Law & Society Blog

Retributive punishment presupposes the existence of a moral community. By moral community I mean a social arrangement that makes moral values real. Moral qualities, qua moral qualities, have no causal power. Yet they are recognizable, either as absolutes or as social conventions. Moral communities create (or detect), observe, and act because of the moral quality of an act or a state of affairs. They thus connect moral qualities to real-world consequences. On the individual level, such connection will be psychological, in that moral properties serve as motivators. On a societal level, the connection will be institutional, for example in enforcing laws and social norms, as in the case of punishment. In any event, moral communities are social environments that treat moral qualities as causes for real-world effects, and thereby give effect to otherwise inert moral qualities.

Of course, since the causality of moral qualities is the result of social institutions and psychological conditions, effects from moral causes are not necessary in the sense that physical objects, without fail, obey the laws of nature. Bridging the gulf between the world of the moral properties and the effects of such properties, or just explaining the nature of that gulf, has rightfully been the focus of much moral philosophy. For Kant, famously, moral properties were noumena, and part of the grandiose concept of the Critique of Practical Reason was to explain how such noumena could cause effects in the phoenomenal world, that is, how “pure reason may become practical.” The intelligible yet causally unreliable nature of moral qualities is reflected in our choice of language. For example, in describing moral qualities without regard to their causal impotence we tend to use simple declaratives (”Murder is wrong.”) But when we take into account the contingent nature of the effects of moral properties, we tend to restate such concepts either in imperative form if our focus is on psychological contingencies (”Thou shall not murder”), or in the form of executable code, addressed to social institutions (”Whoever kills someone else shall be imprisoned for no less than 5 years.”)

The problems, then, for a theory of punishment, are no different from those facing any first-order moral theory. Prominently among them: (i) What states of affairs or actions are good and bad? (ii) To what extent are we justified in requiring compliance with any rules, promulgated (or discovered) on the basis of (i). This view entails a rejection of the (relative) independence of punishment theory from moral theory. It is not possible to give a meaningful account of punishment independent of the normative order that underwrites it. It also explains why consequentialist theories are so immediately and persistently plausible (which is not to deny that they are facing significant objections), namely because they not only give a reason for why we are justified in seeking compliance through punishment, but also, and more importantly, because they contain the outlines of a first-order moral theory, usually a form of utilitarianism, that takes human welfare seriously. And human welfare is likely to be a key concern in every moral community, whatever else that community might be about in addition to human welfare. The question is, of course, whether a focus on human welfare alone is sufficient to identify an adequate set of moral properties and to justify requiring compliance with that set. But structurally, at least, consequentialism is able to address both issues. The same cannot be said for most variants of retributivism. Most variants of retributivism focus on (ii), and try to explain why punishment is an adequate response to the wrongdoing (because a rule has been broken, because fairness requires that the balance of benefits and burdens be restored, because punishment expresses condemnation of the wrongful act, etc.) That endeavor is incomplete (and, consequently leads nowhere) unless coupled with an appropriate first-order moral theory. That’s what writers such as Rawls, Hare, and Hart have done in the context of consequentialist theories. Genuine modern attempts at creating a non-consequentialist first-oder account, however, have been far and few between. Most often, writers on punishment in the retributive tradition refer to Kant and sometimes Hegel to underwrite their partial theories. But these casual hyperlinks are insufficient. If retributivism is to succeed as part of a substantive moral theory, it must present a plausible, fully integrated account of the subject, of society, and of the moral good. I have argued for such a theory, drawing on notions of recognition as developed by Hegel and Fichte, and refined further by writers such as Kojeve and Williams. But are those concepts really more persuasive than a bare-bones consequentialism? Are they even normatively adequate for an increasingly diverse society? Most design choices of consequence in punishment theory will follow from that answer.

License

This work is published under a Creative Commons Attribution-Noncommercial 2.5 License.