Category:Dialogic teaching: Difference between revisions

Introduction to Dialogue

This resource is licenced under an Open Government Licence (OGL).

This resource is part of a larger document (QCA, (2003), New Perspectives on Spoken English in the Classroom), downloadable from http://dera.ioe.ac.uk/6062/

Purposes and characteristics of whole-class dialogue

Tony Edwards, Open University

Introduction: some reflections on the English context

The difficulties of defining dialogue begin with the question of how many can take part before it turns into something else. In ordinary conversation, the managing of turns is a shared responsibility, and competition for 'having one's say' in groups larger than, for example, half a dozen makes a diversion into parallel conversations very likely. Most classroom talk, in contrast, involves a centralised communication system. Teachers direct the talk by doing most of it themselves, combining lengthy exposition with many questions, allocating the right or obligation to answer those questions and evaluating the answers. The transmission of knowledge creates very unequal communicative rights to those who 'know' and those who do not. This is why the sequence of (teacher) initiation - (pupil) response - (teacher) evaluation has emerged from so many research studies as the 'essential teaching exchange'¹ In whole-class questioning, it carries risks that a single right answer will be taken as representing a class-wide understanding and a single wrong answer as a common failure to get the point.

A great deal of teaching is unavoidably a passing on of information and skills. However, it benefits from being complemented by classroom talk that is organised very differently for specific curriculum purposes. It is this 'something else' to which whole- class dialogue contributes, provided it goes well beyond those class discussions which involve few departures from teacher direction and little reduction in teacher talk.² It replaces the usual hunt for answers which the teacher already knows into collaborative searches for solutions or understanding. It blurs those sharp boundaries around school knowledge that largely exclude reference to what pupils know unless they have already been taught it, or at least screens such references for educational relevance. It can provide more opportunities for learners to talk their way into understanding rather than receiving, more and less effectively an already defined version of what they are now supposed to know.³ Dialogue differs from most classroom discussion in so far as the talk is exploratory, that is teacher and pupils see the possibility of conclusions unexpected, and certainly unplanned, when the talk began.

If the potential educational advantages are substantial, why is whole-class dialogue apparently uncommon? It may well be less unusual than classroom research indicates because orderly teacher-centred talk was, until quite recently, so much easier to record audibly and then present in play-script form unpunctuated by gaps and guesses. Robin Alexander and his colleagues show a technically advanced and imaginative capacity to capture many learner voices in classrooms which were not ordered in traditional ways.⁴ But there are powerful managerial and educational reasons why departures from teacher-directed exposition and questioning are unusual.

An absence of untoward noise is still commonly taken as evidence of good classroom control. Opening out the interaction risks disorder. For example, open questions elicit unpredictable responses which are difficult to assess. It is managerially safer to ask the kinds of questions which entitle the teacher (who knows the answer) to respond immediately, thereby exercising the right to speak every other turn, or at least to take a very high share of turns. There has also been a long, well-publicised, war of attrition against progressive teaching that has caricatured it as a laissez-faire indulging of pupils' uninformed opinions. The national curriculum, literacy and numeracy programmes and the high-stakes testing of their outcomes have tended to strengthen the framing of classroom communication. With a great deal to get through, the pace of transmission is likely to be fast. This privileges the teacher's talk, producing not only a great deal of exposition but also a predominance of questions to which the answers are likely to be short and readily 'marked'.

The extent to which whole-class dialogue departs from such normal practice means that it makes unusual pedagogic demands on teachers and learners. Perhaps first among its demands on teachers is that they are willing not to do what they may often take for granted for so much of the time. For example, teachers ask so many questions that innumerable researchers have counted them, timed them, mapped their distribution, categorised them and tried to measure their cognitive level. The pressures to evaluate the consequent answers are so pervasive that there is much to be gained from sometimes replacing them with statements that invite rejoinders, elaboration or disagreement or that even admit perplexity. Dialogue is certainly unlikely to follow either closed questions or those half- or 'pseudo-open' questions which are progressively closed down in ways which make it obvious that an answer is already there for pupils to hunt down. Teachers are extraordinarily skilled not only at redirecting questions in the interests of 'getting on', but also at translating answers into something directly helpful to the lesson's progress that pupils no longer recognise as their own. These are skills to be temporarily put aside. Teachers also need the nerve to tolerate pauses between turns without feeling that any silence is an awkward silence, and that the responsibility for ending it is theirs. A pause at strategic points in the discussion of no more than five seconds (longer than most pauses in whole-class interaction) may be enough to draw in another pupil contribution or encourage the previous speaker to elaborate on what was said. Intervening to answer questions or provide information useful for getting past a sticking-point requires not only the self-restraint not to take the discussion over, but also the willingness to listen to what is being said rather than merely listening for whatever best promotes the teacher's pedagogic agenda.

Corresponding demands are made on pupils. They are usually well practised in listening for clues in how the teacher introduces a question and responds to initial answers. Experience may well have taught them that the clues are often so prolific that even a wild guess will lead the teacher to answer the question for them. They may have much less experience of listening to one another. Indeed, the distance between whole-class dialogue and customary classroom talk is wide enough to make explicit rules of engagement helpful so that the differences are seen as deliberate departures. Doing so applies the notion of a distinctive 'speech event' to whole-class dialogue, recognised by the participants as having its own way of contributing appropriately. Notable examples of recommending clear procedural rules designed largely to curb teachers' usual directing role are the Nuffield Humanities Project and the National Oracy Project, both vulnerable to ill-informed attacks as a progressive descent into 'anything goes'.

Criteria for recognising dialogic talk

Having emphasised the distinctiveness of whole-class dialogue, I end with some criteria for recognising it when it happens. These are offered cautiously, because how classroom talk is used to organise relationships and meanings is too skilful and complex to be treated as a transparent medium. Most obviously, participation is shared around, not monopolised by the teacher and a few confident, willing pupils. Some pupil contributions may be lengthy and most are followed by another pupil. Teacher interventions may well be decisive pedagogically, but are likely to be infrequent and their placing in the interaction unpredictable. Getting and keeping 'the floor', and ensuring that interruptions are constructive not disruptive, are managed as shared responsibilities. Any skewing of communication so that some pupils or a group of pupils remain persistently silent is recognised as a problem and confronted openly. Such normal teacher tasks as clarifying where a discussion has got to or summarising what has actually been learned from it are also shared around. Teacher and pupils take explicit account of what others have said, so that their speech is responsive as well as expressive. Thinking time can be taken without the speaker's turn being lost and re-allocated. Pauses are more frequent, and often longer, than is possible from the driven momentum of most classroom talk. Thinking aloud is encouraged, that is talking one's way into meaning rather than remaining silent until some sort of answer has been formulated.

Conclusion

There is no implication in that brief profile that a consensual conclusion should eventually be reached. Indeed, the sharpest contrast between whole-class question-and- answer, and whole-class dialogue, is that different and even competing ideas can be kept in play without being subjected to one participant's authoritative arbitration. Making good educational use of it raises an obvious question about what to do if the dialogue appears to the teacher to be achieving nothing other than confusion, or is threatening a conclusion (citizenship lessons come to mind) that the teacher is likely to feel an educational or civic obligation to challenge. Contrary to hostile caricatures, whole-class dialogue does not demand that all such responsibility be discarded. It does embody more problem-posing and less solution-giving; a view of learning as enquiry as well as induction into what is already known, and as a social, truly interactive process; and a clear recognition of the educational value of drawing attention from time to time to the grounds for opinions and conclusions, and to how new knowledge can be constructed.

References

1 Edwards, A and Westgate, D, Investigating classroom talk, London, Falmer Press, 1994, pages 44 to 54 and 124 to 133.

2 Dillon, J, Using discussion in classrooms, Buckingham, Open University Press, 1994.

3 Barnes, D and Todd, F, Communication and learning revisited: making meaning through talk, Portsmouth NH, Boynton Cook, 1995; and Mercer, N, Wegerif, R and Dawes, L, Children's talk and the development of reasoning in classrooms, British educational research journal, 25, 1999, pages 95 to 111.

4 See, for example, Alexander, R, Culture and pedagogy: international comparisons in primary education, Oxford, Blackwell, 2000, pages 450 to 461.

Typical Classroom Dialogue

This resource is licenced under an Open Government Licence (OGL).

This resource is part of a larger document (QCA, (2003), New Perspectives on Spoken English in the Classroom), downloadable from http://dera.ioe.ac.uk/6062/

The educational value of 'dialogic talk' in 'whole-class dialogue'

Neil Mercer, Open University

Introduction

In this paper, I will discuss the nature and educational significance of the kind of interaction called 'dialogic talk' and its use in 'whole-class dialogue'. To do so, I must begin by saying something more general about patterns of classroom interaction and teachers' use of questions.

Research in many countries has shown that in whole-class sessions teachers tend to talk much more than their pupils. They also ask the great majority of questions. Moreover, most of their questions will form the first part of an exchange between a teacher and pupil known as an initiation-response-feedback (IRF) exchange.¹ These IRF exchanges give classroom talk its distinctive and familiar form.

There has been much debate amongst educational researchers over the years about the functions and value of this characteristic form of classroom interaction.² In this debate, it was at one time very common to find researchers criticising teachers for talking and questioning too much. However, most classroom researchers would probably now agree that such judgements were too simplistic. One reason is that critics did not properly acknowledge teachers' professional responsibility for directing and assessing pupils' learning of a curriculum, and the ways that they must rely on questions and other prompts to do so. Secondly, they tended to assume that all IRF exchanges were performing the same communicative function. Through the work of sociolinguists, linguistic philosophers and psychologists, we now know that it is dangerous to assume that forms of language have any direct and necessary relation to their functions. By this I mean that, for example, we cannot assume that when someone poses a question to another person, they will always be 'doing the same thing'. At an everyday level, we all appreciate this very well. In a personal conversation we are likely to perceive the question 'Do you really think that you can talk to me like that?' as carrying a very different kind of message from 'Do you want a cup of tea?'. What is more, even an apparently simple and direct question may take on special meanings within a particular setting or relationship.

In the classroom, teachers' questions can have a range of different communicative functions. For example, they can be used to test pupils' factual knowledge or understanding ('Can anyone tell me the capital city of Argentina?'), to manage classroom activity ('Are you all ready now to put your pencils down and listen?') and to find out more about what pupils are doing ('Why did you decide to have just three characters in your play?').

Even the above analysis is an oversimplification, because a question can have more than one function (for example, to find out what pupils are doing and to make them think about it) and because it takes on special meanings in the life of a particular class (have they studied Argentina already or are they about to begin?). But the key point is that the distinction between form and function is important for analysing and evaluating teacher-pupil dialogue.

'Dialogic talk' and 'whole-class dialogue'

Through his comparative research in the primary school classrooms of five countries, Robin Alexander³ has shown that if we look beneath the superficial similarity of talk in classrooms the world over, we will find teachers organising the communicative process of teaching and learning in very different ways. In most of the classrooms he observed, teachers talked more than the pupils; but the balance and nature of contributions varied considerably, both between countries and between classrooms. One of the reasons for this variation was that in some classrooms a teacher's questions (or other prompts) would elicit only brief responses from pupils, while in others they often generated much more extended and reflective talk. The concept of 'dialogic talk' emerged from these observations as a way of describing a particularly effective type of classroom interaction. 'Dialogic talk' is that in which both teachers and pupils make substantial and significant contributions and through which pupils' thinking on a given idea or theme is helped to move forward. It may be used when teachers are interacting with groups or with whole classes.

I can illustrate my understanding of the function of this kind of talk through the example below. It was recorded in an English primary school by Open University researcher Manuel Fernandez, who is investigating the role of computers in children's literacy development. In this extract, the teacher is talking with some members of her year 5/6 class about their current activity; they are communicating by e-mail with members of a class in another local school about the shared curriculum topic 'How to have a healthy lifestyle'.

In the first part of the example, the teacher uses prompts to find out what the children have done. The first actual question comes from a child, on a point of spelling accuracy. When the teacher then begins to question the children, it is not to assess their spelling; it is to elicit their reasons for what they are writing to the children in the other school. She provides feedback on their answers ('OK. Excellent.'), so the episode has some features of the familiar IRF structure; but the teacher's questioning is used to encourage the pupils to perceive more clearly the nature of their task. She then picks up on what they have said to guide the next part of their activity, by suggesting that it will be useful to share their reasoning with their audience (and modelling how they might do it: 'We are sending you this information because ...'). She is using this interaction to build the knowledge foundations for the next stage of their activity - talking with them to guide their thinking forward. So we have here talk in which pupils make substantial and thoughtful contributions, and in which the teacher does not merely test understanding, but guides its development. What is more, all the pupils present are exposed to this reasoned discussion. This may not be 'whole-class dialogue', because the discussion is not shared with all members of the class; but it certainly seems to qualify as 'dialogic talk'.

We can consider further what 'dialogic talk' offers, from an educational point of view. One of the prime goals of education is to enable children to become more adept at using language, to express their thoughts and to engage with others in joint intellectual activity (their communication skills). A second important goal is to advance children's individual capacity for productive, rational and reflective thinking (their thinking skills). Dialogic talk can help achieve both these goals. The work of the Russian psychologist Vygotsky is relevant for understanding why this is so.⁴ He suggested that using language to communicate helps us learn ways to think. As he put it, what children gain from their 'intermental' experience (communication between minds through social interaction) shapes their 'intramental' activity (the ways they think as individuals). What is more, he suggested that some of the most important influences on the development of thinking will come from the interaction between a learner and more knowledgeable, supportive members of their community.

Although developed over half a century ago, Vygotsky's intriguing ideas have only really been put to the test in recent years. Now research has confirmed the validity of some of his claims about the link between language use and the learning of ways of thinking. Research has shown that teachers' modelling of ways of asking questions, offering explanations and providing reasons can have a significant and positive effect on how children use language in problem-solving tasks.⁵ Research by myself and colleagues has shown that a programme of carefully designed teacher-led and group- based activities enables children not only to become better at talking and working together but also at solving problems alone.⁶ The group-based activities of this programme are very important; but equally important is the kind of dialogue a teacher uses in whole-class plenaries and group monitoring. It is no coincidence that the teacher in the example above has been involved in this programme. And this brings us back to 'dialogic talk'.

Conclusion

For children to become more able in using language as a tool for both solitary and collective thinking, they need involvement in thoughtful and reasoned dialogue, inwhich conversational partners 'model' useful language strategies and in which they can practise using language to reason, reflect, enquire and explain their thinking to others. By using questions to draw out children's reasons for their views or actions, teachers can help them not only to reflect on their reasoning but also to see how and why to seek reasons from others. By seeking and comparing different points of view, a teacher can help those views to be shared and help children see how to use language to compare, debate and perhaps reconcile different perspectives. Providing only brief factual answers to IRF exchanges will not give children suitable opportunities for practice, whereas being drawn into more extended explanations and discussions of problems or topics will. This is the valuable kind of educational experience that 'dialogic talk' and 'whole-class dialogue' can offer.

References'

1 Sinclair, J and Coulthard, M, Towards an analysis of discourse: the English used by teachers and pupils, London, Oxford University Press, 1975.

2 Norman, K (ed) Thinking voices: the work of the National Oracy Project, London, Hodder and Stoughton, 1992; Edwards, AD and Westgate, DPG, Investigating classroom talk, second edition, London, Falmer Press, 1994; and Wells, G, Dialogic inquiry: towards a sociocultural practice and theory of education, Cambridge, Cambridge University Press, 1999.

3 Alexander, R, Culture and pedagogy: international comparisons in primary education, Oxford, Blackwell, 2000.

4 Vygotsky, LS, Mind in society: the development of higher psychological processes, Cambridge MA, Harvard University Press, 1978.

5 Brown, A and Palincsar, AS, 'Guided, co-operative learning and individual knowledge acquisition', in L Resnick (ed), Knowing, learning and instruction, New York, Lawrence Erlbaum, 1989; and Rojas-Drummond, S, 'Guided participation, discourse and the construction of knowledge in Mexican classrooms', in H Cowie and D van der Aalsvoort (eds), Social interaction in learning and instruction: the meaning of discourse for the construction of knowledge, Oxford, Elsevier, 2000.

6 Mercer, N, Wegerif, R and Dawes, L, Children's talk and the development of reasoning in the classroom, British educational research journal, 25(1), 1999, pages 95 to 111.

Dialogic Talk

This resource is licenced under an Open Government Licence (OGL).

This resource is part of a larger document (QCA, (2003), New Perspectives on Spoken English in the Classroom), downloadable from http://dera.ioe.ac.uk/6062/

The educational value of 'dialogic talk' in 'whole-class dialogue'

Neil Mercer, Open University

Introduction

In this paper, I will discuss the nature and educational significance of the kind of interaction called 'dialogic talk' and its use in 'whole-class dialogue'. To do so, I must begin by saying something more general about patterns of classroom interaction and teachers' use of questions.

Research in many countries has shown that in whole-class sessions teachers tend to talk much more than their pupils. They also ask the great majority of questions. Moreover, most of their questions will form the first part of an exchange between a teacher and pupil known as an initiation-response-feedback (IRF) exchange.¹ These IRF exchanges give classroom talk its distinctive and familiar form.

There has been much debate amongst educational researchers over the years about the functions and value of this characteristic form of classroom interaction.² In this debate, it was at one time very common to find researchers criticising teachers for talking and questioning too much. However, most classroom researchers would probably now agree that such judgements were too simplistic. One reason is that critics did not properly acknowledge teachers' professional responsibility for directing and assessing pupils' learning of a curriculum, and the ways that they must rely on questions and other prompts to do so. Secondly, they tended to assume that all IRF exchanges were performing the same communicative function. Through the work of sociolinguists, linguistic philosophers and psychologists, we now know that it is dangerous to assume that forms of language have any direct and necessary relation to their functions. By this I mean that, for example, we cannot assume that when someone poses a question to another person, they will always be 'doing the same thing'. At an everyday level, we all appreciate this very well. In a personal conversation we are likely to perceive the question 'Do you really think that you can talk to me like that?' as carrying a very different kind of message from 'Do you want a cup of tea?'. What is more, even an apparently simple and direct question may take on special meanings within a particular setting or relationship.

In the classroom, teachers' questions can have a range of different communicative functions. For example, they can be used to test pupils' factual knowledge or understanding ('Can anyone tell me the capital city of Argentina?'), to manage classroom activity ('Are you all ready now to put your pencils down and listen?') and to find out more about what pupils are doing ('Why did you decide to have just three characters in your play?').

Even the above analysis is an oversimplification, because a question can have more than one function (for example, to find out what pupils are doing and to make them think about it) and because it takes on special meanings in the life of a particular class (have they studied Argentina already or are they about to begin?). But the key point is that the distinction between form and function is important for analysing and evaluating teacher-pupil dialogue.

'Dialogic talk' and 'whole-class dialogue'

Through his comparative research in the primary school classrooms of five countries, Robin Alexander³ has shown that if we look beneath the superficial similarity of talk in classrooms the world over, we will find teachers organising the communicative process of teaching and learning in very different ways. In most of the classrooms he observed, teachers talked more than the pupils; but the balance and nature of contributions varied considerably, both between countries and between classrooms. One of the reasons for this variation was that in some classrooms a teacher's questions (or other prompts) would elicit only brief responses from pupils, while in others they often generated much more extended and reflective talk. The concept of 'dialogic talk' emerged from these observations as a way of describing a particularly effective type of classroom interaction. 'Dialogic talk' is that in which both teachers and pupils make substantial and significant contributions and through which pupils' thinking on a given idea or theme is helped to move forward. It may be used when teachers are interacting with groups or with whole classes.

I can illustrate my understanding of the function of this kind of talk through the example below. It was recorded in an English primary school by Open University researcher Manuel Fernandez, who is investigating the role of computers in children's literacy development. In this extract, the teacher is talking with some members of her year 5/6 class about their current activity; they are communicating by e-mail with members of a class in another local school about the shared curriculum topic 'How to have a healthy lifestyle'.

In the first part of the example, the teacher uses prompts to find out what the children have done. The first actual question comes from a child, on a point of spelling accuracy. When the teacher then begins to question the children, it is not to assess their spelling; it is to elicit their reasons for what they are writing to the children in the other school. She provides feedback on their answers ('OK. Excellent.'), so the episode has some features of the familiar IRF structure; but the teacher's questioning is used to encourage the pupils to perceive more clearly the nature of their task. She then picks up on what they have said to guide the next part of their activity, by suggesting that it will be useful to share their reasoning with their audience (and modelling how they might do it: 'We are sending you this information because ...'). She is using this interaction to build the knowledge foundations for the next stage of their activity - talking with them to guide their thinking forward. So we have here talk in which pupils make substantial and thoughtful contributions, and in which the teacher does not merely test understanding, but guides its development. What is more, all the pupils present are exposed to this reasoned discussion. This may not be 'whole-class dialogue', because the discussion is not shared with all members of the class; but it certainly seems to qualify as 'dialogic talk'.

We can consider further what 'dialogic talk' offers, from an educational point of view. One of the prime goals of education is to enable children to become more adept at using language, to express their thoughts and to engage with others in joint intellectual activity (their communication skills). A second important goal is to advance children's individual capacity for productive, rational and reflective thinking (their thinking skills). Dialogic talk can help achieve both these goals. The work of the Russian psychologist Vygotsky is relevant for understanding why this is so.⁴ He suggested that using language to communicate helps us learn ways to think. As he put it, what children gain from their 'intermental' experience (communication between minds through social interaction) shapes their 'intramental' activity (the ways they think as individuals). What is more, he suggested that some of the most important influences on the development of thinking will come from the interaction between a learner and more knowledgeable, supportive members of their community.

Although developed over half a century ago, Vygotsky's intriguing ideas have only really been put to the test in recent years. Now research has confirmed the validity of some of his claims about the link between language use and the learning of ways of thinking. Research has shown that teachers' modelling of ways of asking questions, offering explanations and providing reasons can have a significant and positive effect on how children use language in problem-solving tasks.⁵ Research by myself and colleagues has shown that a programme of carefully designed teacher-led and group- based activities enables children not only to become better at talking and working together but also at solving problems alone.⁶ The group-based activities of this programme are very important; but equally important is the kind of dialogue a teacher uses in whole-class plenaries and group monitoring. It is no coincidence that the teacher in the example above has been involved in this programme. And this brings us back to 'dialogic talk'.

Conclusion

For children to become more able in using language as a tool for both solitary and collective thinking, they need involvement in thoughtful and reasoned dialogue, inwhich conversational partners 'model' useful language strategies and in which they can practise using language to reason, reflect, enquire and explain their thinking to others. By using questions to draw out children's reasons for their views or actions, teachers can help them not only to reflect on their reasoning but also to see how and why to seek reasons from others. By seeking and comparing different points of view, a teacher can help those views to be shared and help children see how to use language to compare, debate and perhaps reconcile different perspectives. Providing only brief factual answers to IRF exchanges will not give children suitable opportunities for practice, whereas being drawn into more extended explanations and discussions of problems or topics will. This is the valuable kind of educational experience that 'dialogic talk' and 'whole-class dialogue' can offer.

References'

1 Sinclair, J and Coulthard, M, Towards an analysis of discourse: the English used by teachers and pupils, London, Oxford University Press, 1975.

2 Norman, K (ed) Thinking voices: the work of the National Oracy Project, London, Hodder and Stoughton, 1992; Edwards, AD and Westgate, DPG, Investigating classroom talk, second edition, London, Falmer Press, 1994; and Wells, G, Dialogic inquiry: towards a sociocultural practice and theory of education, Cambridge, Cambridge University Press, 1999.

3 Alexander, R, Culture and pedagogy: international comparisons in primary education, Oxford, Blackwell, 2000.

4 Vygotsky, LS, Mind in society: the development of higher psychological processes, Cambridge MA, Harvard University Press, 1978.

5 Brown, A and Palincsar, AS, 'Guided, co-operative learning and individual knowledge acquisition', in L Resnick (ed), Knowing, learning and instruction, New York, Lawrence Erlbaum, 1989; and Rojas-Drummond, S, 'Guided participation, discourse and the construction of knowledge in Mexican classrooms', in H Cowie and D van der Aalsvoort (eds), Social interaction in learning and instruction: the meaning of discourse for the construction of knowledge, Oxford, Elsevier, 2000.

6 Mercer, N, Wegerif, R and Dawes, L, Children's talk and the development of reasoning in the classroom, British educational research journal, 25(1), 1999, pages 95 to 111.

Why Dialogic Talk?

This resource is licenced under an Open Government Licence (OGL).

This resource is part of a larger document (QCA, (2003), New Perspectives on Spoken English in the Classroom), downloadable from http://dera.ioe.ac.uk/6062/

The educational value of 'dialogic talk' in 'whole-class dialogue'

Neil Mercer, Open University

Introduction

In this paper, I will discuss the nature and educational significance of the kind of interaction called 'dialogic talk' and its use in 'whole-class dialogue'. To do so, I must begin by saying something more general about patterns of classroom interaction and teachers' use of questions.

Research in many countries has shown that in whole-class sessions teachers tend to talk much more than their pupils. They also ask the great majority of questions. Moreover, most of their questions will form the first part of an exchange between a teacher and pupil known as an initiation-response-feedback (IRF) exchange.¹ These IRF exchanges give classroom talk its distinctive and familiar form.

There has been much debate amongst educational researchers over the years about the functions and value of this characteristic form of classroom interaction.² In this debate, it was at one time very common to find researchers criticising teachers for talking and questioning too much. However, most classroom researchers would probably now agree that such judgements were too simplistic. One reason is that critics did not properly acknowledge teachers' professional responsibility for directing and assessing pupils' learning of a curriculum, and the ways that they must rely on questions and other prompts to do so. Secondly, they tended to assume that all IRF exchanges were performing the same communicative function. Through the work of sociolinguists, linguistic philosophers and psychologists, we now know that it is dangerous to assume that forms of language have any direct and necessary relation to their functions. By this I mean that, for example, we cannot assume that when someone poses a question to another person, they will always be 'doing the same thing'. At an everyday level, we all appreciate this very well. In a personal conversation we are likely to perceive the question 'Do you really think that you can talk to me like that?' as carrying a very different kind of message from 'Do you want a cup of tea?'. What is more, even an apparently simple and direct question may take on special meanings within a particular setting or relationship.

In the classroom, teachers' questions can have a range of different communicative functions. For example, they can be used to test pupils' factual knowledge or understanding ('Can anyone tell me the capital city of Argentina?'), to manage classroom activity ('Are you all ready now to put your pencils down and listen?') and to find out more about what pupils are doing ('Why did you decide to have just three characters in your play?').

Even the above analysis is an oversimplification, because a question can have more than one function (for example, to find out what pupils are doing and to make them think about it) and because it takes on special meanings in the life of a particular class (have they studied Argentina already or are they about to begin?). But the key point is that the distinction between form and function is important for analysing and evaluating teacher-pupil dialogue.

'Dialogic talk' and 'whole-class dialogue'

Through his comparative research in the primary school classrooms of five countries, Robin Alexander³ has shown that if we look beneath the superficial similarity of talk in classrooms the world over, we will find teachers organising the communicative process of teaching and learning in very different ways. In most of the classrooms he observed, teachers talked more than the pupils; but the balance and nature of contributions varied considerably, both between countries and between classrooms. One of the reasons for this variation was that in some classrooms a teacher's questions (or other prompts) would elicit only brief responses from pupils, while in others they often generated much more extended and reflective talk. The concept of 'dialogic talk' emerged from these observations as a way of describing a particularly effective type of classroom interaction. 'Dialogic talk' is that in which both teachers and pupils make substantial and significant contributions and through which pupils' thinking on a given idea or theme is helped to move forward. It may be used when teachers are interacting with groups or with whole classes.

I can illustrate my understanding of the function of this kind of talk through the example below. It was recorded in an English primary school by Open University researcher Manuel Fernandez, who is investigating the role of computers in children's literacy development. In this extract, the teacher is talking with some members of her year 5/6 class about their current activity; they are communicating by e-mail with members of a class in another local school about the shared curriculum topic 'How to have a healthy lifestyle'.

In the first part of the example, the teacher uses prompts to find out what the children have done. The first actual question comes from a child, on a point of spelling accuracy. When the teacher then begins to question the children, it is not to assess their spelling; it is to elicit their reasons for what they are writing to the children in the other school. She provides feedback on their answers ('OK. Excellent.'), so the episode has some features of the familiar IRF structure; but the teacher's questioning is used to encourage the pupils to perceive more clearly the nature of their task. She then picks up on what they have said to guide the next part of their activity, by suggesting that it will be useful to share their reasoning with their audience (and modelling how they might do it: 'We are sending you this information because ...'). She is using this interaction to build the knowledge foundations for the next stage of their activity - talking with them to guide their thinking forward. So we have here talk in which pupils make substantial and thoughtful contributions, and in which the teacher does not merely test understanding, but guides its development. What is more, all the pupils present are exposed to this reasoned discussion. This may not be 'whole-class dialogue', because the discussion is not shared with all members of the class; but it certainly seems to qualify as 'dialogic talk'.

We can consider further what 'dialogic talk' offers, from an educational point of view. One of the prime goals of education is to enable children to become more adept at using language, to express their thoughts and to engage with others in joint intellectual activity (their communication skills). A second important goal is to advance children's individual capacity for productive, rational and reflective thinking (their thinking skills). Dialogic talk can help achieve both these goals. The work of the Russian psychologist Vygotsky is relevant for understanding why this is so.⁴ He suggested that using language to communicate helps us learn ways to think. As he put it, what children gain from their 'intermental' experience (communication between minds through social interaction) shapes their 'intramental' activity (the ways they think as individuals). What is more, he suggested that some of the most important influences on the development of thinking will come from the interaction between a learner and more knowledgeable, supportive members of their community.

Although developed over half a century ago, Vygotsky's intriguing ideas have only really been put to the test in recent years. Now research has confirmed the validity of some of his claims about the link between language use and the learning of ways of thinking. Research has shown that teachers' modelling of ways of asking questions, offering explanations and providing reasons can have a significant and positive effect on how children use language in problem-solving tasks.⁵ Research by myself and colleagues has shown that a programme of carefully designed teacher-led and group- based activities enables children not only to become better at talking and working together but also at solving problems alone.⁶ The group-based activities of this programme are very important; but equally important is the kind of dialogue a teacher uses in whole-class plenaries and group monitoring. It is no coincidence that the teacher in the example above has been involved in this programme. And this brings us back to 'dialogic talk'.

Conclusion

For children to become more able in using language as a tool for both solitary and collective thinking, they need involvement in thoughtful and reasoned dialogue, inwhich conversational partners 'model' useful language strategies and in which they can practise using language to reason, reflect, enquire and explain their thinking to others. By using questions to draw out children's reasons for their views or actions, teachers can help them not only to reflect on their reasoning but also to see how and why to seek reasons from others. By seeking and comparing different points of view, a teacher can help those views to be shared and help children see how to use language to compare, debate and perhaps reconcile different perspectives. Providing only brief factual answers to IRF exchanges will not give children suitable opportunities for practice, whereas being drawn into more extended explanations and discussions of problems or topics will. This is the valuable kind of educational experience that 'dialogic talk' and 'whole-class dialogue' can offer.

References'

1 Sinclair, J and Coulthard, M, Towards an analysis of discourse: the English used by teachers and pupils, London, Oxford University Press, 1975.

2 Norman, K (ed) Thinking voices: the work of the National Oracy Project, London, Hodder and Stoughton, 1992; Edwards, AD and Westgate, DPG, Investigating classroom talk, second edition, London, Falmer Press, 1994; and Wells, G, Dialogic inquiry: towards a sociocultural practice and theory of education, Cambridge, Cambridge University Press, 1999.

3 Alexander, R, Culture and pedagogy: international comparisons in primary education, Oxford, Blackwell, 2000.

4 Vygotsky, LS, Mind in society: the development of higher psychological processes, Cambridge MA, Harvard University Press, 1978.

5 Brown, A and Palincsar, AS, 'Guided, co-operative learning and individual knowledge acquisition', in L Resnick (ed), Knowing, learning and instruction, New York, Lawrence Erlbaum, 1989; and Rojas-Drummond, S, 'Guided participation, discourse and the construction of knowledge in Mexican classrooms', in H Cowie and D van der Aalsvoort (eds), Social interaction in learning and instruction: the meaning of discourse for the construction of knowledge, Oxford, Elsevier, 2000.

6 Mercer, N, Wegerif, R and Dawes, L, Children's talk and the development of reasoning in the classroom, British educational research journal, 25(1), 1999, pages 95 to 111.

Pedagogical Strategies⁽ⁱ⁾

Teaching Approaches\Dialogic Teaching