In this paper I lay out the bare bones of my conception of practical reasoning, which I understand as similar in all relevant respects to theoretical reasoning except that (as it is put) the conclusion of practical reasoning is either action or intention, while the conclusion of theoretical reasoning is belief. I then turn to ask how, on this conception, moral education is possible--understanding moral education as more practical than theoretical. We want people to do the right things, not just to believe that they ought to do them. But the first task is to identify the right thing to do and this requires practical reasoning since it involves tracing reasons to act in one way rather than another, the right act being the one there is most reason (of a certain sort) to do. I maintain that a certain form of casuistry can take people who already have some competence with the notion of a reason and enable them to develop that competence by a process of explicit questioning, without our needing to see this process as a form of indoctrination.