Abstract

In this monograph the author attempts to explain what speakers say in using language and how what they say is understood by developing a generative theory of illocutions. Such a theory consists of a finite set of rules and a finite vocabulary structured together in such a way that there is one and only one description produced of what is said by a speaker in each and every case in which what is said can be understood differently. The theory is to mark all features of the utterance situation that produce these differences. In more conventional terms, Travis presents a kind of pragmatic theory for language use in which sentence-tokens can have characteristics in specific circumstances which are not features that belong to their sentence-types. In developing this model the author exploits features of transformational linguistic theory—specifically the notion of a generative description and the requirement that an adequate theory provides one and only one description of each distinct linguistic form. However, the author would not accept transformational theory itself as sufficient for a theory of illocutions because the former does not take into account contextual and nonlinguistic elements.