Over the past ten years or so, there has been a renewed interest in the moral education theory of punishment. The attractions of the theory are numerous, not least of which is that it offers hopes for a breakthrough in the apparently intractable debate between deterrence theorists and retributivists. Nevertheless, I believe there are severe problems with recent formulations of the theory. First, contemporary educationists all place great emphasis on autonomy, yet fail to show how continued respect for autonomy is (...) compatible with achievement of their stated punitive goals. Second, educationists have, and possibly must, take incarceration as the best available punitive sanction. Yet it is unclear how morally educative such a punishment will be. Third, contemporary educationists view punishment as a benefit to be conferred on an offender. But educationists have not succeeded in arguing that society is obligated to confer such benefits, nor have they adequately defended the Platonic moral psychology necessary to show that moral education is always a benefit to justly punished offenders. Fourth, contemporary educationists are hopeful that an indeterminate sentencing policy can be avoided, but I argue that such a policy is an ineliminable component of an educationist justification of punishment. Finally, I raise some doubts about the scope that educationist goals ought to have in any comprehensive theory of punishment. (shrink)