fbpx
Design @teflninja

Mike Long’s Task Based Language Teaching and Native Speakerism

In April I had the pleasure of finally reading Mike Long’s Second Language Acquisition and Task Based Language Teaching. I was incredibly impressed with the academic rigour, the breadth and depth of the writer’s knowledge, but most of all (as a practising teacher) with the far-reaching practical implications. Having said that, there was one aspect which kept on cropping up throughout the book that made me uncomfortable, namely the idea that authentic language and texts are those produced by ‘native speakers’, and that these ‘native speakers’ are by definition better models of the language and task performance.

To me this is a prime example of how deeply ingrained the ideology of native speakerism is in the minds not just of students who demand classes with ‘native speakers’ or recruiters who refuse to hire ‘non-native speaker’ teachers, but also in the minds of ELT and SLA professionals.

Before I move on to show a few examples of native speakerism that I encountered in Long’s book, let’s first define what native speakerism is.

What is native speakerism?

The term native speakerism was originally coined by Holliday (2005, 2006), who used it in reference to the notion that the linguistic and pedagogical ideals of teaching English spring from Western culture, which a ‘native speaker’ embodies. Houghton and Rivers (2013a) point out that native speakerism has its roots in the dichotomous discourse of us and them, ‘native speakers’ and ‘non-native speakers’, where the former are usually seen as the norm and ideal both in terms of language use and teaching skills, while the latter as deficient and inferior. Thus native speakerism can be understood as

a prejudice, stereotyping and/or discrimination, typically by or against foreign language teachers, on the basis of either being or not being perceived and categorized as a native speaker of a particular language. (…) Its endorsement positions individuals from certain language groups as being innately superior to individuals of other language groups (Houghton & Rivers, 2013a, p. 14).

Of course, as any ideology, native speakerism does not spread in a vacuum, but is maintained, supported and normalised by powerful discourses which make it seem justifiable and acceptable. These are then used as a basis of social practices and actions.

To give one example, native speakerism is supported by the discourse that ‘native speakers’ are better models of the language for our students, and therefore students should be exposed to ‘native speaker’ language in class in order to improve their proficiency. This might lead us to select predominantly materials created by and for ‘native speakers’.

Native speakerism and Long’s TBLT

So how is native speakerism manifested in Long’s discussion of TBLT?

The first clue is how authenticity is construed in the book. For example, Long defines genuine tasks as those “originally designed for native speaker – native speaker communication, not LT” (p.21).

Later he defines authentic materials as “genuine texts, such as song lyrics, news broadcasts, films, newspaper articles, and textbook chapters, originally created by and for native speakers (NSs), not for LT to non-natives” (p.249).

You could argue that in both cases Long’s emphasis is on the fact that authentic texts are not created specifically for language teaching, which is something that I think we’d all agree with.

However, if this was the case, why mention that authentic texts are created by and for ‘native speakers’? Wouldn’t it be enough to say that authentic texts are those originally not intended or created for language teaching and learning?

It would unless you believe that only ‘native speakers’ can be the choice of authentic material and real language.

An interesting indication that this might indeed be what Long believes can be found on p. 271, where Long presents a task whose aim is for students to learn to obtain and provide directions. The first pedagogical task involves listening to three conversations and is incidentally called “The real thing”.

Guess who recorded the conversations? A ‘native’ or a ‘non-native speaker’?

If you answered the former, then well done!

Indeed long writes that the three conversations are to be “real examples of NS giving directions” (p. 271).

So perhaps Long does indeed believe that ‘native speakers’ are by default better language models for our students?

A further clue to answering this question can be found on p. 313, where Long discusses the fifth methodological principle of TBLT, which involves promoting inductive learning of chunks. He suggests that an extensive reading and listening program should be added to the main classroom course.

That per se is perfectly justified and empirically sound given the evidence. However, what is highly questionable in my opinion is his suggestion that students should listen to and read “lively recordings of the texts made especially for language learning by a native speaker [emphasis mine]” (p. 313).

By now, it seems to me that it is impossible to argue that Long is unaware of the implications of his adding the word ‘native speaker’, neither on p. 313, nor in any of the previously quoted examples.

His thesis is otherwise incredibly detailed, his claims based on VERY extensive reading, and his arguments are always phrased carefully and eloquently.

Therefore, I’d argue here that he’s well aware of the implications. In fact, I’d go further and say that he actually believes that:

  • students should be primarily exposed to ‘native speaker’ input
  • only ‘native speakers’ can be a source of authentic language input.

In fact, when a fellow teacher emailed Long to clarify what his position was, his answer was very clear: ‘non-native speakers’ are inappropriate as task models (unless the target task typically involves ‘non-natives’) and ‘native speakers’ are better models of the language.

This is further evidenced by Long’s views on who should conduct a needs analysis.

On p.136, in reference to Selinker, Long writes that an expert informant for a needs analysis “should be a native speaker, well trained and competent in the field of interest”. Again, this begs the question why it should be a ‘native speaker’? Wouldn’t any sufficiently proficient speaker do?

They probably wouldn’t to Long.

When he discusses the use of elaborated input in tasks (rather than simplified or graded input) on p. 253 and 254, he writes that the addition of “to earn money as an implicit paraphrase of to provide for (to earn money to provide for his family)” would be redundant for a ‘native speaker’.

Really?

I’d argue that it would be redundant for a proficient speaker, regardless of their L1.

Having said that, it could also be necessary and appropriate to add it there in a natural conversation to facilitate understanding. There’s plenty of lexical redundancy and paraphrasing in natural speech.

So, bearing all of the above, it seems clear to me that the implicit idea in Long’s version of TBLT is that a ‘native speaker’ is simply by definition always more proficient and as a result would make a better language model.

Interestingly, however, authentic and real tasks will involve ‘non-native speakers’ interacting with other ‘non-native speakers’, rather than exclusively ‘native speakers’. Therefore, if we are to promote authentic input and authentic tasks, these can’t be restricted to ‘native speakers’.

In fact, in the majority of contexts, save a few rare cases where our students for some reason are going to exclusively interact with ‘native speakers’, restricting the input and task models to ‘native speakers’ might not appropriately prepare our learners to use English effectively outside the classroom. In addition, focusing only on ‘native speaker’ language input can give students the idea that ‘non-native speakers’ are inappropriate language models.

Finally, the idea that any input from any ‘native speaker’ is always a better and more authentic model seems to me to be completely erroneous and evident of how deeply embedded native speakerism still is both in ELT and SLA.

References:

  • Holliday, A. (2005). The struggle to teach English as an international language. Oxford: Oxford University Press.
  • Holliday, A. (2006). Native-speakerism. ELT Journal, 60(4), 385–387. https://doi.org/10.1093/elt/ccl030
  • Houghton, S., & Rivers, D. J. (2013). Introduction: Redefining Native-Speakerism. In S. Houghton & D. J. Rivers (Eds.), Native-speakerism in Japan. Intergroup dynamics in foreign language education (pp. 1–16). Bristol: Multilingual Matters.
  • Long, M. H. (2014). Second Language Acquisition and Task-Based Language Teaching (1 edition). Wiley-Blackwell.

Would you like to learn how to tackle native speakerism, promote equality and teach English for global communication?

Check out TEFL Equity Academy and start your FREE 30-day trial today.

What’s included?

  • 10+ courses
  • interviews with experts
  • new content added monthly
  • LIFETIME access
  • downloadable pre-recorded video presentations
  • lesson plans and teaching ideas

Start your FREE 30-day trial now.

Show this post some love. Share it with colleagues.

Share on facebook
Share on whatsapp
Share on twitter
Share on linkedin
Share on reddit
Share on vk
Share on pinterest
Share on email

12
Leave a Reply

Leave a Reply

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

  Subscribe  
newest oldest most voted
Notify of
Jason Anderson
Guest

Hi Marek, Very well argued. You raise some important points here that I agree with. This old fashioned notion of ‘authenticity’ that Long perpetuates in his writing is implicit within a number of definitions of TBLT (and therefore in the way some authors have constructed the notion of ‘task’), where it tends to sit dangerously close to rather dated concepts of proficiency=native-speaker-like usage. I have offered an alternative perspective in my writings on translingualism (see my piece in ELT Journal this year: Anderson 2018 Reimagining English language learners from a translingual perspective) where I argue that “reimagining the language classroom… Read more »

eflnotes
Guest

hi Marek did you also have a look at Long’s uses of ‘foreign[er]’? would such a look show an //implicit idea in Long’s version of TBLT that a ‘foreigner’ is simply by definition always less proficient and as a result would make a worse language model//? your point re genuine/authentic text – how does using the term ‘native’ in Long’s description of differences between synthetic and analytical syllabuses contribute to ‘native speakerism’? similarly how does adding the qualifier that Long does in ““lively recordings of the texts made especially for language learning by a native speaker with excellent diction and… Read more »

Neil McMillan
Guest
Neil McMillan

To Jason Anderson: for the record, contrary to what you imply, Long nowhere argues that “authentic is best”, at least not for anyone other than the most proficient speakers.

Jason Anderson
Guest

Hi Neil, Thanks for your comment. I didn’t say that Long claims that ‘authentic is best’. My reference was to the arguments of Guy Cook who points out this tendency and the dangers it carries. Cook’s 1997 article “challenges the widespread belief in contemporary ELT that students should be exposed to authentic or natural language, and that such language is primarily focused on making meaning and achieving practical purposes.” (p.224) Nonetheless, I think, even if it’s not explicitly stated, there is an underlying argument in Long’s work that implies that authentic is preferable. I think Cook would agree with me,… Read more »

Neil McMillan
Guest
Neil McMillan

I recommend you read Long’s 2015 book where his position on authentic materials is pretty unequivocal, although I do agree with Marek that the definition of “authentic” adopted leaves something to be desired.

Robert Taylor
Guest
Robert Taylor

Hi, Marek,

I’ve posted a brief response to your post under the title ‘Long’s Natives’,

https://wordpress.com/view/englishteachingandissues.wordpress.com

Robert Taylor
Guest
Robert Taylor

(3rd attempt)

Hi, Marek,

I have responded to your post on my blog at englishteachingandissues.wordpress.com

trackback

[…] the beginning of his blog post, located here, Marek notes Long’s association of ‘authentic texts’ with ‘native […]

Joe
Guest
Joe

Hi Marek. I was just reading Long’s book when I came across a section on page 170 and I remembered reading your blog post. In discussing analysing texts for classroom use, he says:

“Conventional linguistic analyses are replaced with analysis of target discourse, sample authentic texts by target discourse samples (with authenticity not necessarily judged according to native speaker norms).”

So while there are is a certain amount of native-speakerism in the examples you have highlighted, there is certainly also an awareness, at least in this example, that native speaker norms are not necessarily appropriate for all learners.

Join our tribe

Support Equality. Tackle Native Speakerism.

Share on facebook
Share on twitter
Share on whatsapp
Share on linkedin
Share on pinterest
Share on reddit
Share on vk