The dead end of deterrence, and beyond.

• Author: Huigens, Kyron

INTRODUCTION

The deterrence theory of punishment has been taken a long way down a new path.(1) The descendants of Bentham now argue that the criminal law addresses "the moral goodness of citizens' motivations."(2) The heirs of Hume and Holmes maintain that the criminal law is a "[p]reference-[s]haping [p]olicy."(3) These developments are as welcome as they are surprising. One wonders, however, if the deterrence theorists who have taken the new path know where they are, how they got there, or where they are going. Dan Kahan, the leader of this hardy band of scholars, eschews "deep theorizing,"(4) and scoffs at the notion that some "abstract `theory' supplies the `truth' about criminal law."(5) As it happens, Kahan and company inadvertently have taken deterrence theory to its inevitable dead end. It is time to move beyond the notion of deterrence to a new conception of punishment's justifying purpose.

This Article consists of a brief statement of a virtue ethics theory of punishment,(6) which follows an extended examination of three versions of deterrence theory: those of Paul Robinson, H.L.A. Hart, and Kahan. The logic of this organization can be stated briefly. The virtue ethics theory of punishment answers a question--What is the nature and significance of fault in just punishment?--into which deterrence theory never has offered a persuasive answer. Kahan's "new path" deterrence theory almost gets it right. This Article picks up where Kahan's work leaves off.

The dominant consequentialism of Anglo-American legal theory gives the deterrence theory of punishment credibility,(7) but this theory has always suffered from a prominent weakness. Fault--also known as desert, culpability, or blameworthiness(8)--is the distinctive feature of the criminal law,(9) but consequentialism has no independently viable conception of fault. Fault is an irreducibly retrospective concept, and the inveterately prospective orientation of deterrence theory's underlying consequentialism cripples its efforts to give an adequate account of fault.(10)

One possible response to this evident shortcoming in deterrence theory is simply to adopt and adapt the retributivist conception of fault as intentional wrongdoing--a strategy that Paul Robinson has developed.(11) The requirement of intentional wrongdoing can be imposed as a side-constraint on punishment, for reasons that are distinct from punishment's justification, but that are perfectly sound. However, this side-constraint interpretation of fault and the liberal and efficiency-minded reasons given in support of it cannot explain the frequent and apparently just punishment of nonintentional wrongdoing that is not merely strict liability. The explanation of nonintentional fault requires an understanding of genuine fault--"genuine" in the sense that it is an affirmative, justifying reason to punish in a given case, and not only a necessary condition. This conception of fault, like any adequate conception of fault, lies beyond the theoretical resources of deterrence theory.

H.L.A. Hart takes a second, fundamentally different, and admirably straightforward approach to the question of fault in the criminal law.(12) He denies forthrightly that fault in the sense of "moral culpability" is either a necessary condition for punishment or an affirmative, justifying reason to punish.(13) Hart retains the requirements of voluntariness and the absence of excusing conditions, such as duress and insanity, as side-constraints on punishment.(14) He calls these the "criteria of responsibility,"(15) and he collapses mens rea elements of proof into the same liability-limiting category.(16) One result of these steps is a persuasive defense of nonintentional criminal liability in the form of negligence liability. But even in his deservedly influential analysis of negligence, Hart deliberately denies the existence of nonintentional fault--indeed, of any kind of fault. I rely on J.L. Austin's classic essay, A Plea for Excuses,(17) to argue that this rejection of fault is untenable.

Hart might have developed, rather than obscured, the concept of fault.(18) Hart argued elsewhere that law does not govern only by way of conscious instrumental reasoning, and that people can be governed by legal rules at the level of their motivations.(19) This idea can premise the theoretical account of genuine nonintentional fault that deterrence theorists have failed to provide. Kahan's "new path" deterrence theory is important because it picks up this promising line of thought, albeit not from Hart. Kahan attempts to describe the ways in which the criminal law shapes motivations, and in his account of mistake he has sketched the outlines of a deterrence theory of genuine fault.(20)

Kahan's efforts in this direction fall short.(21) He, too, fails to develop the concepts of genuine fault and nonintentional fault, because he stays within the confines of consequentialist ethics. Under the less than benign influence of the law and economics movement, he fails to work out the full implications of his own insights concerning value, practical reasoning, and the criminal law. Kahan's difficulty is that he has adopted Lawrence Lessig's theory of the regulation of social meaning as a basic theoretical structure.(22) Lessig's theory itself is an attempt to cross the economic analysis of law with certain conceptions of value and practical reasoning that, heretofore, have had no place in economic thinking.(23) Lessig's difficulty is that these notions come from the nonconsequentialist tradition in philosophical ethics, and they simply cannot be incorporated into economic analysis, which is an inveterately consequentialist enterprise. The incoherence of Lessig's theory obstructs Kahan's "new path" deterrence theory.

Yet, the difficulties of Kahan and Lessig are not my main concern. I am interested in the nonconsequentialism that they fail to use persuasively. The features of nonconsequential value and practical reasoning that interest Kahan and Lessig form the foundation of virtue ethics.(24) Virtue ethics is an ancient body of thought that begins with Aristotle's Nicomachean Ethics,(25) but that has been revived in philosophical ethics only within the last fifty years.(26) It has never played a significant role in modern legal theory, which has been dominated for at least the last one hundred years by consequentialism.(27) This, however, is what makes Kahan's and Lessig's efforts so interesting and, for my purposes, so useful. Their efforts at hybridization demonstrate the inadequacy of consequentialist legal analysis and of law and economics in particular; and demonstrate, at the same time, the plausibility and importance of nonconsequentialist conceptions of value and practical reasoning. It is a short step from here to the consistent nonconsequentialism of virtue ethics.

A virtue ethics theory of punishment does provide an adequate account of fault.(28) It portrays fault as an inference, drawn from the particular circumstances and manner of the wrongdoing of the accused, to the effect that the practical judgment of the accused is inadequate or flawed. This defective practical reasoning justifies the incapacitation of the guilty person, in proportion to the defect as it is manifested in minor or major wrongdoing. The adjudication of fault is a prerequisite to just punishment precisely because punishment's justification in any individual case rests on the defect in practical judgment that the inference of fault reveals.

As a final note of introduction, before I make the foregoing arguments in more detail, let me point out one critical contrast between deterrence theory and a virtue ethics theory of punishment. Consequentialism's emphasis on the production of states of affairs in which social welfare is optimized leads the deterrence theorist to overlook the criminal law's concern with the particularities of individual cases.(29) Not coincidentally, the particularities of individual cases are the special concern of those features of the criminal law that deterrence theory chronically neglects: fault and adjudication. One can only marvel at the long dominance of a theory of punishment that, at best, finesses the task of explaining two of the criminal law's essential features. My hope is that the reader's appreciation of the magnitude of these omissions will motivate her to see through to the end the present, lengthy exploration of an alternative theory of punishment.

Part I of this Article lays out the basic concepts of the theory of punishment, and describes the retribution and deterrence theories, with an emphasis on the latter. Part II argues that deterrence theory's adoption and adaptation of retributive theory's construction of fault is a failure because deterrence theorists have not developed the notion of nonintentional fault. The best version of deterrence theory moves away from the concept of fault altogether, even though its author, H.L.A. Hart, offered elsewhere good reasons to move in the opposite direction. Part III examines Kahan's "new path" version of deterrence theory and Lessig's "social meaning" version of law and economics. The main burden of this Part is the introduction of the features of nonconsequential value and practical reasoning that Lessig and Kahan fail to address persuasively. Part IV begins with a description of the normative theory that is implied by these features of nonconsequential value and practical reasoning: virtue ethics. The second section of Part IV draws these several threads together in a brief but comprehensive statement of a virtue ethics theory of punishment that explains the justification of punishment, the nature of criminal wrongdoing, and the meaning of fault in the criminal law.

1. THE BASIC CONCEPTS OF THE THEORY OF PUNISHMENT

   An adequate theory of punishment should give an account of the two different sets of norms that make up a criminal justice system: the primary norms that...