7. TOWARD A DIFFERENTIATED STANDARD OF PENAL JUSTICE

The self-contradictoriness as grounds of law of both formal liberty and positive freedom produces the phenomenon of mixture to which interpretive skeptics are fond of pointing as indicative of law's essential incoherence. Thus an accessory may be liable for the consequential offenses of the principal on a foreseeability standard, even though the principal is liable

only on a subjective standard.100 Automatism or an honest mistake of fact excuses from a crime of basic intent unless the mistake or automatism was attributable to the accused's drunkenness.101 Mental disorder negating intent excuses from murder if the accused is insane but results in a conviction for manslaughter if he is not.102 To the same point are legal fictions designed to feign a commitment to the subjective standard even as that standard is eroded; thus the rule that intoxication is no defense to a crime of basic intent is supported on the theory that the intention to get drunk supplies the mens rea for the offense charged.103

The skeptic's claim that such inconsistencies are endemic to the penal law can be justified only on the premise that the antagonism of the paradigms is a necessary and constant reality. This would be the case if there were indeed a logical imperative to absolutize one or the other principle as the ground of law. Yet the process whereby each fell into contradiction when absolutized in isolation shows us that there is no such imperative. If the elevation as an end of both formal agency and positive freedom leads to their self-contradiction, then neither can alone be the ground of law. That each suffers inversion when pursued at the other's expense shows that each needs the other for its own self-consistency. This should not be surprising. Formal agency was objectively an end only insofar as it manifested itself as such in property and bodily autonomy, so that it already implicitly contained the welfare dimension it purported to exclude. Similarly, both the general happiness and the common good were interpreted in terms of individual autonomy, so that they too implicated the end-status of the individual that their absolutization negates. Because each principle implicitly contains the other within itself, its domination of the other nullifies an element of its own nature and so results in its self-destruction.

The contradiction into which each principle falls when pursued intransigently discloses a conceptual whole wherein each is preserved as a constituent principle informing a distinctive but bounded subsystem of law. The genuine ground of law is neither formal agency nor the common good but the whole of which both are mutually complementary parts. This whole evinces the structure of mutual recognition we have called dialogic community and is indeed the archetype of which the relationships seen earlier are instances. The common good is both authentically common and authentically a good only insofar as it respects the discrete end-status of the individual agent and hence the distinctive legal paradigm ordered by that principle. Conversely, the individual agent is an end only insofar as it recognizes the authority of the common good whose self-inadequacy alone first establishes its end-status as a reality. That this structure of reciprocal deference is the true ground of law was revealed in the self-realization of the pretenders to that rifle; for it was, as we saw, the subtextual theme of both paradigms. And it was decisively revealed in the process whereby

each of these pretenders collapsed when absolutized to the exclusion of the other. Accordingly, the validation of dialogic community as the ground of law is just the process of its manifestation in, and emergence from, paradigms ordered to rival principles; and since the justification of a foundational end must be internal to its notion (or it would not be foundational), these paradigms are, despite their limitations, constituent elements of the foundation (129-130).

If the ground of law is dialogic community, then the principle of each paradigm is normatively resistant to the expansionist claims of the other. Though each principle is deficient insofar as it pretends to be the absolute ground of law, each is nonetheless preserved as an essential component of an adequate conception. The implication is that the penal law is a differentiated totality composed of two principal subsystems of law, each characterized by indigenous principles of justice and standards of fault. There is thus no "general part" of (in the sense of a single set of principles applicable throughout) the penal law; and yet the penal law is a unity, for each special system embodies in a distinctive way the ground of law that connects the systems to each other.104

That the penal law is a differentiated whole implies that the principle of each paradigm is valid only within limits consistent with the distinctive existence of the other. Once the whole comes forward as the ground of law, the hegemonic claims of the constituent principles cease to appear logically natural and become indicative instead of conceptual pathology. Conversely, their mutual respect is no longer a compromise of principle but a demand thereof. Neither is there any longer a "choice" between paradigms to undermine ab initio the rationality of legal discourse; for the principle of each model applies only within definite boundaries. The border between the systems is marked by the difference between the private rights of atomistic persons and the public goals of citizens united for the enhancement of their freedom. What judges call true crimes are denials of the end-status of personality as embodied in the exclusive authority it exercises over its body and external possessions and in the reasonable care of "neighbors." Public welfare offenses are breaches of statutes aiming at the general happiness taking individual values as given or protecting the social conditions of effective autonomy.105

Because each system is part of a whole, each must be actualized by judges and legislators with a moderation that reflects this constituent status. It is a mistake to extend the principle of either the pure agency or the welfarist paradigm over the whole penal law, for this absolutizes a principle whose validity is inherently relative to a specific context. Hence it is a mistake to make negligence the basis of liability for crimes against personality or to weaken the strict retributivist understanding of desert for such offenses. Subjectivist orthodoxy is appropriate for true crimes. However, it

is also a mistake to make subjective fault the standard of blameworthiness for welfare offenses or to assess the justice of penalties for such offenses from the standpoint of desert. Because welfare laws are justified instrumentally, the concept of desert strictly understood has (with one exception I will mention presently) no intelligible application within this sphere.

Even within the sphere of welfare offenses, the principle of penal justice is not monolithic. It varies, as we have seen, depending on the meaning of welfare. Where welfare means the satisfaction of preferences undetermined by the idea of effective autonomy, the appropriate principle of justice is the priority of autonomy over happiness, for the latter becomes normatively significant only through the end-status of personality. Applied to statutory interpretation, this principle rules out offenses of absolute liability (for persons but perhaps not for corporations) and favors defenses of reasonable mistake of fact and of due diligence in seeking to comply with or ascertain the law. At the legislative level, it screens out utilitarian (or other aggregative) justifications for penal laws as well as what Feinberg calls legal moralism, the penalization of activity (e.g., the consumption of alcohol) to advance a relative view of the good life; for these justifications signify the coercion of persons for the benefit of others. Insofar, however, as welfare means the satisfaction of values essential to effective autonomy, the appropriate principles of justice are that rights of formal liberty are validly limited by the common good to the extent necessary to achieve the aims of the law and to the extent that the legislative means do not contradict the end they purport to further. The latter proviso excludes absolute liability offenses in this context as well.

The principles of right applicable to welfare statutes yield a position regarding the legitimacy of imprisonment within this sphere. The priority of autonomy over happiness rules out prison sentences for offenses against nonfundamental values unless the statutory breach is intentional or advertently reckless. Imprisonment is justified for willful breaches, for although the intentional act would not have deserved punishment apart from positive law, the intentional breach of statute posits a principle subversive of law and so also of the lawbreaker's liberty. Thus imprisonment is here justified by desert. Incarceration is also ruled out for inadvertent breaches of laws promoting fundamental values, for the common good of autonomy cannot coherently be furthered through a total deprivation of liberty unwilled by the agent; here the means would contradict the end. We may conclude that the penalty of imprisonment understood as a qualitative deprivation of freedom is justified by desert (i.e., by willful lawbreaking) or not at all.

However, this result need not render welfare laws ineffectual. Where a statute protects interests essential to autonomy, the legitimacy of the means least restrictive to liberty implies that partial or quantitative restrictions of

liberty are justified not only by a willful breach but also where monetary penalties sufficiently heavy to deter would be beyond the financial capacity of most lawbreakers. Partial restrictions include probation orders and—at the extreme—confinement for intermittent periods. The latter penalty is justified, however, only if there exists a distinct class of penal institution for the noncriminal offender. Where restrictions on liberty are justified by a goal but not by desert (e.g., in the case of inadvertently dangerous driving), special institutions and administrative leniencies are required to reflect the categorical difference between welfare offenders and criminals. In their absence, confinement even for intermittent periods is wrong as a penalty for regulatory offenses. In general, there should be no implication of depersonalization for the welfare offender; his penalty is the intermittent confinement and nothing else. Further, the maximum overall term should be lower than that of the lowest maximum for crimes; and since the point of the sanction is deterrence rather than retribution for a wrong, the penalty need not vary with the consequences of the conduct.