How much do you pay attention to how and what you say (how you model language) in the language classroom?
The voice as classroom technology
Afterwards, when I talk about a lesson with myself or my colleagues, I usually describe what we did, or didn’t do, and what technology we used or didn’t use. Thinking about it, the great irony is that VOICE is often the bedrock classroom technology used to take action, and it may very well be the one that busy, dedicated, reflecting teachers reflect on the least. Yet another reason to adopt video recording as a reflective tool.
While I’d heard about teachers as language models before, I’d never thought too deeply of myself as one until I started paying attention to the words coming out of my mouth in interaction with others while doing an MS. Looking back, I’m a little surprised it wasn’t addressed on my Dip.TEFLA course. Perhaps this is because at the time DipTEFLA was ‘native speaker’ centric, and focused on authentic materials (not produced for language learners) — I wonder if we ‘native speakers’ were deemed less authentic than the songs, news broadcasts, and TV shows we could use if we accommodated our talk to our learners…
During my MS, I started reading about modelling and noticed (even more these days) that the trend in ELT is to poo-poo this role because it a) encourages teacher-centered-ness, b) hearkens back to audiolingualism and its attention to accurate native speaker models, and c) encourages teacher talk, which are all generally thought of as bad things.
Voiced comprehensible input
Well, I see all those points, but isn’t it difficult to disagree with the following facts/principles?
- whether highly or not-so-highly proficient in speaking/listening, a teacher’s L2 talk should be heard and understood by students
- exposure to meaningful and comprehensible (to them, not us) L2 input of any sort (hello again, ELF and EIL) is beneficial to L2 learners
- any L2 coming out of an L2 teacher’s mouth is input, and as such should be considered in terms of the learners’ varying levels of listening proficiency. Failing that, perhaps it should be eliminated, as incomprehensible input generally produces a lot of undesirable outcomes.
Then throw in the different classroom cultural expectations around the world, and we can understand further in affective terms that students with particular backgrounds pay much more attention to teacher modelling than others, and feel more comfortable when they can. In these environments I’ve found students highly receptive to modelling as an element of activity setup (and subsequent activities likewise successful, in terms of both accuracy and fluency).
Some ways to model language
- write partial or complete prompts (partially elicited) on WB, with choral repetition
- backchain (chunking an utterance in reverse) to get at reduced speech, again with choral rep.
So, instead of casually/nonchalantly telling students to find out from their partners three things they’ve never done, I’ll elicit the basic question structure (“have you ever”) first, practice the reduced form intonation chorally (with visuals), get everyone to try one question and check around, then turn them loose on each other for a specific, amount of time.
Or, instead of telling them to find out what their partners did the night before, I might lead them to backchain “wadidjadoo las’night” first, then ask and answer for a specific listening purpose.
To modify or not to modify
An example: a couple of days ago I was watching teaching video of myself with some students in a TESOL course, and taking feedback on it, and we noticed that the pace and clarity of my talk varied with the degree of spontaneity of my utterance; I’d rush and slur/mumble when following up on conversational utterances from students. Chances are, because most if not all spontaneous action is in fact the result of loads of practice, I learned to slur/rush my talk in daily life, and hence this is probably a common feature of my speech that I wasn’t aware was this bad.
That said, one student wondered if occasional brief utterances of this sort aren’t such a bad thing, because they make the situation more authentic for the student: they will need to construct meaning from a few prominent sounds and gestures in context. That makes sense to me, but unintentional moments like this need to be the exception to the rule.
Perhaps if language teachers concentrate on using their talk as models, “superior” and “non-superior” L2 users alike can in fact drastically reduce the amount of t-talk while simultaneously raising its usefulness to learners and giving them the floor for more minutes of the classtime.
What are your thoughts about the teacher as language model?