Reviews of devolution in Wales have been focussed so far on empirical description and positive analysis; there has been almost no discussion of how it should be judged or evaluated. I redress this imbalance by considering the arguments for devolution and the various normative values that may be used to evaluate the process. The arguments for devolution are of two basic kinds. Some are consequentialist: that devolution is desired on the grounds that it is believed to have good or desirable effects. Others are deontological: that devolution is thought to be intrinsically valuable. Any attempt to justify and evaluate devolution by consequentialist reasoning amounts to post hoc ergo propter hoc rationalisation. The social sciences are a long way from allowing the prediction of the global long-term net equilibrium effects of a major institutional change such as devolution in Wales. I further argue that the nonconsequentialist value of autonomy is the underlying justification for the devolution process. Devolution should be judged according to how it appeals to, supports or embodies this value.