Aspects of Speech Act Categorisation: Towards Generating Teachers’ Language

In this paper we examine a possible method for classifying speech acts produced by human teachers, with a view of informing the designs of intelligent natural language tutors and of providing the basis for a formal analysis of the effects that teachers’ language has on students’ learning. We argue that traditional means as initiated by the Ordinary Language Philosophers such as Austin (1962), Grice (1975) and Searle (1979) are not sufficient to account for all types of linguistic phenomena occurring in educational dialogues. Two such phenomena are of particular interest to us: (1) speech acts which combine the qualities of several other less complex types of speech acts, and (2) the existence of many different speech act forms which can be used by a teacher to fulfil similar communicative functions in identical educational circumstances. We present our analysis of two sets of dialogues which shows that the main difficulty with using the traditional approaches is that they treat speech acts in discrete terms. We argue that categorising speech acts in such a way is not useful in modelling teachers’ language in that it does not explain the mechanisms involved in teachers’ linguistic choices. Following Givón (1989), we suggest that rather than being classified in a discrete manner, all speech acts should be explained in terms of how close they are with respect to one another on speech act continua derived from an interaction of many different communicative factors. We explain Givón’s proposal for an alternative, more flexible, approach and we take the first steps towards extending this approach to account for the linguistic phenomena of interest.