C is for Conversation

8 05 2011

A core tenet of the Dogme philosophy is that classroom learning should be ‘conversation-driven’, and that, out of the language that emerges from this conversation, language learning episodes can be co-constructed.

But what do we mean by ‘conversation’?

In a recent comment on Diarmuid Fogarty’s blog, Luan Hanratty wrote:

Conversation is the word we really need to define. Jack Richards wrote that interactions fall into three basic categories: small talk, transactions and performances. For me, small talk is the most important because it constitutes a difficult social skill that is often least practised among learners. But small talk is too mundane to base a whole class around, hence the need for materials.  Dogme has the danger of becoming like the people who tweet about what they had for lunch. Pleasant but not very inspiring, especially in an learning context. Surely we can do better by giving more time to transactions and performances, i.e. speech acts rather than coffee chats.

This is a fair criticism, especially if we construe conversation as being synonymous with ‘chat’, which by definition is largely interpersonal in terms of its function, and local – even trivial – in terms of its field. If learning opportunities are based solely upon this fairly restricted register, it’s unlikely that most students will find their communicative needs are satisfied – especially if these needs include more formal registers, such as academic or technical writing. (At the same time, it’s worth noting that even written registers are becoming increasingly ‘conversationalized’, especially since the advent of digital media).

So, effectively, a classroom conversation needs to be more than chat. And it also needs to be more than the teacher-led question-answer sequences that characterised Direct Method courses, and which are so easily ridiculed: How many fingers do I have? Do I have a nose on my face? Is this your neck? etc.

Causeries avec mes élèves

Incidentally, one of the prototypical Direct Method courses was called Causeries avec mes élèves [Conversations with my students, 1874)]. Its author, Lambert Sauveur, describes the first lesson: “It is a conversation during two hours in the French language with twenty persons who know nothing of this language. After five minutes only, I am carrying on a dialogue with them, and this dialogue does not cease.”

While this dialogue might, in many ways, not have resembled natually-occuring conversation (the first five lessons of his course dealt with parts of the body), one principle that Sauveur rated highly was coherence, his intention being “to connect scrupulously the questions in such a manner that one may give rise to another”. As Howatt (1984) comments: “This principle probably explains his success in communicating with his students better than anything else. They understood what he was talking about because they were able to predict the course of the conversation” (p. 201).

Conversation is predictable, because one turn follows from the other. At the same time, because it is locally assembled, and takes place in real time, it is unpredictable. This tension between the predictable and the unpredictable makes conversation – real conversation –  an ideal medium for instruction.  As Leo van Lier (1996) argues, “learning takes place when the new is embedded in the familiar, so that risks and security are in balance… Conversational interaction naturally links the known to the new. It creates its own expectancies and its own context, and offers choices to the participants. In a conversation, we must continually make decisions on the basis of what other people mean. We therefore have to listen very carefully… and we also have to take great care in constructing our contributions so that we can be understood” (p. 171).

At the same time, for such conversations to provide a site for learning, there need to be strategic interventions on the part of the teacher – interventions that distinguish normal conversation between peers from what has been called ‘instructional conversation’ (Tharp and Gallimore, 1988):

The task of schooling can be seen as one of creating and supporting instructional conversations… The concept itself contains a paradox: “Instruction” and “conversation” appear contrary, the one implying authority and planning, the other equality and responsiveness. The task of teaching is to resolve this paradox. To most truly teach, one must converse; to truly converse is to teach” (p. 111).

The notion of instructional conversation has been further developed by scholars such as Neil Mercer (1995), who writes of the ‘long conversation’ that constitutes the dialogic curriculum, and Gordon Wells (1999), who calls it ‘dialogic enquiry’.    In fact, dialogue may be a better term than conversation, not least because it echoes Paulo Freire’s insistence on putting dialogue at the heart of pedagogy: “Whoever enters into dialogue does so with someone about something; and that something ought to constitute the new content of our proposed education” (1993, p. 46). In a similar manner, Sylvia Ashton-Warner yielded to – and exploited –  the hubbub in her infant classroom: “I harness the communication, since I can’t control it, and base my method on it” (1963, p. 104).

Individual presentations

So, really, conversation stands for all the talk, the dialogue, the communication (both spoken and written) that is generated by the people in the room, and that is shaped, scaffolded, supported and signposted by the teacher. It could take the form of formal debates, individual presentations, small group tasks, or a plenary discussion. It could be mediated by means of an online chat function, or Twitter, or SMS messages, or pieces of paper that are traded back and forth across the class. In the end it is simply the ‘stuff’ (to use Ashton-Warner’s phrase) out of which learning episodes are moulded. In its most basic, common-or-garden form it is simply conversation – the most natural form of communication we know.

Formal debate

And, as Gordon Wells concedes, “conversation may not be perfect as a means of information exchange… but when engaged in collaboratively, it can be an effective medium for learning and teaching. In any case, since there is no better alternative, we must do the best we can” (1987, p. 218).


Ashton-Warner, S. (1963, 1980). Teacher. London: Virago.

Freire, P. (1993). Education for Critical Consciousness, New York: Continuum.

Howatt, A.P.R. (1984). A History of English Language Teaching. Oxford: Oxford University Press.

Mercer, N. (1995). The guided construction of knowledge: Talk amongst teachers and learners. Clevedon: Multilingual Matters.

Tharp, R.G., & Gallimore, R. (1988). Rousing minds to life: Teaching, learning, and schooling in social context. Cambridge: Cambridge University Press.

van Lier, L. (1996). Interaction in the language curriculum: Awareness, autonomy and authenticity. Harlow: Longman.

Wells, G. (1987) The Meaning Makers: Children learning language and using language to learn. London: Hodder and Stoughton.

Wells, G. (1999). Dialogic Inquiry: Towards a sociocultural practice and theory of education. Cambridge: Cambridge University Press.



42 responses

8 05 2011

Hi Scott

This is a well-researched explanation and I think you are right that ‘dialogue’ is a more accurate word in a language learning context. I think the difference between the two is that conversation is an art. You might say that someone was a good bar-room conversationalist but you wouldn’t say they were a good dialogician.

Good conversationalists are chameleons – perhaps there is a machiavellian connotation to that but people do need to adopt different codes for different people; from the duke to the dustman as it were. Good conversationalists are likewise erudite, articulate and urbane, i.e. book clever, word clever and people clever. Beyond this, good conversation should be open to controversial ideas and taboos as long as they are expressed with logic, reason, facts, humour and politeness. But this is a stylistic and artistic end to the skill. For the purposes of the language classroom the priorities are different. Effective communication is the bedrock and the rhetorical adornments are a luxury perhaps more suited to advanced learners.

8 05 2011
Mr Darkbloom

Thanks; very nice, Scott. (and nice post-dogme symposium video, which I enjoyed yesterday)

Personally, I think in a general class of low-level learners – where conducting small-talk is obviously a struggle – one of best things teachers can do is foster natural, everyday small-talk. Jason Renshaw’s unplugged lesson video on his blog is a great example of this.

Let’s face it – and so many students say its important – small talk is one of the most common and socially vital ways we communicate with each other. If only for the first time we meet people, say at a party… These conversations contain our first (and lasting) impressions of someone and in a way, are fairly predictable (tiresome after a while even!)… so, where do you come from? What do you do? Have you ever been to…? Yes, I’ve been living here for…?

I remember early on in my teaching, being surprised that many ostensibly business English learners were often as interested in the art of coffe-break small talk as they were their presentations or conference calls. Why? Well, it’s such an integral part of getting to know who you’re really dealing with in business (sussing them out, as it were!).

So, in my view, small talk definitely needs to be fostered in practically any kind of class, regardless of what else the learners are focusing on.


9 05 2011
Scott Thornbury

Thanks David for the comment. I agree that most students hunger for the conversational gambits that comprise small talk, and that this is an important area of language to target – ‘phatic communication’, if you like.

Nevertheless I think it’s important to stress that the focus on conversation in the dogme classroom is not so much for the purposes of teaching conversation, per se, than to provide a matrix for the devleopment of overall language proficicency. If speaking skills improve as a consequence, so much the better.

Over a hundred years ago, Henry Sweet wrote that “conversation in a foreign language may be regarded from two very different points of view: (1) as an end in itself, and (2) as a means of learning the language and testing the pupil’s knowledge of it.” (Sweet, 1899, 1964, p. 210). It is this latter purpose – although with less emphasis on testing – that underlies the dogme dictum that classroom teaching should be ‘conversation driven’.

8 05 2011
Dennis Newson

Scott, you invite us to consider what “conversation” means. I think about this question often in social German life where I am always quoting: “What is truth? ” said Pilot, and did not wait for an answer.” So often, it seems to me, women here especially, ask you a question and then turn away to speak to someone else when you begin to answer. They are very bad, in other words, at turn-taking (and rude). And if more than four people are eating together at a table conversations typically split into two or three and no-one can hear easily what is going on. A crucial point, obviously, is that a “conversation” must be attended to by the participants and turn-taking rules must be adhered to.

Apart from that while I cannot define it helpfully, I can certainly distinguish between a conversation that is merely polite: “Do you come here often? How was the drive from Berlin?” and one in which at least one participant genuinely wants to report something, find out something, make a statement of genuine feeling or committed opinion. Speaking just to fill a gap won’t do. I am sure there is room here for exploring the difference between day-to-day conversation and therapeutic discourse. Anyone who has had therapy will know the profound difference between mere chat and effective conversation – typically a monologue with suggestive comments by the therapist. It is Halliday I think who writes about learning how to mean. Somehow, it seems to me, we need to school dialogic language learners to mean what they are saying – as well as saying what they mean. Any satisfactory definition of “conversation” I would suggest must include this element – the talk must be genuine, felt, meant, whether it is about food, computers or feelings.

For therapy and conversation see: Therapeutic Discourse Psychotherapy as Discourse William Labov and David Fanshel Academic Press 1977 ISBN 0 12 432050 3


9 05 2011
Scott Thornbury

Thanks Dennis, for drawing that interesting connection. Certainly, I think many teachers in a one-so-one situation have experienced a certain (uncomfortable?) morphing of their role as teacher into that of therapist. And there is a long tradition in humanistic education that encourages the use of the kinds of dialogic interventions – such as reflecting back the content of the speaker’s utterance – that were developed originally for the purposes of therapy. It’s worth noting that the Eliza program – which was developed to mediate human-to-machine therapy sessions – served as a model for a whole generation of ‘chatter bots’, some of which have been used in language teaching. Here’s a random site I just found by googling chatter bot: http://www.athena.blueinfos.com/

8 05 2011
Luke Meddings

Really enjoyed this Scott, thank you. Just a quick lateral connection.

Your lovely formulation: ‘This tension between the predictable and the unpredictable makes conversation – real conversation – an ideal medium for instruction’ chimed with something I was just reading about John Dewey’s desire to find a balance between excessive focus on subject-matter and exclusive focus on the child.

That strikes me also as a tension between the predictable (subject content) and the unpredictable (human beings). And facilitation through dialogic teaching and conversation as the best way to negotiate, explore and – maybe! – resolve it.

9 05 2011
Scott Thornbury

Thanks, Luke – slightly tangential, but your mention of Dewey reminded me that he also said that the experienced teacher “has acquired the requisite skill of doing two or three distinct things simultaneously — skill to see the room as a whole while hearing one individual in one class recite, keeping the program of the day and, yes, of the week and of the month in the fringe of consciousness while the work of the hour is in its centre” (The Relation of Theory to Practice in Education,1904, p.318).

“The work of the hour” — or indeed the work of the minute –is probably where the teacher deals with the unpredictable, while having one eye cocked on the broader agenda.

8 05 2011

Great post. Hope it’s OK that I share what happened in a workshop I gave yesterday in Bern Switzerland. The workshop was on improving fluency in Speaking and a few days beforehand I found out that over 70 people had signed up and so it would have to take place in the main plenary hall which luckily had chairs, plenty of space and bare walls!

I was pretty nervous before hand for a few reasons. Firstly I planned much of the 75 min workshop around participant conversation and emergent language – I wanted to introduce the ideas of a) listening to our own language to demonstrate that we have all the authentic scaffolds for our students and b) the skill of listening closely to other peoples conversations and making notes about the real language they use and c) what to actually “do” with the emergent language and language scaffolds. So the shape of a “lesson” so to speak.

Secondly, I was expecting maybe 15 people – certainly not 70+

Finally I knew a high percentage of the teachers at the workshop would be Swiss state school teachers who were perhaps not used to or familiar with such ways of working.

After splitting the room into 7 groups of 10 I assigned the roles “language detectives” to 2 of the group and “conversationalists” to the rest. I then gave 2 minutes of silent time and asked the conversationalists to think about 5 things about themselves they’d be happy to share with others. I asked the language detectives to get organized with pens and paper.

Then I asked language detectives to circulate and listen in on various conversations and make notes of useful language. I gave examples of the type of language this could be. Discourse features of speech, formulaic expressions, natural grammar structures like “I really like verb+ing” etc.

The conversationalists simply had to tell each other about themselves, ask questions and generally speak as normal. The conversation phase lasted about 15 mins with several change of partners, giving the language detectives the chance to jot down enough language.

Then I asked them to sit in their groups with a big piece of paper and decide together on the 6 most useful structures they felt would help students carry out a similar conversation and if poss focus on an area of language like discourse features or phrasal verbs – if they’d come up. This resulted in lots of conversation (which I joined in) Here are some examples of what groups came up with.

filler expressions.

“like she said”
“as she said”
“it’s also the fact that”
“things like that”

“that’s a shame”
“What about you?”


I pointed out how this is often the stuff that contributes to sounding fluent but is so difficult to teach. Then I introduced Swain’s notion of producing the push in order to aid fluency with the following points. Repeat tasks, have public performances, incorporate new language items into the task and increase the memory load. We talked about how once we had these language items – how could we best help students incorporate them into their own productive speech?

We then did some activities which might help with that. So we did a Speaking Bingo game using discourse features of turn taking. A disappearing dialogue with attention to aspects of rhythm, stress and intonation and finally a jazz chant using phrases to express agreement and disagreement.

The 3 aims of my workshop were to show:

Some practical ways of working directly with the language your students need.

Some practical ways to help ‘anchor’ those selected items into students memory.

Some linguistic theory which may influence our expectations and activities in class.

I opened the floor for Q+A for the last 10 mins I wanted to be sure that teachers of whom many were non natives felt comfortable using their own language in this way and we talked about how initially it was difficult to listen in on conversations for both meaning and structure.

I was pleased with the way it went, mainly that it showed that it is indeed very possible to work in a conversation driven way, with emergent language with large groups (of teachers at least)

8 05 2011
Anthony Gaughan

Heh, heh, sounds like you nicked my plan for one of my workshops at the upcoming SIG Day in Zug 😉

Seriously, sounds like a great session and illustrates nicely one of the recurring challenges I hear about unplugged approaches in the classroom – it is too hard to listen to genuinely interesting conversation and listen out for linguistic interest.

It is unfamiliar for most people, but – as your workshop experience shows – it is far from impossible!

8 05 2011

Really looking forward to yours at ZUG Anthony! And yes, the two main “difficulties” people had were getting so caught up in listening to the conversations that they found it tough to listen for linguistic items and also as there were 70 people – some people found it difficult to listen intensively with the obvious noise level.

9 05 2011
Scott Thornbury

Steph – another case of synchronicity – how did I know that you were about to do a workshop on speaking when I posted on conversation!? Your generous description of the workshop would serve as a model for any interactive, materials-light teacher training session. I can’t wait to try it!

9 05 2011
Anthony Gaughan

Heartily seconded!

10 05 2011
Karenne Sylvester


8 05 2011
Nick Cherkas

Thanks Scott – a great post. I particularly liked the idea from Ashton-Warner (“I harness the communication, since I can’t control it, and base my method on it”) suggesting that having a class which is difficult or impossible to “control” is an asset rather than an obstacle, as the learners have a collective attribute, or gift, of natural interaction which can be exploited.

I’ve always used and taught small talk to classes, especially to lower-level and business ones. Lower-level students need the speech acts to be able to interact, keep a conversation going and signal interest, whereas business students, as Mr Darkbloom pointed out, need to be able to interact using small talk just to communicate appropriately with colleagues and clients in most English-speaking business contexts. Indeed, this aspect of spoken language is frequently needed (and used) more by these learners than being able to use financial lexis and give presentations. Materials writers seem to agree, as most business coursebooks and resource books now feature sections on small talk.

Loved the workshop idea, steph. Is it available for sourced borrowing?

8 05 2011

Hi Nick, thanks and feel free, everything that I used in the workshop came from ideas/inspiration from others and I think every workshop of this nature ends up unique anyway as the mix of people are never the same.

8 05 2011
Ana lía

I wish I could have done your workshop , it sounds great ,I’ll put your ideas into practice in my school , thanks from Agentina !

9 05 2011
J.J. Sunset

A couple of too-much-Coke-at-night-induced dilemmas:

1. Would we define as a “dialog” an interaction among four speakers?

2. Might all the possible conversations have already taken place in the history of a language? A sort of finite virtual conversation corpus?

Sorry about the unnecessary headache…

9 05 2011
Scott Thornbury

In the book I wrote with Diana Slade on conversation we use the rather clumsy term ‘multilog’ to capture the fact that conversation often occurs between more than two speakers. As for your point about all conversations having occurred, this is, in essence, what the Russian scholar Bakhtin claimed — that everything we say is an echo of, or response to, something that has been said before, and carries with it the history of all previous conversations. “Our speech, that is, all our utterances are filled with others’ words, varying degrees of otherness and varying degrees of ‘our-own-ness’, varying degrees of awareness and detachment. These words of others carry with them their own expression, their own evaluation tone, which we assimilate, rework and reaccentuate” (Bakhtin, Speech Genres and Other Late Essays, 1986, p.89).

Not sure if that’s what you meant, but…

9 05 2011

everything we say is an echo of, or response to, something that has been said before, and carries with it the history of all previous conversations.

And if we talk into consideration the full range of environmental factors with a language event occurs, we can juxtapose that with the mind-bending notion that a word never means exactly the same thing twice 🙂

Nevertheless I think it’s important to stress that the focus on conversation in the dogme classroom is not so much for the purposes of teaching conversation, per se, than to provide a matrix for the devleopment of overall language proficicency.

Absolutely, and we may call this conversation with a capital C


9 05 2011
J.J. Sunset

Yes it is, Scott, thanks.

Varying degrees of otherness, history, predictibility…

12 05 2011

‘A favorite theory of mine—to wit, that no occurrence is sole and solitary, but is merely a repetition of a thing which has happened before, and perhaps often.’ ~ Mark Twain

9 05 2011
Mark Kulek

I would say that teachers teach grammar because it’s easier than teaching speaking. It can’t be because it’s fun. 

As for young learners, there are many teachers here in Japan who base their curriculum on phonics. Phonics is a powerful tool in an EFL setting, but shouldn’t we be giving kids a good start at speaking? My observation has been that a majority of teachers teaching kids are taking the easier road.  Ironically, the kids who come to my school who had a non-native teacher have strong speaking skills. Could it be that non-native teachers are not teaching phonics, but speaking skills?

Mark in Gifu

9 05 2011
Peter Hourdequin

Hello Scott,

Thanks as usual for a thought-provoking post. This is timely because I have just been thinking a lot about how to best structure (or not structure) a weekly speaking opportunity that I have for students (Japanese) in our university’s foreign language study center. In previous years, foreign teachers like myself have each been scheduled to be in the center for a 40 minute session once a week. Students could talk to us freely, ask questions, converse, whatever. The problem we had was that one student could easily dominate the instructor’s time, leaving more timid students without speaking opportunities. The proposed solution was, this year, to make the time more structured–adding a sign-up sheet, and a rule against participating in the same teacher’s time slot two weeks in a row. This has worked well so far in terms of distributing the talk opportunities, but it may have had a negative effect on “conversation.”
I also want conversation to be “emergent” and want to encourage that, but without a topic, and without some planning (on the students’ part), I feel this has been difficult (especially with lower level students). Students are shy and hesitant, and though they answer questions when asked, the whole drill can feel like a day at the bowling lanes: everyone takes a turn while the others watch and listen quietly (polite or nervous laughter takes the place of applause). But the bowling can be slow, and as a teacher I sometimes feel like I have to guide students to the lane, lift the ball, and place it in their hands. . . . pull the arm back.
Things have not actually been that difficult. But there is sometimes the feeling (despite the table being nice and) of everyone looking to me for conversational guidance or help, when the ideal of course is a lively, egalitarian discussion. Since this isn’t a class per se, and since my time is limited, I want to find a structure that puts more of the onus back on the students, and yet creates conditions for greater fluency. Having just completed a course with Marc Helgeson, the solution I was thinking about was planning/think time. That is, I have the feeling that one of the impediments to speaking for these students is that they have not had time in advance to think through a) what they might want to say about a given topic and b) what words / phrases they might use to say these ideas in English. Therefore, I am making a form for them to receive when they sign up which explains the importance of planning, and gives them some space to write/organize their thoughts in advance.

I know personally, for example, that before I pick up the phone to make a call to the office staff in Japanese, I do a similar kind of language planning exercise myself. I sometimes even jot down notes about what I will say, words that may come up, etc. I am fluent in Japanese in certain domains, in a sense, but when I have to use language in a domain where I am not accustomed to speaking, I always find planning useful. When I don’t plan, I always wish I had. So this is the theory (or, I should say, anectode) behind creating a structure that encourages planning on the students’ part.

And then I read your post . . .

Where do you (and others here) place planning within this conversation on conversation?

9 05 2011
Anthony Gaughan

I think that’s a great question, Peter. I also know from my own experience that planning conversations in an L2 (German, in my case) helps a lot, especially when:

a) the topic is new to me
b) the topic is emotionally charged for me

Both of these conditions place a premium on my being able to say what I want to say clearly and powerfully. But I also want to be able to focus on what my interlocutor is saying, so that I can accommodate that – and if I am making up what I want to say entirely on the fly, I have less brainpower left for that, it feels like.

And both of these conditions, I think, are often met in an effective classroom conversation – or they should be. (And note, by emotionally charged, I imagine joyful, amused, concerned, as much as anything else – discussing how people behave on stairs in central train stations during rush hour would have been an “emotionally charged” topic for me this morning, I can tell you – so the topic can also be general and mundane!)

So I’ve been encouraging learners in classes I teach to take half a minute or so to consider what they would like to contribute to an upcoming conversation, and to research or request required lexis in advance. Nothing new here (Scott and others put me on to the idea ages back) but it still impresses me how much this seems to act as a leveller in an otherwise mixed proficiency/confidence group of students.

So I find your idea of a tool to encourage your learners to make a plan of attack for a conversation very interesting and would be very interested to hear more about what they put down on paper and how they feel about its impact.

I appreciate that I haven’t actually answered your question, but hopefully this has some relevance, anyway 🙂

9 05 2011

Hi Peter. I think planning is a good idea, but with a group where you don’t have the same students each week, I think I’d be inclined to give a few minutes planning time in the course of the session, once you know what they’re interested in talking about that day.

For groups such as you describe, to give a bit of focus and structure, I find it very useful to have a conversation stimulus, or a series of stimuli, to get the conversation going. This could be a map of the world or their country, photos around a topic (see ELTpics :-)) or a series of questions, often around a topic but not necessarily. With the questions, introducing a very loose game-like quality works well, relaxes the students, makes it a bit more fun, as well as giving a bit of structure. For example, using a pack of cards, students pick a card and have to answer the corresponding question. (All groups I’ve used this with love finding out what the different cards and suits are called). Or, questions are cut up and turned upside down for students to take turns picking one, answering, asking others, etc. Even, board games with dice and counters, such as you can find on OneStopEnglish.com. With each question, allow/encourage the conversation to develop, asking what other students think, asking follow up questions etc, reformulating contributions and focussing on language as you go. When the conversation lulls, it’s the next student’s turn to pick a card/question or throw the dice. Some questions won’t generate much conversation at all, but others can produce very enlightening and interesting information and discussion, and provide useful speaking practice!

9 05 2011
Scott Thornbury

Hi Peter, Anthony, Dennis and Carol have already said more (and said it better) than I could, but I’ll just add this extract from my book on conversation (co-written with Diana Slade, CUP 2006):

One of the problems with providing opportunities for real conversation in the classroom is that, as Bannink (2002) points out, conversation resists planning: “Genuine conversational interactions cannot be the outcome of preplanned lesson agendas, they have to emerge – and so, by definition, cannot be planned” (p. 271).

Nevertheless, there is some evidence that “programmed” conversation is feasible. Ernst (1994), for example, describes the stage of a conversation class that she calls the “talking circle”:

The talking circle is a total group activity what generally takes place at the beginning of the 45-min conversational English class. Almost every day, teacher and students gather in the talking circle to share and discuss experiences, anecdotes, news, special events, introduce the weekly theme, and the like. Although the teacher might open the discussion by suggesting a general topic, the overriding assumption is that the talking circle provides a place and an audience for students to discuss anything of interest to them. (p. 299)

Ernst notes that these talking circles “provide rich opportunities to practice the L2 and to engage in direct and meaningful interaction” (p. 294). It may be that the best fluency practice that learners can have is when the pre-lesson chat is “ritualised” in this way, i.e. when it is incorporated into the lesson as a regular feature.

Ernst, G. (1994) “Talking Circle”: Conversation and negotiation in the ESL Classroom. TESOL Quarterly, 28, 293-322

9 05 2011
Scott Thornbury

I should add that there is an extremely lucid article about the experience of learning Chinese by means of a dialogic pedagogy, by Howard Vickers on his blog at: http://www.avatarlanguages.com/blog/pedagogical-conversation/#more-655

Howard recounts how the conversations with his Chinese teacher often revolved around objects and photos (that Howard, the student, supplied), and the advantage of these – apart from being a stimulus for the conversation – is that they offered a point of re-entry after a digression into grammar or vocabulary.

9 05 2011
Dennis Newson

Peter. My answer to your question is really also a question: “This is
what I have done in the past. How does this sound?”

With German University, nominally teachers- in- training students –
round about 30 at a time for regular weekly sessions – with the aim
of generating as much talk amongst them as possible, and gathering
examples of language to be worked on, and using the format of a task
(though I saw the talk as the point and not the task) I divided them
into groups of 4. Each group received just one copy of an assignment I
had written – only one copy so that one of the four had to read it out
and the other three had to request repetitions and explanations.

The assignments were tasks rather than “What do you think about….”

Amongst the most successful were:

1. Following the description of a flat on the outskirts of the town
the 4 had to draw up a list of regulations for living together,
including allocation of rooms – one without a window etc.
2. The flat sharers decided they needed to find a fifth person to
share and interviewed people from other groups.
3. The group were the editorial board of a student magazine and had
to decide on which poems received should be included/excluded and to
draft the letters of acceptance, rejection.

The students focussed on the task, but the real point was the language
that was generated.

I eavesdropped and noted “points arising” – language for exploitation

I have to admit that there was rarely time for dealing at all
adequately with the language arising and I got the distinct impression
that dealing with emergent language – though I did not call it that –
was my preoccupation and not the learners’. They appeared to be
satisfied with the liveliness and fun of the group negotiations.

At the beginning I used to end with a plenary session, as it were,
with one person from each group reporting to the others what had
happened. Compared with the group discussions, these reports were flat and unproductive.

At a later stage I took to making videos of the groups at work as a
better way of taking notes.

As I was doing this one day a student said something like: “This is
the way it should be.”

“And what am I supposed to do?” I asked, feeling almost redundant as
teacher turned camera man.

“You are just supposed to motivate us”, said the student.

In summary, my suggestion, choose small group discussion and
performance of tasks rather than individual preparation..
Conversations, surely, can only take place between a very small number
of people. And, wouldn’t you agree, conversation needs to be spontaneous. Prepared, rehearsed conversation is virtually a contradiction in terms. Dennis

9 05 2011
Anthony Gaughan

“As I was doing this one day a student said something like: “This is
the way it should be.”

“And what am I supposed to do?” I asked, feeling almost redundant as
teacher turned camera man.

“You are just supposed to motivate us”, said the student.”

Dennis, I love this! A “Paradebeispiel” for establishing the conditions 🙂

However, I am not so sure that preparation and rehearsal can’t have a role in a classroom geared to developing a language learner’s ability. It’s a contrivance, certainly, but one which we use in real life quite frequently (perhaps more often in transactional interaction, but if it has value there, why not elsewhere?)

Or are dogme and contrivance “virtually a contradiction in terms”? Hmm…(puts thinking cap on…)

9 05 2011
Mr Darkbloom

I got the distinct impression that dealing with emergent language […] was my preoccupation and not the learners’. They appeared to be satisfied with the liveliness and fun of the group negotiations

This is a very interesting point. When to interupt the flow? In classes that are chatty and lively and a good discussion is underway, I often feel less like a teacher and more like a party pooper, just calling for a little break to pick up on some language that was used. The learners were just happy to carry on chatting for maybe the entire lesson! This is great, right? Maybe we’ve got flow of the Csikszentmihalyi kind. Still, I feel some balance – and formal feedback – is in order…

Dear ELT agony aunts, what should I do?


9 05 2011

The powers-that-be in the institution I work at could do with reading this as they have a very strange definition of ‘conversation’ – my official job title is that of ‘English Conversation Teacher’ and yet the syllabus I’m handed at the start of each year covers working with story books, preparing students for the Cambridge Young Learner tests and developing writing skills… Why they can’t just call me a plain old ‘English Teacher’ I’ll never know 🙂

10 05 2011
Sue Lyon-Jones

Very interesting post Scott and immaculate timing as well, as I’m going to be back in the classroom in a few weeks time running an unplugged conversation course, and I was leafing through your “How to Teach Speaking” book just before I saw this 🙂

I hadn’t come across Ernst’s article before, but I often use conversation-circle type activities to start off speaking and listening lessons. Like Carol, I go in with some discussion prompts just in case we need them, and I would echo the points that people have made about small talk, and not just at lower levels, either – I’ve had advanced students in my classes recount stories about coming unstuck because they didn’t know the rules of social interaction for the situation they found themselves in.

I would say that register, tone, and body language also play a significant part as well, as these can make a big difference in terms of how learners are perceived by other people. For instance, fillers can be misconstrued as sarcasm by listeners if the tone of voice is not quite right.

10 05 2011
Scott Thornbury

Thanks, Sue, for that – and good luck with your class!

Of course another tried-and-tested format is the Counselling Language Learning one, which, although not real conversation (because it slows the whole process down) is good for allowing students time to formulate their utterances. Here is a description (for those who are not familiar with the technique, taken from the book that Sue was kind enough to mention):

A more formalised way of assisting performance is by means of a technique that derives from a teaching method called Community Language Learning (CLL). Instead of addressing the teacher directly, the learners sit in a circle and address each other, building up a conversation which, utterance by utterance, is recorded on tape. The teacher’s role is to act as a kind of language consultant, providing the language the learners need to express their intended meanings. At beginners level, this will involve the teacher translating the learners’ meanings. At higher levels, it may simply be a question of reformulating what the learner wants to say. Once each utterance has been ‘tidied up’ in this way, it is committed to tape. Here, for example, is a segment of a conversation that occurred between a small group of adults in a language class in Spain:

Student 1: Emma, where are you going tonight?
Student 2: Tonight I am going to have supper out.
Student 3: Where are you going to have supper?
Student 2: I don’t know. I am being taken out.
Student 4: Who are you going with?
Student 2: I am going with — with a guy, but he isn’t my boyfriend.
Student 1: And where is your boyfriend?
Student 2: Do you mean now?
Student 1: No, not now. Where will – erm he be this evening?
Student 2: He’s going to play water polo.
Student 1: Hmmm, water polo – very interesting! Is your boyfriend hunky?
Student 2: Yes, he is very hunky…

The above segment lasts less than a minute on tape but took around ten minutes to put together, each line having been tried out and rehearsed before being recorded. Once a sufficient amount of conversation has been recorded, it is played back and transcribed on to the board or an overhead transparency, and it is then available for reading aloud, for some kind of analysis, or for further refinements, such as the addition of discourse markers, backchannel devices, (e.g. really? uh-huh, etc.) and so on.

10 05 2011
Mark Kulek

I would like to have more student led topics and conversation in my classroom. Creating a classroom culture where it is the norm. I use lexical chunks for prompts, question cards where the ss ask and move, picture prompts, etc. But it’s all my doing. I mean, I’m creating the topics. I need ideas to get students to decide the topics for the class and then to act on them.

Just last week, I made an activity sheet for students to introduce their own topic for discussion. On the sheet there are two boxes. The first box is for the student to state his or her topic and to draw a picture that represents their topic. The second box is for their notes. The other students have a sheet to take notes and to write down questions to ask the Topic Master. However, Scott mentioned that conversation is not real if planned. I think I know what is meant by that, but not clearly.

I teach small groups at a conversation school. Of course, I want real conversation happening. I want my students leading the way. Like that which was stated earlier, I want to be just the camera guy.

Mark in Gifu

10 05 2011
Karenne Sylvester

If I weren’t on a blog break I’d be very tempted to write a post entitled

“The importance of being mundane..”


“why non-intelligent conversations are the most important conversations of all when it comes to teaching language learners…”

…I have to admit to being more than a wee bit squirmish about this notion of declaring what a conversation is… what instructional conversation is, zwiglididliwhat…

but then I am so not and never have been fond of academics playing around with what they have theorized life to be (often stating the bleeding obvious while missing the real point)…instead of indulging in the sheer pleasures of living in life…


1. For me, small talk is the most important because it constitutes a difficult social skill that is often least practised among learners. But small talk is too mundane to base a whole class around, hence the need for materials.

Yes – ever listened to people sitting at a bar, two women chatting in the Sauna? Husband and wife nattering on about those wee miracles they somehow, magically, gave birth to?

No materials necessary.

Small talk, my friends, is what makes the world tick and it can be and is done for hours and hours and hours between those who know each other as it is in the very mundanity of the average conversation that repetition of the most frequently used words in the English language occur.

Ideally a language classroom should provide the tools and create the environment to engage in what the whole world engages in (not the edge of the edge) – not entirely sure where the idea has arisen that mindless-mindnumbing small-talk cannot be done in an entire class nor even in fact an entire semester!

And who is it that determines the mundane?

The every-single-day conversations are always boring to someone and fascinating to someone else…. of an IT developer with his team; the dull and dreary review of the balance sheets of the company being audited with its CTO… the mindnumbing dullness of a director convincing his actors to act?

Life itself is small talk and it very much does not matter that the majority of this discourse is senseless and without any real function (to whom? to whom? I cry)… the real function of daily interaction is to serve humanity, to create community, to say ..you know what, this is my life… not this is what I theorize life to be – and for the listeners, the co-creators and sharer of those brainless tidbits to say, I know -hey, this is my life also and we are okay.

I suspect the problem is, therefore, more that the teacher has confused his/her role in the classroom as being “master-of-the-universe-of-the-English-language” or “most wisest one” rather than a “facilitator” and “open-doorer” of knowledge).

The teacher (who while reading this, who just scoffed at my making up a new words)… in the ideal, nirvanaesque world we should, minute-by-minute, spout knowledge is probably what led us to coursebooks in the first place, 😉 😀

It’s because some teachers can’t be bothered to stoop low, to indulge in the repetitive nonsense that makes up conversation that they completely miss the point that these nonsense structures hold up the beams to all the rest of language.

2. If learning opportunities are based solely upon this fairly restricted register, it’s unlikely that most students will find their communicative needs are satisfied – especially if these needs include more formal registers, such as academic or technical writing.

Um, no… see above. The formal is supported by the informal. Don’t throw out baby with bath water.

3. Conversation is predictable, because one turn follows from the other.
– not always, especially not in groups where A types are sitting.

4. We therefore have to listen very carefully… and we also have to take great care in constructing our contributions so that we can be understood.
– interesting thought but not life realistic :D.

5. The concept itself contains a paradox: “Instruction” and “conversation” appear contrary, the one implying authority and planning, the other equality and responsiveness. The task of teaching is to resolve this paradox. To most truly teach, one must converse; to truly converse is to teach.”
-ya what?? I get this and I don’t this. I agree and I don’t agree… it’s a bit weak, a bit… not quite right. Apples are red and green. If the fruit is red or green, it is an apple. Nope. … – to truly teach, one must be able to converse and be able to allow the students to converse without directing the flow and objective of the conversation… to truly be able to allow conversation to flow while staying conscious to the intricacies of the language and being able to provide feedback is teaching language..”

My overall thoughts on reading this post are basically…

Lose control and all this academic investigation and allow conversation to simply occur – scaffold, support, correct – helper, facilitator, aid, guide… not all this controlling master of the universe stuff… this sort of thinking process leads us on to unuseable, unnecessary, convoluted flow-charts -taking you to far away from what’s really happening in the room, like. Language is flow, untameable, unknowable…delicious.

I just keep coming back to the thoughts that the best teacher is he who not a teacher.


12 05 2011

Just been re-reading my Krashen, who says that one valuable function of the teacher could be teaching “conversational-competence” because it leads to more comprehensible input.
It also seems to me that conversation in the class allows for acceptance into social groups etc – Krashen also mentions this social and psychological distance in terms of affective filters. Bonny Norton too talks of less successful migrant language learners not feeling ‘accepted’ as part of the society.
For me this is where conversation in the classroom comes to the fore, as acceptance of a person and a way of negotiating entry into the target society/peer group via someone who is there to smooth the process – the teacher.

26 05 2011

Here is a quotation I bumped into yesterday morning, while I was revising my English literature. It is from Laurence Sterne’s Tristram Shandy, and I unconditionally subscribe:

“writing, when properly managed, is but a different name for conversation”.

(The Life and Opinions of Tristram Shandy, Vol.1, Ch. XI)

I’d say this sentence economically epitomizes all text and discourse analysis, even before linguistics was born. And indeed it broadly applies to all kinds of writing, including blogs!

26 05 2011
Anthony Gaughan

Sterne (good Irishman that he was) knew a bit about the blarney – as your quote shows!

I must admit, it sent me back to my Shandy, and I came across this gem:

“I do not know what envy is: for never do I hit upon any invention or device which tendeth to the furtherance of good writing, but I instantly make it public; willing that all mankind should write as well as myself.
– Which they certainly will, when they think as little”

Sterne, L. (1759-67), The Life and Opinions of Tristram Shandy, Gentleman, Vol. 9, Chap. 14

27 04 2012
Felipe Braz

I could notice that Sylvia Ashton Warner was mentioned in your artice. I would really love to find a film about her life, shot in the 1980s. Does anybody know where I can find it. I’m in Brazil and I cant’ find it here.
Thanks a bunch!

27 04 2012
Scott Thornbury

Felipe, I did once see a documentary about Ashton Warner, available for private viewing in the film library of the New Plymouth (NZ) art gallery, but I have no idea who made it (possibly the New Zealand Broadcasting Corporation) or if it is currently available. It was a good 15 or 20 years old, but it did include some footage of the lady herself. That’s the best I can do, I’m afraid. Google?

28 04 2012

Google comes up with severaal hits for Sylvia Ashton Warner, including:



29 04 2012

You can find things on her at the NZ Film Archive here:


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 )

Connecting to %s

%d bloggers like this: