Dissing Dogme Part Two: Carrying TBL to its demented logical extreme
So, ding ding, gloves on! Here we again with Round Two of cheap pops at Dogme. I bet you can’t wait.
Today what I’d like to focus is an issue right at the heart of Dogme: when and how students are given language, the medium through which new language is delivered and the issues arising from this. The title of this post refers to TBL, and the reason for this is that in many ways I think Dogme has ended up painting itself into a bit of a corner by adopting much of the orthodoxy of Task-Based Learning. Interestingly, though, it has done so without really acknowledging as much, and even without (often) doing it in as focused a way as the best TBL manages. Now, I’m certainly no advocate for TBL, and tend to agree with much of what Anthony Bruton has said against it, but at least it had a clear aim – usually (in its early forms, anyway) geared around asking students to do meaningful tasks often related to real-world necessity or utility. These tasks might be something like visiting a doctor, having a job interview or booking tickets.
In TBL, language may be fed in during some kind of pre-task stage, where structures and lexis are either explicitly focused on or are merely hinted at, but the bulk of the language comes in response to the students’ own ability to perform the given task. Inevitably, when the whole thrust of a task is to communicate (and perhaps negotiate) a particular message successfully, students are basically free to use whatever grammar and lexis they wish. Some tasks may suggest or prompt the use of certain items more strongly that others, but at the end of the day, its the completion of the task that matters. All too often this means that students – scared of drawing attention to themselves by making what they may still perceive as mistakes – stay within the confines of words and structures they already know; in short, they get by, but do so without always bothering to make the extra effort and take the risks needed to utilise new language.
The feedback then depends both on what students themselves feel they needed, if they are willing – or able – to ask about this, and then on whatever the teacher was able to notice students having problems with. In reality, often what this means (and this is especially true if the teacher is setting up and running tasks they’ve done before and thus have some sense of the linguistic requirements of) is that teachers have an already prepared set of language ready to be wheeled out and looked at; at best, they may well focus on a mixture of previously anticipated areas (which may or not have actually been problematic, but which nevertheless will make students feel they’re getting some input and upgrading of their output) AND actual responses to things students really tried to say.
Much of the feedback will be done on a board, possibly (and ideally, I suppose) using examples written up whilst students are busy communicating and doing the task. There may well be gaps in the examples / boardwork, which the teacher then elicits, and probably some time for commentary and also questions on the linguistic input.
In many ways, this basic template, developed and honed by Prabhu during the Bangalore Project in the early 80s, fed into both the TTT (Test-Teach-Test) paradigm and, more pertinently for the purposes of this post, Dogme. However, whilst TBL at least focuses on tasks which may have some real-world utility, in Dogme, there may not necessarily even be a task, or if there is, it may just be ’emergent’, just as the students’ language is supposed to be. In reality, this may well mean one of two things: (1) a teacher walks into a room and starts chatting to students – or, in an even more Dogme twist, students start chatting to the teacher – and sooner or later either a task suggests itself or some language that needs to be fed in order to help the communication along gets given to the class (orally, or via the board) or – and I suspect this is by far the more common approach adopted by many self-proclaimed Dogmeticians – (2) a teacher walks into a classroom with a task of some kind that they want students to do. Maybe the students are to brainstorm ideas or debate a series of moral questions or listen to and discuss the meaning of a pop song or whatever. There’ll be some kind of ‘task’ which results in some kind of talking, which in turn results in some kind of input (which may well be, as stated earlier, not based exclusively on what was really actually heard, but also on past experience and prediction).
There may possibly also be some kind of learning that occurs as a result of students interacting with each other. In Dogme presentations, I’ve seen this kind of thing dressed up as the true fruits of a Vygotsky-ian social constructivist approach, but to most people it’s called accidental learning – and certainly nothing the teacher set out to actually teach!
Now, my beef here is this: why on earth do the Dogme folk think that language can only be given to students AFTER they actually need it? What grounding is there for this in any theoretical approach to learning? And what kind of load does it place on both the teacher – as sole provider of linguistic input – and on the students, who have to sit through a kind of presentation stage post hoc after every single chat / task / debate / bit of speaking? The teacher inevitably ends up being the source of input, thus increasing the risk of their own ideolect colouring the language they pick up on, whilst the medium of delivery for feedback will invariably be the board, meaning students then have to copy down what they’ve seen written up. If you then want to check the degree to which students have learned from the feedback, there needs to be a repeat stage built in later on, or some kind of parallel task, though of course 9as stated above, again) if the goal is purely driven by an interest in communication, it ultimately doesn’t matter if learners try to take new language on board or not, so long as they get their point across again!
Now, I’m all for teachers being able to do the above, but to elevate this to the be-all-and-end-all of language teaching is idiocy. What most annoys is the denial of the notion that we are actually able to predict language students may well need to do tasks. Surely one of the things students pay us for is to make informed decisions about what tools might best help them perform tasks. Oops. Wait a minute! This is actually what COURSEBOOKS do – or good ones do, anyway! Silly me. Nearly forgot.
Once coursebooks are focusing on helping students to perform certain kinds of tasks better – and with the increased influence of the Common European Framework more are, or are at least paying lip service to the idea of doing so – then the next step is to predict what lexis and what grammar might best aid students in their attempts; what kind of listening (or reading) models it might be useful for students to meet before attempting tasks – and how the task itself might best be framed in order to ensure that students are best positioned to attempt to take some of this new input on board as they attempt it. This means students have written records already provided, teachers can see how much they already know and teach the gaps by using exercises designed to get new language to students ahead of tasks and tasks can be more fruitfully and richly attempted (and, of course, as students do this, the teacher is STILL free to monitor and get new language up on the board to round up with).
Are Dogmeticians seriously suggesting that feeding in language – or testing students’ knowledge of language – before getting them to try some kind of communication is a no-no? If yes, then why? And if not, then why the coursebook hate?