Standard International English? Part 2 ELF

Standard International English? Part 2 ELF

by Charles Lowe

Part 2 - English as a Lingua Franca


So far in Part 1, I have argued that Kachru’s concentric-circles model is inappropriate to the discussion of pedagogy, and I suggest a replacement. In Part 2, I move on to ELF. The research base for ELF has provided valuable insights into the non-native-speaker use of English.

The Cons and Pros of ELF

Having compared Emmerson and Modiano to Kachru, I would now like to explore how Jenkins (2007) justifies the centrality of ELF in language teaching. According to Jenkins, 1 to 1.5 billion non-native speakers of English (Kachru’s Expanding Circle) use English in their lives, and 80% of all interaction in English is between non-native speakers. I have no dispute with these figures.

The contention of Jenkins (2002), and Seidlhofer (2001), is that, because there are identifiable linguistic phenomena in the corpus data of ELF, therefore these phenomena should constitute an aspect of the syllabuses of English language students. Their contention is that ‘interaction among non-native speakers from different first languages’ (Jenkins 2003:129) should provide the features of the emerging lexico-grammar of ELF. The VOICE data (Vienna-Oxford International Corpus of English) has thrown up some interesting features that do not interfere with L2-L2 communication, such as treating ‘who’ and ‘which’ as interchangeable relative pronouns.

However, there is a wide logical gap between discovering interesting features of pragmatic competence in L2-L2 communication, and proposing these features, in principle, as influencing the components of an English language teaching syllabus. Indeed, the rationale for the proponents’ degree of certainty is that, if 80% of the world ‘gets by’ in English, cobbling together a combination of ‘near-enough’ pronunciation (my coinage), ‘near-enough’ grammar, and ‘near-enough’ lexis, along with an appropriate amount of negotiation of meaning, real objects, visual gesture and signage, then why teach more than is necessary for these interlocutors to ‘get by’? The argument seems to have become: why teach accurate language at all, since so much of non-native to non-native interaction (i.e. ELF), is ‘inaccurate’, in terms of a standard form of the language? Hammond (2007), with appropriate reservations about ELF, suggests a weaker model in which language learners are taught ELF as one of several registers of English, along with formal and informal speech and writing, simply because it would be useful to many of those heading to British universities where non-native students often constitute more than 50% of the total.

I believe Jenkins’ position is untenable. My objections are not, primarily, made on theoretical grounds, but on practical grounds, given my experience of talking to learners in many teaching contexts. Indeed, it is on the practical level that ELF is being touted as an alternative syllabus paradigm, so an initial practical critique is appropriate.

The practical objections to Jenkins’ ELF assumptions are: firstly, the mere fact of the number of non-native users of English over the relatively smaller number of native-speaker users does not render her model more ‘right’; secondly, it is simply true that most students of English want a native-speaker teacher where possible; and finally, non-standard language use is not considered well by gate-keepers, such as university-entrance-administrators and employers.

But more importantly, there is a clear theoretical dimension. The main theoretical objection is about a key flaw in the Jenkins argument, which is the apparent refusal to project any descriptive model of exactly what ELF is. It is true that a number of language phenomena have been identified from the corpora. These are interesting and the conclusions are fair. For instance, when a non-native speaker says ‘A girl which has red hair’, mistaking ‘which’ for ‘who’, it is fair to conclude that this hardly interferes with communication. It is also true (Jenkins 2003) that generalisations can be made about the way non-native interlocutors pragmatically ‘get their meaning across’, and that, according to the data so far collected, problems impeding successful communication are rare. However, the researchers, Seidlehofer et al, have only narrowed down a few areas of error, from an as-yet limited corpus, so we are left with two serious shortfalls in the proponents’ arguments. Not only are the specifics of the potential errors completely unpredictable (i.e. no-one can know what non-standard forms non-native speakers are going to produce in the future, and therefore cannot predict the extent to which communication will or will not be impeded), but we must also theoretically multiply the few identified aberrations by ‘N’, making the number of potential errors vast. Given the joint flaws of unpredictability and range in this non-standard but apparently communicatively ‘unproblematic’ body of not-quite-English language, just what would the proponents of ELF have us teach our students?

Having presented the cons, it is important to mention the pros, and to say that, clearly, there is a lot of support in the literature for Jenkins’ basic idea, and arguments-in-favour generally revolve around the following issues: any form of standardisation will be imperialistically imposed and based on ‘idealised native speaker norms’; mistakes made are simply non-standard forms, and as such are only important if they impede communication; we need ELF because linguistic nativeness is culturally-tainted; two earlier versions of international standardisation, (i) Quirk’s Nuclear English (which was over-prescriptive, not really international, and inadequately focussed on learner problems), and (ii) Grzega’s ‘Basic Global English’ (which has yet to be fully formulated), have, despite their drawbacks, provided useful bases for a new teaching paradigm; any form of standardisation will prevent organic growth; a learner-corpus will provide ‘real data’ about communication breakdown between non-native users of English, and so will give us important information to help ‘weight’ our syllabuses; ELF can be presented to students as just one version of English out of many.


To conclude this second part of my article, I want to say that the notion of ELF has clearly made a significant mark in the discussion of English Language Teaching, especially by foregrounding the plight of non-native=speaker teachers. But in my view it has been misunderstood, because it cannot and should not provide the basis for any language teaching syllabus.


Hammond B (2007) Should the ELT goal be native-speaker-like use of English? MA Paper, Kings College London University
Jenkins J, Modiano M, Seidlhofer B (2001) ‘Euro-English’, in English Today 17/4
Jenkins J (2002) ‘A sociolinguistically based, empirically researched pronunciation syllabus for English as an international language’ in Applied Linguistics 23/1
Jenkins J (2003) World Englishes: A Resource Book for Students, London, Routledge
Jenkins J (2006) ‘Testing and EIL’, in ELTJ 60/1 January 2006
Jenkins J (2007) English as a Lingua Franca: Attitudes and Identity, Oxford, Oxford University Press
Kachru B (1986) The Alchemy of English: The Spread of Alternative Englishes, Oxford, Pergamon Press, reprinted 1990, Urbana: University of Illinois Press
Kachru B et al (2006) Handbook of World Englishes, Oxford, Blackwell
Seidlhofer B (2001) ‘Closing a conceptual gap: the case for a description of English as a Lingua Franca’, in International Journal of Applied Linguistics 11/2

Author Biography

Charles Tim Lowe has been involved in ELT since 1975. He has taught and trained in several countries, written extensively on ELT, and, after completing his MA, lectured at the Institute of Education London University. He established the original Distance DTEFLA, and managed two flagship language schools. Now, having taught EAP and CLIL for 6 years at Sophia University Tokyo, he has returned to teach English at IH London, where he once again teaches both General and Business English.