Document Type


Publication Date



Retributive tbeory has long held pride of place among theories of criminal punishment in both philosophy and in law. It has seemed, at various times, either much more intuitive, or rationally persuasive, or simply more normatively right than other theories. But retributive theory is limited, both in theory and practice, and in many of its versions is best conceived not as a theory of punishment in its own right, but instead as shorthand for a set of constraints on the exercise of punishment. Whether some version of retributive theory is a live possibility in the contemporary world remains very much an open question.

In my essay, I consider three interrelated lines of attack against retributive theories of punishment: first, that it relies on philosophical assumptions that are either unrealistic or false; second, that the notion that offenders deserve to be punished, whatever its intuitive appeal, is possibly an empty idea and in any case one unsuited to a politically liberal state; and third, that tile abstractness of most versions of retribution render it unable to offer much in the way of useful, I or concrete policy advice. I f retributivism is to be taken seriously as a robust theory of punishment, it needs to provide replies to each of these lines of criticism.