About three weeks ago I decided to start recording my students telling stories. The reactions ranged from horror to outright refusal (they changed their minds after everyone else did it), but the interesting thing was that afterwards, when we analysed the recordings, they were all extremely positive about the whole process. I used two basic approaches.

1. I had found a radio programme where a woman talked about her first experience in the 1980s with the 'new generation' of Marks and Sparks sandwiches, talking about how they seemed extraordinary at the time. I decided to record my students talking about their first experiences of food or, failing that, an anecdote or powerful memory related to food. I gave them some time to prepare it (with use of dictionaries, without writing anything). We analysed the extract from the radio programme for typical features of storytelling and spoken discourse and then tried to apply some of these things to improve what we heard on the recording. We also went through the recordings looking for other improvements that weren't necessarily exemplified in the radio programme. Basically we deconstructed and reconstructed their stories as a group.

2. For lower levels, simple analysis of the story without reference to L1 model. I usually tended toward analysis of hesitation with these Ls.

Some of the typical problems that arose, even at higher levels, were:

a) no pronoun substitution and therefore unnatural repetition

b) unnatural hesitation

c) failure to 'close' the story so that the listener didn't know if the speaker was finished or not

d) failure to provide proper emphasis and intonation for key moments of the story

e) lack of 'vague' language

I decided not to ask them to retell the story after the analysis as I thought they would be too self-conscious and fluency would suffer. I don't know if that was the right decision.

I'm obsessed with hesitation at the moment. When we heard a particularly long and distracting 'errrr' in the recordings (and there were a great many), I asked them to tell me why. If their answers are to be believed, there are three main causes of unnatural hesitation:

1. They don't know the word and have to paraphrase

2. They do know the word but it's not immediately accessible

3. There are two or more words competing for use

The first problem is fairly easily remedied. The second shouldn't have been an issue since I asked them to tell themselves the story mentally in English before recording it. My guess is that very few people actually did this. On the other hand, in the heat of the moment, we tend to want to improvise, or something comes to mind that we hadn't thought of before ( as E.M. Forster once said, "How do I know what I think until I see what I say?"). For many of them it may have been the first time they had told that story in any language. Maybe the next time I do this the only specification will be that it has to be a story they've told many times already.

Intonation and emphasis at key moments in the story came up only towards the final recordings and I realised only then how crucial it is to successful communication. Unfortunately, it also seems as if it's the most difficult thing to master. I really don't know how we can teach it without ending up with speakers who sound like bad actors. Intonation is a kind of emotional fingerprint and therefore is something that can only be absorbed where there is a 'whole person' relation, i.e. face to face contact with a great many different L1 speakers. Or maybe not. Then there is always the question of just how much a learner is willing to change to sound authentic.

I think particularly with storytelling, teachers find themselves dealing with issues that are not strictly connected to L2 competence. Some people are naturally better at it than others and some may be entirely incapable of coherent longer turns even in L1. Perhaps this makes a lesson of this kind doubly valuable for Ls - they get two skills for the price of one. On the other hand, perhaps the general strain of having to really think about what you say before you say it makes your overall production in L2 a little better in some respects.

Another problem was the unnatural setting. Recording yourself is stressful, but I told them that this simpy mimicked the pressure of speaking in L2 to people you don't know very well. The cause of stress in both cases is fear of judgement (it also occurred to me that it seems a little ironic that many people like talking but no-one wants to listen to themselves as they do it). The setting also made the listeners understandably reluctant to interrupt or ask for clarification or take part in some collaborative work with the speaker, something that is completely natural and essential in L1 storytelling, especially in 'rounding off' a story. The next time I do this, I think I'll try and make it a little more natural by inviting the other students to be a little more interactive.

I should say again now just how much the Ls liked this activity. There was usually always a feeling of "Wow, it's not actually that bad!" We all had a lot of fun.