I need bad models

A bad model can be more useful than a great many words … though it has to be bad in exactly the right kind of way for that person at that moment (or does it being bad in the right way in fact make it into a good model ?).

I’ve long pondered this in Module Two. Participants attend sessions which we deliver using lots of classroom management techniques we hope some of them will also apply (and those techniques are usually delivered well as most of us tutors have had a fair bit of practice and multiple opportunities to hone sequences, to know precisely what will elicit the right example or which check question / comment will work). Very often, teachers don’t notice much of this happening simply because it works. Things that are done well are seamless, they look effortless. Only a very practised observer is likely to see constituent parts. The odd time when I forget a key point (now put that paper in your pocket and don’t get it out again while you are in the circle), something goes wrong, and people notice it going wrong and realise something was missing (though not always what). But they don’t notice all the steps that made crossing groups over happen smoothly twice in that same afternoon (tell them what they will do when they are settled in new group, label people off into new group, check everyone knows which group they are going to, make sure enough tables and chairs will fit in new position, tell them not to move yet, announce where new groups will be – while exhorting the odd eager mover to sit down again, finally say move into new group). Or perhaps some notice, but it is hard to remember to actually do all of it if you don’t do it very often. But they do notice and remember (and this also comes across when you are reading peer observations), when there is a problem. If people see something not working (at least if it is the right bad model), they can often also see what should have happened, in a way that if you just ask ‘what should be done for X’, they often seem much less able to identify.

My ‘bad’ teaching models in Delta sessions are only moments when something really does go wrong, mid session, and up to now they have only ever been impromptu, so I haven’t really tested the bad model theory out there as such. But it works with phonology, with tone units. My language learners often need a lesson to help with delivery of presentations. They have the ideas and the content sorted and have checked their language, but the actual delivery is still halting or badly broken up as they are trying to remember so much at once (including even just to smile). So I show them how to check stress, chunk ideas for pauses and think about whether their voice is going up or down or neither and where it should. If they do this for their opening paragraph, perhaps more of the rest will come smoothly once they have relaxed a bit and even if not, at least they give a good first impression. I’ve found if I play them a recording of a past student, sometimes they can identify some of these elements are needed, but the way that is quickest to get them to see what a feature is and why you want them to think about it, is to read it with that thing missing. So I read a couple of simple first sentences from a presentation, but shift four or five key stresses deliberately (and only that). Then when I ask them whether it sounded good (no) and why not, they are usually quick to tell me and then it is easier to get them to identify rules, mark up text and try it for themselves as the need for it to be right is apparent to them. So here a bad model (with one feature adjusted to being ‘bad’ each time) has turned out to be the most effective tool I have to get learners to notice something.

This has also come up again recently with the group I have been teaching for iGCSE First Language English. Because their language ‘works’, because they find it easy to express needs or opinions that others understand in immediate situations (more effortlessly than many of their ‘second language’ speaking contemporaries), it has been hard to convey the need to augment and finesse it. In the exam they need to show they can operate in different registers (and they need to be able to do it in things like interview situations if they apply for schools / courses / internships, all of which have been mentioned as possibilities). I have shown them ways of adding to their stock of language and choosing from it and if they add enough, they should be able to achieve this, be able to move from style to style as the situation (or the exam question) calls for it, but it is proving much harder to get them enthused about actually going on and doing many (or even any) of the activities that would make a difference in the longer term independently. The easy, informal language they use at home answers the need to actually respond at all, and they seem to feel that if a question can be answered with ‘it’s big’. where is the need to retrieve or use magnificent, vast, huge, impressive or anything more complex or colourful.

However, one recent lesson that worked seemed to be another variant on ‘bad’ models. A lesson on TED sparked the ideas, but that one spark led me to realise how much there is on YouTube – either people saying things ‘badly’ for comic effect, or (as with Russell Brand) just kind of sabotaging their own delivery. The actual lesson plan is here, but the two threads I was trying to get across were (a) if you want people to listen to you, you have to speak to them at least partly in the way they want to hear things and (b) to do that effectively you need to work out what that entails, what features your language and delivery need to include. So really, I was just putting across the same message about having a range of language to choose from so you can achieve a style appropriate to the context / task, but coming at that message from a different angle in trying to get them to see for themselves why it might be effective. This was aided and abetted by that fact that a school debate was coming up.


YouTube has so much once you see how you can use it. I used an Armstrong and Miller sketch to elicit the ideas of different groups of people using particular language. The comedy lies in the disjunct between seeing 1940s fighter pilots on screen chatting, but hearing lexis / phrases and the concerns of a couple of contemporary schoolboys, dialogue that is rife with teenage gripes, recent slang and comments and rhythms in their speech that hum with the here and now. They were able to pick up on this easily and break quite a lot of it down to give me examples (despite not seeming to find it in the least funny).

That meant the ideas of groups of people having a particular way of talking was clear ( If I ever do this again, I think I’d add in snippets of Catherine Tate’s Posh people, or Vicky Pollard or Lauren meets Tony Blair, but the questions to use (or the worksheets if I get them to write stuff down as they watch) need thinking through, so they are looking for examples of specific things.

Then having walked them through to the fact that it you are trying to persuade someone of something in argument (essay – that’s coming soon) or debate (as they were about to participate in one), you need to speak to them in a way they expect and respect, Russell Brand provides a truly great bad model in that he does several of the things they do that undermine the impact of what they are saying and not only that, but does them to varying degrees in different situations. In the last five minutes of his delivery to the House of Commons committee on addiction, he is dressed in a way that seemed designed to get the backs of committee members up, slouches, sniggers, cracks daft jokes, gets distracted and generally seems to be grandstanding more than trying to influence policy, yet they were able to recognise that not only did he seem to have a serious message, but that he was potentially undermining it by doing these things if he alienated his audience. I had prompts ready to try and bring out things like addressing someone as ‘mate’ as being unlikely to get his message ‘heard’, but I didn’t need them. Just for once they were falling over themselves to tell me what he was doing wrong (and with just a little questioning to acknowledge that it was not that he didn’t have anything sensible or useful to say). It continued to work with the second clip, where (by luck or by judgement) he seemed to have tightened up his delivery on Question Time. There are still some asides that don’t help, but overall, he does much less to trivialise the impact of what he is actually saying and as a result seems to get a proportionately greater response from those listening to him. Again, they were able to tell me what he had changed and what was still there (both in terms of things that would be good to adopt and of things he’d probably do better to lose).

Perhaps it was just luck, perhaps the group were just having a particularly switched on day, but it did seem as though those ‘bad’ models hit the spot in getting them to notice the impact that conscious use of language (as opposed to just going with your default mode) can have. They can’t be a whole syllabus and a person needs good models too ( to borrow from and compare themselves with), but a few well placed, well chosen bad models really can have an impact. Now how do I get some for teacher training ? Endless filming and requests for permission to use ? But people won’t want their mistakes frozen in time, however useful. Or might the fame outweigh that ? Tutors acting ? Would that be convincing in any way  … ?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s