Two Faces of Intention

We do things intentionally, and we intend to do things. Our commonsense psychology uses the notion of intention to characterize both our actions and our mental states: I might intentionally start my car, and I might intend to start it. My intention to start it clearly does not guarantee that I do. But beyond that it is not obvious how these two phenomena are related. A central problem for a theory of intention is to provide a plausible account of this relation. One thing seems clear: it is part of our commonsense psychological framework that these phenomena are not completely unrelated. In classifying both our actions and our states of mind in terms of some root notion of intention, commonsense psychology clearly assumes that there is some important commonality. Our problem is to say what this commonality is, by spelling out the relation between intentional action and intending (or, having an intention) to act. There are two common approaches to this problem. The firstthe desire-belief model-sees intentional action as action that stands in appropriate relations to the agent's desires and beliefs. ' This is a reductive model: it sees intentions to act as reducible to certain desires and beliefs.2 On this approach, the problem of the relation