L2 users is “to be competent users of the language, rather than try to achieve the unattainable “native speaker” status; NNSTs are excellent role models for this process.” Han (2004, p. 184) suggests that the idealised native speaker can serve as the goal, while an idealised L2 user serve as the measure of success. This measure is less than ideal, since an ideal is unattainable by definition.

According to Jenkins (2000, p. 96), a standard of some sort is necessary. Without a standard, there is danger of unintelligibility far greater than within SL or FL interaction. “Participants in EIL need to be able to tune into each other’s accents and adjust both their phonological output and their receptive expectations accordingly.” Davies (1989, p. 457) agrees, saying “The keystone is intelligibility; without that there is no speech community, no standard.” This is an important point: without a native speaker model, other ways of ensuring intelligibility must be found. One option is, as Jenkins proposes, a standard L2 user pronunciation. I argue that a grammar as well as pronunciation might benefit from being standardised in this case. Dobson (2001, p. 69) points out the relationship between language and culture and promotes a dialogue between the L1 and L2 cultures, rather than a “teacher lead, learner follow” attitude. Perhaps a similar dialogue would be useful between languages.

An appropriate measure of L2 proficiency might be as impossible to define as the term “native speaker.” It might also be unnecessary. It might be sufficient to describe an L2 user’s proficiency in terms of whether he or she is able to achieve his or her goals in using the L2. It is widely believed that the goal of L2 learning and all language use is communication with others. Perhaps a proficiency can be measured in terms of how well a language user (L1 or L2) is able to accommodate to their interlocutors. To that end, part of a language teacher’s job may be raising awareness in his or her students of other varieties of the target language and how other users communicate. This can involve explicit teaching of lingua franca communication (Holliday, 2005).

On a less theoretical level, it can also involve practice – in multilingual classes the setting is ripe for using the common language to understand one another without judgements on “correctness.” Monolingual classes in the foreign language setting can make use of available technology to connect to classes based in other countries. Taking evaluation out of the equation may help change the focus from determining the final state of SLA to describing how and why L2 users interact with other L2 users successfully. In the final analysis, Alan Davies said it best: “language is always subservient to our communication needs” (1989, p. 465).


Connor, U. (1999). How like you our fish? Accommodation in international business. In M. Hewings and C. Nickerson (Eds.), Business English: Research into practice (pp. 115-128). Harlow: Longman.

Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33(2), 185-209. Retrieved from

Cook, V. (2000). Comments on Vivian Cook’s “Going beyond the Native Speaker in Language Teaching”. The author responds. TESOL Quarterly, 34(2), 329-332. Retrieved from

Davies, A. (1989). Is English an international language? TESOL Quarterly, 23(3), 447-467. Retrieved from

Davies, A. (2003). The native speaker: Myth and reality. Toronto, Canada: Multilingual Matters, LTD.

Deters, P. & Swain, M. (2007). “New” mainstream SLA theory: Expanded and enriched. The Modern Language Journal 91(1), 820-836. Retrieved from

Dobson, A. (2001). Teaching of culture within foreign language teaching and its relationship to nationalism. Australian Review of Applied Linguistics, 25(1), 63-73.

Drewelow, I. & Theobald, A. (2007). A comparison of the attitudes of learners, instructors, and French native speakers about the pronunciation of French: An exploratory study. Foreign Language Annals, 40(3), 491-520. Proquest.

Fioramonte, A. & Thompson, A. (2013). Nonnative speaker teachers of Spanish: Insights from novice teachers. Foreign Language Annals, 45(4), 564-579. doi:10.1111/j.1944-9720.2013.01210.x

Graddol, D. (1999). The decline of the native speaker. In D. Graddol and U. Meinhof (Eds.), English in a Changing World. AILA Review 13(1), 57-68.

Han, Z. (2004). ‘To be a native speaker means not to be a nonnative speaker’. Second Language Research, 20(2), 166-187. doi:10.1191=0267658304sr235oa

Hatoss, A. & Saito, A. (2011). Does English ownership rest with us? Global English and the native speaker ideal among Japanese high school students. International Journal of Pedagogies and Learning, 6(2), 108-125.

Holliday, A. (2005). The struggle to teach English as an international language. Oxford: Oxford University Press.

Jenkins, J. (2000). Intelligibility in interlanguage talk. In The phonology of English as an international language: New models, new norms, new goals. Oxford: Oxford University Press.

Kachru, B. B. (1996). World Englishes: Agony and ecstasy. Journal of Aesthetic Education, 30(2), 135-155. Retrieved from

Kachru, Y. (1994). Monolingual bias in SLA research. TESOL Quarterly, 28(2), 795-800. Retrieved from

Kramsch, C. (1997). The privilege of the nonnative speaker. PMLA. Publications of the Modern Language Association of America 112(3), 359-369. Proquest.

Kramsch, C. & Whiteside, A. (2007). Three fundamental concepts in second language acquisition and their relevance in multilingual contexts. The Modern Language Journal, 91(1), 907-922. Retrieved from

Mesthrie, R. (2010). New Englishes and the native speaker debate. Language Sciences, 32(1), 594-601. doi:10.1016/j.langsci.2010.08.002

Pakir, A. (1997). Standards and codification for world Englishes. In L.E. Smith and M.L. Forman (Eds.), World Englishes 2000 (pp. 169-181). Honolulu, Hawai’i: University of Hawai’i Press.

Sridhar, S.N. (1994). A reality check for SLA theories. TESOL Quarterly, 28(4), 800-805. Retrieved from

Virtual Exchange Projects. (2013). Discussing Virtual Exchange Projects in ELT. Retrieved from

Leave a Reply

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

You are commenting using your 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

%d bloggers like this: