T is for Task-based Learning

13 11 2011

I’m off to this conference next week, where I’ll be attempting to situate Dogme/ Teaching Unplugged within the wider orbit of task-based language teaching (TBLT).

To tell the truth, I find the thought of it rather daunting, given not only the calibre of the other presenters (see the programme here) but also the fact that Dogme doesn’t have a shred of hard research evidence to support it.  TBLT, on the other hand, seems to be all research and very little actual practice. Yet I’m also intrigued as to why I’ve been invited, and wonder if this isn’t a sign that either Dogme has come of age, or that it is in danger of losing its edge. Or both.

It also comes at an opportune moment, as dogmetists start to engage with the need for serious research. In my presentation I will be indicating the kinds of research questions that I hope to see addressed. This in turn will involve highlighting, in the burgeoning research into TBLT, those particular studies that might also validate a Dogme approach. It’s always been my claim that Dogme shares many core principles with TBLT, but without the more elaborate ‘architecture’ usually associated with the latter. As Luke and I say, in Teaching Unplugged, “where a Dogme approach parts company with a task-based approach is not in the philosophy but in the methodology” (p. 17). Hence, a lot of the research that underpins TBLT, especially with reference to the basic claim that ‘you learn a language by using it’,  has more than passing relevance to Dogme.

Dogme in relation to TBLT and content-based instruction

All this has led me to re-visit the entry for task-based learning in An A-Z of ELT, in which I claim that TBLT

has been influential more at the theoretical and research level than in terms of actual classroom practice. One reason for this is that a focus on tasks requires a totally different course design, not to mention the implications for testing. Also, for many teachers, a task-based approach represents a management challenge.  How do you set up and monitor tasks in large classes of unmotivated adolescents, for example? And how do you deal appropriately with language problems that emerge spontaneously from the task performance?  A grammar-based syllabus and a PPP approach offer greater security to teachers with these concerns (p. 224).

This is a little ironic – cheeky, even – given that the same criticisms have been levelled at Dogme, i.e. how do you cope with unpredictability, not to mention students’ – and other stakeholders’ – need for a syllabus?  More to the point, are these criticisms of TBLT justified?  Is it really a laboratory artefact, or does it have a life of its own?  And is it so difficult to implement?

Information gap task

The literature suggests that it is. Rod Ellis (2003, p. 322) concludes that “overall, task-based teaching, while superficially simple, is complex”. One reason that it is complex – according to Ellis – is that, if their potential to promote language acquisition is to be realised, tasks need to have a linguistic focus as well as a communicative one. That is, it’s not enough that you describe this picture to me and I draw it. Rather, the task should require that you or I, or both of us, focus on some linguistic feature of the interaction that we haven’t yet internalised.  Engineering this dual focus is no mean feat.

It’s not just a management issue (e.g. how do I draw learners’ attention to form when their primary concern is on meaning?), but a course design issue: how do I design tasks that require the use of specific linguistic items, and how do I design a syllabus of tasks that covers the items that I assume the learners will need?

This is where the Dogme takes a more relaxed attitude, perhaps. By banking on the fact that, if you use language purposefully, intensively and communicatively, you will ‘uncover’ the syllabus that you need, the requirement for ‘focused tasks’ (i.e. tasks that target a pre-selected language feature) is obviated. The learners’ linguistic needs are met (so the theory goes) if their communicative needs are met.  And their communicative needs are met if they’re given the space, and the incentive, to realise them.

Besides, it seems to me that a lot of the literature on TBLT is aimed at finding the optimal configuration of task design factors – such as rehearsal, planning time, collaboration, and so on  – that in turn impact on accuracy,  fluency and complexity. Calibrating these different factors requires an almost obsessive attention to detail. Yet, as Michael Breen (1987, 2009) pointed out:

Perhaps one of the most common experiences we have as teachers is to discover disparity between what our learners seem to derive from a task and what we intended or hoped the task would achieve. Whilst the objective of the task will have been reasonably precise, actual learner outcomes are often diverse, sometimes unexpected, and occasionally downright disappointing (p. 334).

If task-based teaching is so fundamentally unstable, why not opt, instead, for maximising those features of the classroom ecology that really do have strong and predictable effects, i.e. granting learners some control of the agenda?  Where learners have some ownership of, and investment in,  their language learning  program, the fact that it’s task-based, or text-based, or even grammar-based, is of relatively little consequence.

But do I dare say this at the conference!?


Breen, M. 1987. ‘Learner contributions to task design’. In Candlin, C., & Murphy, E. (eds.) Language Learning Tasks. London: Prentice Hall. Reprinted in van Branden, K., Bygate, M., & Norris, J. 2009. Task-based Language Teaching: A Reader. Amsterdam: John Benjamins.

Ellis, R. 2003. Task-based Language Learning and Teaching. Oxford: Oxford University Press.