DAVID VELLEMAN REPLIES TO DISCUSSION ON THE POSSIBILITY OF PRACTICAL REASON The critiques of Jonathan Dancy and Al Mele give me an opportunity to clear up some misunderstandings of my view. Nadeem Hussain’s critique gives me an opportunity to revise the view in some crucial respects. I am grateful for both opportunities. 1. DANCY Dancy asks whether the self-knowledge at which an agent aims, in my view, is knowledge of what he is currently doing or knowledge of what he is going to do. In the first instance, I think an agent aims at the former. But Dancy may find my view confusing because I also think that an agent’s primary means of attaining the former is by attaining a short-term version of the latter. In my view, an agent wants it to be the case that he knows, at any given time, what he is doing at that time, but he ensures this outcome by not doing anything until he knows that he is going to do it. This knowledge is of course practical knowledge of the sort that causes what is known – causes it, in my view, by way of the agent’s inclination to do what he thinks, so that his thinking it will constitute knowledge of what he is doing. But the fact remains that the agent attains contempo- raneous knowledge of his actions by attaining anticipatory knowledge of them. Dancy then objects that I frequently shift between saying, on the one hand, that an agent wants to know what he is doing and, on the other, that he wants to know why he is doing it, or to make sense of what he is doing. As Dancy later acknowledges, I raise and answer this very objection in the Introduction to the book (26–27). My view is that an agent wants to know a Philosophical Studies 121: 277–298, 2004. Ó 2004 Kluwer Academic Publishers. Printed in the Netherlands.