Initiation-response-feedback

The seminal work on discourse analysis was carried out in classrooms, and it is from this that an understanding of structure of the exchanges that make up spoken discourse is drawn. Sinclair and Couithaerd (1975) identified three levels of discourse: the exchange, a turn-taking interactional sequence; the move, or contribution of a participant to the exchange in a turn; and the act, identifiable within the move and playing a specific linguistic function such as questioning or instructing. The three-part exchange teacher initiation / student response / teacher feedback (IRF) was proposed as the basic unit of classroom interaction, since referred to by Lemke (1985,1990) as dialogue. Mehan (1988, p.121) describes what are arguably features of classroom speech genre:

The instructional phase … is composed of characteristic interactional sequences. This exchange of academic information in interactional units is called ‘elicitation sequences.’ These units are interactional in that they are a joint production of teacher and student; they are sequential in that they occur one after the other in interaction. These sequences have three interconnected parts: an initiation act, a reply act, and evaluation act.

Students and teachers share an understanding of the social situation in terms of appropriate discourse and accompanying behaviour, i.e. teachers and students come together in classrooms knowing that certain types of interaction – including the IRF sequence – are appropriate in the situation, and it is their role to participate using the register of the classroom. The IRF sequence is generally equated with exchanges where the teacher is responsible for the initiation of interaction with the aim of eliciting ‘information’ ,and the exchange is continued, possibly “though extended sequences until the raply called for is obtained” (Mehan, 1988 ,p.127).

However , IRF sequences can be initiated by students, and it can be argued that at tertiary level this is both acceptable, and desirable, in tutorial situations as described earlier in this chapter. Newman et al. (1989, p.127), describe the IRF sequence as ideal for ‘instruction’ due to its “built-in repair structure in the teacher’s last turn so that incorrect information can be replaced with the right answer.” Likewise, it can be considered ideal for students in tutorials to initiate and continue clarification or discussion in the process of social construction of meaning and knowledge to the point that satisfies their need for interaction.

Since Sinclair and Coulthard’s (1975) identification of the three-part exchange research on classroom discourse moves has tended to focus on teacher initiations and the ‘art’ of teacher questioning, assuming students are primarily responders. It must be acknowledged, however, that students can volunteer, or initiate, in response to teacher initiations that are not directed to specific students, and some studies do show teacher-initiated IRF sequence can be, and are, more varied than the surface structure reveals ZAdger, 2001, p.505). In a complex classroom “ecology of social and cognitive relations” (Erikson, 1996, p.33) classroom interaction is rarely a sustained series of dyadic teacher-student IRF exchanges, and Christie (2002) argues that entire sequences of classroom discourse need to be examined in order to make Judgements of the value of IRF exchange, and that too many studies of classroom discourse focus only on exchange patterns, ignoring

The discourse level of speech act was used by Ramirez (1988) to analyse teacher-initiated IRF sequence, and revealed the initiation in a typical three-part exchange exhibited a “higher density of acts” (Ramirez, 1988, p.139) than the response and feedback moves. Ramirez called for finer distinctions of initiation, arguing the function of acts may lie in their prosodic features as much as the lexico-grammar. Basturkmen, (1999) further classified initiations as one of two types of elicitations: elicit of information – a request to the other speaker to supply missing information about which the speaker for information/ ideas about which the speaker has assumptions and for which the speaker is seeking confirmation. Elicits of confirmation are further classified using varying levels of speaker commitment to the act. This study adopts Basturkmen’s classification, which, in systemic linguistic terms, lies in the interpersonal function, and it is discussed in more detail in the section on methodology.