David K. Chan

In "After Anscombe," I argue that, although Bratman's account of intention "has provided a conceptual tool for many directions of research in philosophy and cognitive psychology," it cannot do the work in ethics that moral philosophers, especially Kantians, use it for. This can be shown by considering the problems in using intention to make a moral distinction in cases of double effect. If so, Bratman's is not the same concept of intention that Anscombe had in mind when she wrote her book. I show that Ancombe's account of intention is a much broader concept, with deeper historical roots. This does not mean that Bratman's concept of intention should be abandoned. Rather, we need to make a distinction between desire and intention that is implicit in Anscombe's account. That is, the aspects of intention that make her concept suited for ethics do not belong to Bratman's narrower concept and should be attributed to a concept that can be distinguished from intention. Evidence for what the latter concept is can be found in Aquinas' account of intentio from which the concept of intention has borrowed its name. Intentio for Aquinas is an act of will in regard to the end, whereas Bratman's intention is of the means that are chosen in practical reasoning. Intentio seems to be an intrinsic desire for the end. If this is right, then it is desire that is relevant for ethics, and not intention in its modern use. Recognition that desire and not intention is the relevant concept for ethics, in both Anscombe's and Aquinas' accounts of human agency, dovetails perfectly with their ethical theories, for it is Aristotle's ethics that has guided both philosophers in their ethical writings. After all, Aristotelian virtue ethicists are concerned with moral evaluation of the agent's character, which consists in a set of dispositions to seek ends that are desired for their own sake.

The problem of the nature of action is to say what else there is to action besides the occurrence of the event brought about in acting. My approach is to distinguish action from non-action on the basis of a special mental cause of the physical event brought about in acting. ;I begin with accounts that make do only with reasons, i.e. beliefs and desires, as the mental causes sought. But the belief-desire model of action does not give the necessary condition for action because non-intentional actions are actions done for no reason whatsoever. Sufficiency is ruled out by cases of deviant causality. Intentions do not escape the problems that prevented beliefs and desires from serving as the mark of action. Volitionalist theories of action need to avoid the dilemma posed by Ryle, one horn of which is the infamous problem of a regress of volitions needed to explain volitions. The accounts of volition that evade Ryle's dilemma do so at the cost of leaving the property of being an action as something sui generis and unanalysable. ;I isolate a special notion of trying which has as its object only the performing of teleologically basic actions. I argue that an agent has an immediate awareness that he is trying when he is trying. But trying is physical as well as mental. One of the distinctive functions of trying is that of control. It is a deep metaphysical truth about agents as we know them that the control they exercise over the physical actions they perform goes by way of a physical mechanism. To avoid the problem posed by causal deviance, I argue that any causal route from trying to movement is admissible so long as the agent is immediately aware that he is trying when the mechanism is activated and it is this trying that is the cause of the movement. To complete the case for trying being the mark of action when it causes the requisite events, I show that when one performs non-intentional actions, one always tries and the trying is causally related to the events one brings about