In Opening Dialogue, Nystrand (Nystrand, 1997) draws on the Bakhtinian contrast between monologic and dialogic discourse, together with Gutierrez’s concept of instructional scripts (Gutierrez, 1994), to develop the notion of dialogic instruction. In monologic recitation, classroom talk is closely controlled by the teacher, with the aim of transmitting knowledge which students are required to remember. Dialogically organised instruction, on the other hand, is based on a different kind of relationship between teacher and students, in which students are asked to think, not simply to remember. For Nystrand, the study of classroom discourse is important because different modes of interaction place students in different positions as learners (p. 29):
Specific modes or genres of discourse engender particular epistemic roles for the conversants, and these roles, in turn, engender, constrain, and empower their thinking. The bottom line for instruction is that the quality of student learning is closely linked to the quality of classroom talk.
Opening Dialogue reports the findings of a large-scale study of the effects of patterns of classroom discourse on student learning in 400 English lessons in 25 US high schools. The major source of evidence was structured classroom observation in which teacher questions were coded on a series of dimensions. The research team also tape-recorded lessons and used this evidence to explore unexpected findings from the coded observations in more detail. They also interviewed participating teachers, and tested student learning outcomes by a written examination, scored against a number of criteria. Their results support the hypothesis that dialogically organised instruction is superior to monologically organised instruction in promoting student learning. Recitational patterns of talk were found to be overwhelmingly prevalent, and to have a negative effect on learning; they were particularly strongly concentrated in lower-track classes. Important aspects of the alternative, dialogic approach to instruction highlighted by the study were: the teacher’s use of authentic questions (where what counts as an acceptable answer is not prespecified); uptake, where the teacher incorporates students’ responses into subsequent questions; and the extent to which the teacher allows a student response to modify the topic of discourse, a strategy which Nystrand terms ‘ high-level evaluation’. He identifies a number of specific classroom methods which may help to promote the development of dialogic forms of understanding, including the use of learning journals, position papers drawn up and presented by students to the class, and peer response conferences (where students meet in small groups to review each other’s work).
Nystrand makes a particular contribution to our understanding in his discussion of the relationship between patterns of classroom discourse and the nature of the pedagogic contract established between a teacher and his/her students. As we have explained, the findings of his study do document that particular styles of interaction have an effect on student learning, for better or worse; but he goes on to argue that understanding this relationship cannot be mechanically reduced to measuring the relative proportion of authentic vs. ‘display’ questions over the course of a lesson, for example. He quotes transcripts of extracts from lessons by two teachers with contrasting styles to illustrate that the inappropriate use of authentic questions can be counter-productive; and that the skilful use of a lecturing style can on occasion be effective. For example, if the teacher asks many authentic questions which are unrelated to the topic of the lesson, then this is unlikely to help develop students’ understanding fruitfully; whereas a concise, clear exposition by the teacher may be the most efficient way of explaining the nature and purpose of a task before the class moves on to a new activity. Dialogic instruction will be supported by an increased use of authentic, topic-relevant questions on the part of the teacher, but more fundamental is the quality of the interaction which surrounds those questions. What matters most is not simply the frequency of particular exchange-structures in classroom discourse, but how far students are treated as active epistemic agents, i.e. participants in the production of their own knowledge.
Nystrand’s work marks the first sustained attempt to explore the significance of the Bakhtinian theory of dialogism for our understanding of the language of classroom instruction. His study demonstrates that choices made by the teacher can influence the conditions for learning established in the classroom, and in particular that the teacher does exert a measure of control on the structure and organisation of classroom discourse. He goes on to show that the preferred mode of interaction adopted by the teacher carries consequences for the epistemology of the classroom: broadly, the teacher can orient towards controlling what knowledge is produced, or towards structuring the activities through which students produce knowledge. The study is impressive in scope and makes a strong case for the superior effectiveness of dialogically-organised instruction: students taught in this way tend to do better in written tests than those taught using a monologic, recitational approach.
One drawback of the methodology used in the study is that the central plank of evidence is a record of the coding of classroom interaction made by observers in real time. Although simultaneous tape recordings were made of the lessons observed, these are treated as supplementary evidence rather than the chief source on which the findings are built. Consequently, with the exception of a small number of short transcribed extracts, the original discourse which was spoken cannot be reconstructed; rather we have a global summary of the tendencies in the data (e.g. the preponderance of test questions from teachers and the infrequency with which authentic questions are used). However, as Nystrand’s own findings indicate, in understanding how the structuring of classroom discourse operates, the devil lies in the detail. For example, he notes how the research team’s initial coding of the data threw up some unexpected results, such as the fact that the use of group work appeared to have a negative effect on student learning. When the research team inspected the data more closely, including checking their coding against the recordings they made, it emerged that activities which had been coded as group work were often, in practice, individual work by students who were merely seated in groups. A re-analysis of the data showed that group work was effective when the activity required genuine collaboration, and when the teacher specified the goals clearly, but gave groups autonomy in carrying out the task. Whilst Nystrand makes a convincing case for this general interpretation, putting the flesh on the bones of a theory of dialogic instruction will require closer attention to the detailed analysis of transcripts of the discourse actually spoken by participants in classroom exchanges, since it is at this level of granularity that we can see talk at work in shaping the learning process that students experience. A further question which his study raises, and which future research in this area needs to explore, is why recitational approaches to teaching continue to be so prevalent, given their apparent ineffectiveness in engaging student interest or in securing improved outcomes in attainment. Since it seems unlikely that the majority of teachers would choose to rely on a pedagogic style calculated to depress student learning in the absence of strong constraining factors, we need to investigate the structural conditions which reproduce monologic patterns of instruction on a social scale.