ITRI-01-01 Relating Imperatives to Action

The aim of this paper is to provide an analysis of the use of logically complex imperatives, in particular, imperatives of the form`Do A1 or A2' and`Do A, if B'. I argue for an analysis of imperatives in terms of Classical Logic which takes into account the innuence of background information on imperatives. I show that by doing so one can avoid some counter-intuitive results which have been associated with analyses of imperatives in terms of Classical Logic. In particular, I address Hamblin's observations concerning rule-like imperatives and Ross' Paradox. The analysis is carried out within an Agent-based Logical framework. This analysis explicates what it means for an agent to have a successful policy for action with respect to satisfying his or her commitments, where some of these commitments have been introduced as a result of imperative language use.