Demystifying Task-Based Language Teaching

One of the hundreds of things that annoys me about language teaching1 is the use of jargon to the exclusion of all else when explaining how something works in a relatively simple way. Sometimes it’s not all that simple, but what could happen if we take the approach of keeping things simple and explaining everything as we go? Perhaps people will stop saying that the alternatives to the status quo in language teaching in general and English language teaching in particular are unworkable.

night-of-the-living-dead-group
Image: Night of the Living Dead, Public Domain.

With this in mind, I decided to write a book about Task-Based Language Teaching (TBLT) to demystify it and make it accessible to ordinary teachers who would like to try it but got put off it by the complexity and the density of the research literature. Then I put it off because my colleague in SLB Co-op, Neil McMillan, told me about his and Geoffrey Jordan’s plan for an online course to make TBLT a bit more viable. He then asked if I wanted to be a part of it. I jumped at the chance.

I’ve wanted to make TBLT less of a messy learning process since I had a messy learning process with it myself during my DipTESOL. It is with hindsight that I realised it really should not have been that way. If I had a mentor to guide me through the contradictory information regarding different task-based models (and there are some different ones, which I’ll look at below), and talk about how to sequence lessons and what the prerequisite work is before you even start teaching, I am sure that my stress levels would have been lower and my students might have had a couple of more straightforward lessons.

Generally speaking there are three main ‘task-based’ approaches.

1. The Nunan (2004) model, which uses tasks but really is just a Present-Practice-Produce lesson done as Produce-Practice-Present because the teacher has preset plans for focus on language items.

1a. The Willis & Willis (2007) model, which does pretty much the same as the Nunan model (in my opinion) but uses a task cycle and greater reflection but still with the focus on language items but advocating more focus on lexis.

2. The Ellis (2003) model, which mixes ‘real-world’ tasks with tasks that exist solely for pedagogy, using what Skehan (1998) calls “structure-oriented tasks” (Skehan, 1998. p. 122-123). Essentially, there will be tasks that are only there to induce use of certain grammar, vocabulary or functions. However, there is more room for Focus on Form (looking at language that learners have shown they need, rather than what the teacher presumes they need).

3. The Long (2014) model. The syllabus is created by needs analysis, and the typical language used to do it is sourced by an “analysis of discourse” (Long, 2014. p.??) Put simply, find out what learners need to do with the language and how it would normally be done. There is no looking at language just to shoehorn a language point into the lesson. Instead, there is only Focus on Form, again focusing on language learners have displayed a need for in their output (bits of speaking or writing that are not quite what would be considered an appropriate way to communicate within the group of people one intends to communicate with) or displayed a lack of understanding of in their input (i.e. bits of reading or listening that have not been understood).

SLB Co-op prefer the Long model, but acknowledge that it can be somewhat difficult for teachers without a university department full of applied linguists supporting them. This is why we are going to look at how we make TBLT workable in the real world.

The course, created by Neil McMillan, Geoffrey Jordan and I with guest contributions from Mike Long and Roger Gilabert, starts next year. To be a part of it you can find out more here.

1. Actually there probably are hundreds but there are thousands of things that I love – but anyway, I digress

References

Ellis, R. (2003) Task-based Language Learning and Teaching. Oxford: OUP.

Long, M. (2014). Second language acquisition and task-based language teaching. John Wiley & Sons.

Nunan, D. (2004) Task-Based Language Teaching. Cambridge: CUP.

Skehan, P. (1998). A Cognitive Approach to Language Learning. Oxford: OUP.

Willis, D. & Willis, J. (2007) Doing Task-Based Teaching. Oxford: OUP.

"This is gold!"


I’ve been using Saboteur with an adapted Kotoba Rollers framework by James York with my university classes. I want talking with authentic tasks, which games provide. There is also transcription of language used. It isn’t all fun and games though.
In the game, players are either good, hardworking miners or saboteurs. None of the players know the roles of the others but they hardworking miners need to work together to get the gold. The saboteurs need to ensure the pack of cards is exhausted before the treasure cards are reached. There are also action cards such as breaking tools, fixing tools, causing rock falls and checking maps for gold, which may lead to cooperation or subterfuge.
The published rules are a bit tricky to understand. I had set the reading for homework, figuring that if there were a lot of difficulties the students would use dictionaries or Google Translate. This means my students skim read them superficially and did not bother to understand the rules fully before game play. Dictionaries and Google barely got looked at.
However, the rules needed a bit of clarification. This led to some good negotiation of meaning (Long, 1983). There are cards used to destroy the mine path above or break other players’ tools but they weren’t always easily understood.
The transcription is the main part I changed. I ask students to write three parts.

What did your partner say? Did they say it differently to how you would say it? How would you say it?

This has been done pretty well and is usually the best part of my RPG-based classes’ sheets, too.

What communication problems did you have? Why?

This sometimes ends up being a wishy-washy “I need to speak more fluently” but a lot of my students have gone a bit deeper.

If you spoke Japanese, what did you say? How can you say it in English?

This has an obvious function but students do sometimes half-arse it and just use Google Translate one way without checking the translation in a (monolingual) dictionary or Skell.
Still the work got done and there was another game of Saboteur in the following lesson to review. I was satisfied with this little Kotoba Rollers cycle, and so were my students, though I needed to buy 4 lots of the game for my big class.

References

Long, M. (1983) Native speaker/non-native speaker conversation and the negotiation of comprehensible input1. Applied Linguistics, 4 (2) pp. 126–141.

Rod Ellis: Moving task-based language teaching forward

Regular readers know that I am a big advocate of Task-Based Language Teaching. In this video Dr. Rod Ellis discusses some of the problems/issues/misconceptions in TBLT. One of my favourite parts is task complexity, another is teacher education, though the latter is rushed through much more than it ought to be.
If you like this video, or just don’t have an hour to spend on it (which is a shame), you can have a look at my previous post on how to actually do task-based teaching, which is a rough and dirty guide.

How I Plan Lessons

Massive disclaimer: plan like this and you will almost certainly fail certificates and diplomas. This is how I plan lessons in a task-based framework that’s a bit Long/Skehan-influenced. However, if you wish to reclaim time for leisure, read on. Many thanks to Kamila and Sarah for the spurring on.
img_1991
First thing, refer to your syllabus and notes from the previous lesson. What did you plan for in the syllabus? Does this need to change?
Syllabus: persuade people. (Authentically vague note). Remember a weakness in dealing with difficult people.
Next, what is your target task/exit task, that is what do you want learners to be able to do by the end of the lesson? How much time do they need to do it? How much time do you expect to need for reflection and feedback?
Task: Persuade a colleague to visit a disagreeable client. Estimate: 12 mins task, 5 mins reflection and feedback.
Regarding the exit task, can it be broken up into smaller components? What are they?
Greet, broach a difficult topic, hedge, point out advantages, bargain.
Would a text be useful as an example? Do you expect to do decoding, vocabulary, pragmatics, semantics, grammar, pronunciation or discourse work?
Yes. I’d love an authentic text. Unlikely though. Maybe something like a documentary or the BBC version of The Apprentice. Use an excerpt. Note time codes for difficult words/elements of connected speech. Likely 15-20 mins.
Can you cover all the smaller tasks and the text in one lesson or do you need longer, once the exit task is added to the end?
Probably in a 90-minute lesson. Greetings are fine. Broaching needs 5 mins + 5 minutes Focus on Form (likely discourse markers so prepare some corpus lines, perhaps). Hedging probably 3 mins FonF 2 mins, combine with broaching 7 mins and 2 mins FonF. Point out advantages – maybe 5, seems good for schema activation. FonF might be intensifiers. Bargaining, 6 mins with FonF around  7, possibly syntax with conditionals/modality. The FonF is just predicted. It might be totally different depending upon task performance. Component tasks may be cut as needed (see below).
What will you do to activate schematic knowledge? What about differentiation?
Brainstorm a list of advantages of talking to difficult people. Choose most persuasive three. 7 mins.
Put stages in order. 
Schema activation, 7 mins
Attempt task, 10 mins. FonF 5-10 mins. If task OK, add complexity.
Text work.
Decode these words:


tough /tʌf/

(the) first /ðə fɜːst/

I wanted to /aɪ wɒnɪtʊ/
Listen, summarise, check.

Till 13:09
FonF
Component tasks with FonF
Exit task, feedback, homework.
Gather materials.
Probably copy and paste the corpus lines (linked above) into a document, blank out the adverbs. Give it to my student if required.
Cheers, I planned my lesson.