Legal Moralism and Liberalism
Jeffrie G. Murphy*

The law must protect ... the institutions and the community of ideas, political and moral, without which people cannot live together.  Society cannot ignore the morality of the individual any more than it can his loyalty; it flourishes on both and without either it dies.
                                                                 Lord Patrick Devlin

I. Introduction

        One of the few benefits of growing older is that one is offered the opportunity for pleasures of nostalgia.  This conference provides many such pleasures for me.  It allows me to return to a university at which I spent many happy and productive years and to honor Joel Feinberg -- a man for whom I have great esteem and affection.  I am also pleased to have as my commentators two persons for whom I hold similar feelings: Herbert Morris and Jean Hampton.  To Joel Feinberg and Herbert Morris, I also owe debts of intellectual and personal gratitude, for these two men have both played important roles in my philosophical life -- roles that they have perhaps forgotten or of which they were never even aware.

        In 1965, I was a doctoral student at the University of Rochester -- finishing a dissertation on Kant's moral and political philosophy.  During that year, Joel Feinberg read a paper at a departmental colloquium -- an early version of his essay The Expressive Function of Punishment.   That paper was my first introduction to the philosophy of law, and Joel's presentation on that occasion was so stimulating that I made a mental note that here was a kind of philosophy that excited me more dramatically than any other kind ever had and that someday, when I could put my Kantian labors on hold, I would have to return to it.

        Two years later, I was teaching as a visitor at the university of Michigan.  I had by then taught legal philosophy a couple of times at the University of Minnesota, but I was still a rank amateur at it.  Herbert Morris presented a version of his essay Persons and Punishment to the Michigan department and later met with several faculty members at Dick Brandt's apartment.  I spoke to Herb at length about my developing interest in the philosophy of law.  As a result of that conversation, and some equally supportive and helpful correspondence, I decided to take a year off to study law at UCLA.  There Herb Morris taught me, among other things, how to learn and think about the criminal law.  My time at UCLA -- including many stimulating conversations with Herb and with Dick Wasserstrom -- transformed my intellectual life and future.  From then on, legal philosophy was my dominant interest; and my approach to the subject was deeply influenced by my UCLA experience.

        I was present at the birth of two of the classic essays in 20th century legal philosophy, and I was inspired and guided by the men who wrote them.  This conference thus has great personal as well as intellectual meaning for me, and I am delighted to be a participant in it.

II. Philosophy of Law Circa 1965

        What, then, am I going to talk about?   Even my topic, I fear, carries with it the strong aroma of nostalgia.  When I first began teaching the philosophy of law, one of the primary topics being discussed -- perhaps the main topic -- was the issue of legal moralism, an issue that was then often referred to simply as "the Hart-Devlin debate."  Should the criminal law be limited, as Mill and (with qualification) H.L.A. Hart had argued, by the harm principle (the principle that state coercion is justified only to prevent one citizen from violating the rights of another citizen and in that sense harming him) or could the criminal law in principle be used to enforce any of the important moral convictions of the community -- even if the targeted conduct (e.g., private sexual behavior between consenting adults) was not in any obvious way rights-violative or harmful to others?  This latter position, a kind of communitarianism often called legal moralism, had been defended by the English judge James Fitzjames Stephen in the 19th century and by the English judge Lord Patrick Devlin in 1965.1  They conceded that there would often be good practical reasons for not using the criminal law to enforce certain moral prohibitions, but they refused to grant that it was ever wrong in principle to do so.  Like most good liberals I sided with Hart and Mill against Devlin and Stephen and even published a perfectly dreadful and justly forgotten essay on the topic.2   I believed, along with most of the people with whom I talked about legal philosophy, that legal moralism had been properly killed off, that liberalism had once again been vindicated against the forces of superstition and oppression, and that legal philosophy could now move on to new and more important topics.

        I no longer retain my confidence in the complete victory of liberalism on this issue.  I have now come to believe (perhaps simply as a further sign of growing older) that the conservative position of legal moralism retains some philosophical power -- that it can, at the very least, pose some serious problems for the liberal theory of morality and the state.  My remarks are in part prompted by some recent writing by Joel Feinberg on the topic -- who apparently also agrees that the issue is not quite as dead as most of us believed it to be a few years ago -- and thus my paper to honor Feinberg will in part be about Feinberg.

III. A Return to the Hart-Devlin Debate

        In 1987 Joel Feinberg published an essay entitled "Some Unswept Debris from the Hart-Devlin Debate."  An earlier version of this essay had been presented at the meetings of the now defunct Arizona Philosophical Association (where I served as the commentator) and an expanded version was later published as a part of his book Harmless Wrongdoing, the final volume of his masterful four volume work The Moral Limits of the Criminal Law.3

        In his discussion of the Hart-Devlin debate, Feinberg contrasts two theories of criminalization: "liberalism" and "legal moralism."   Liberalism is the view that "the prevention of harm or offense to [nonconsenting] parties other than the actor is the only morally legitimate reason for a criminal prohibition."4  Legal moralism is the view that it is sometimes legitimate to use the criminal law to prevent actions simply because those actions are "inherently immoral even if those actions cause no harm or offense to nonconsenting third parties."5  (I am not sure what the phrase "inherently immoral" means, and I will try out various interpretations of it as I go along.)

        Given this characterization of legal moralism, I am not persuaded that Lord Devlin is a legal moralist.6  Devlin does spend a great deal of time arguing in defense of criminal prohibitions against such activities as oral and anal copulation (sodomy) and even seems to support the idea that these statutes may be aimed specifically at male homosexuals.  His reason for this, however, does not seem to be that he regards these activities as inherently immoral if by "inherently immoral" we mean "immoral on the basis of properties of the act itself rather than on the basis of how the act is perceived or judged by others."   Thus Devlin might, for all I know, be prepared to concede that it may be difficult to find plausible secular (and thus constitutionally acceptable7) grounds for regarding occasional homosexual sodomy between consenting adults as in itself immoral.  Take your favorite moral theory.  Occasional homosexual sodomy, unlike (say) occasional torture, does not seem disutilitarian.  It does not seem to rest on a nonuniversalizable maxim.   It does not seem -- at least in any obvious way -- to extinguish the possibility of living a virtuous life of human flourishing -- unless, of course, one draws one's theory of virtue and flourishing from a religious source.8  A life devoted exclusively to homosexual sodomy would, of course, get low marks on all these moral score cards; but so would, I suspect, a life devoted exclusively to any one passion or activity.  Thus the person setting out to justify criminalizing consensual sodomy might have a great deal of difficulty in persuading anyone (not already committed to the view on religious grounds) to adopt the view that such activities are inherently immoral and thus would, I think, find it very difficult -- though not impossible -- to be a legal moralist on this issue.9

        It is interesting in this regard that Devlin never seeks to show that consensual sodomy is inherently immoral (it is perhaps a strength of his position that he does not have to try), nor do any of his critics spend much time in attempting to show that it is not.  It seems to be understood that this is simply not the central issue.

        What then is the central issue?  It is, I think, Devlin's conviction (shared by James Fitzjames Stephen, Emil Durkheim and others) that society is bound together, not by moral truth, but simply by shared moral beliefs -- however irrational and unenlightened those beliefs may be.  Devlin is thus able to construct the following argument: the criminal law is legitimately concerned with the preservation of society, violations of a society's shared morality tend (like treason) to undermine society even in cases where these violations have no direct personal victims, and thus the criminal law may legitimately prohibit such violations in those cases where the majority judges this to be a prudent course of action.  Viewed in this way, Devlin is not challenging liberalism but is rather exploiting a tension within liberalism itself -- the tension between the value of individual liberty, on the one hand, and the value of democratic rule toward utilitarian ends on the other.  Devlin is not a legal moralist.  He is rather a utilitarian, democratic cynic with some controversial empirical views.  Utilitarian because he regards social harmfulness, in some very extended sense, as the only factor relevant in justifying a criminal prohibition.   Democratic because he believes that the majority has a right to have its preferences enacted into law absent some compelling reason why they should not be.   Cynical because he believes the social importance of a moral belief is not a function of its truth or reasonableness but is solely a function of its pervasiveness and the degree to emotional intensity with which it is held.  And empirically controversial for this reason: he holds extremely confident beliefs about the extent and depth of the moral repugnance to sodomy and about the harmful impacts of challenges to those feelings of repugnance and he holds these beliefs, so far as I can tell, on the basis of little more than hunches or anecdotal evidence.  His belief that private acts of consensual sodomy tend, like treason, to undermine society seems to Hart to be evidentially on a par with the Emperor Justinian's belief that homosexuality causes earthquakes.  I do not think that Devlin's view is quite that silly, but I do think that the presented evidential basis for his view is fairly flimsy.

        One could make a case for something like his view, however, that would at least be worth discussing.  One might, for example, argue that open toleration of the flouting of sexual norms threatens the honorific position historically accorded the traditional nuclear family and that such a threat risks undermining the social stability generated by such family units.  I have no idea to what degree conservative claims of this nature are true; and, indeed, I tend to be quite skeptical about many of them -- particularly skeptical when they are directed against homosexuality, but somewhat less skeptical when they are directed against routine divorce, adultery, and promiscuity.10  Such claims are not just plain silly, however.  They merit a rational response based on a fair weighing of all available evidence.  It is possible, for all I know about ancient science, that Justinian's belief about the causal connection between homosexuality and earthquakes also deserved a rational response in his own day.  It no longer does, of course.

        If I am generally correct in the characterization I have given of Devlin's view, then the differences between Devlin and liberals such as John Stuart Mill seem not as interesting as one might have initially thought.  Most of the differences are not on deep issues of ultimate principle, but rather on issues of empirical evidence of social harm.  And Devlin on these issues seems to rely more on hunches than on solid evidence, a tendency exhibited by many utilitarians.  Mill's own case for freedom, after all, is based largely on empirical hunches about the long-range social benefits of freedom, and I am not at all sure that there is overwhelming evidence in support of his extremely strong views on the matter.

        It is perhaps worth noting in passing, by the way, that the current AIDS epidemic might provide a prima facie public health and thus purely secular rationale for a legal concern with sodomy, but this would be based on claimed actual harmful effects of the practice itself -- not on harm supposed to flow from challenges to beliefs about the practice -- and thus would provide no support for the kind of view Devlin seeks to defend.

        In spite of the shortcomings of Devlin's own positive views, however, two important issues of principle that are worthy of serious thought and reflection can be generated from his negative discussion -- his counter-attack against Hart and other liberals who have attached him.  These issues arise as consistency challenges to  moral and political liberalism -- challenges that a rejection of legal moralism, demanded by liberals in the area of sexual freedom, is inconsistent with two other doctrines that many (if not all) liberals hold dear: first, the idea that criminal punishment should be based, at least in part, on the retributive notion of desert or blameworthiness and, second, the idea that democracy constrained by a set of fundamental rights is the preferred form of government.11  It has often been claimed, by followers of Marx and other philosophical radicals, that the ideology of liberalism is filled with internal contradictions and will eventually explode as a result of the tensions generated by such conflict.  Some of Devlin's challenges can also be read in this spirit.  I will thus spend some time exploring these two possible inconsistencies or contradictions -- contradictions that may beset, if not all forms of liberalism, at least liberalism of a certain common and compelling sort.  (It is a kind of liberalism, for example, that I have found compelling.)  The liberalism in question involves the following three claims:

        (1) The Harm Principle: Legal coercion is justified only to prevent one citizen from violating the rights of another.  Any other basis for state coercion -- particularly the attempt to promote personal virtue -- would itself violate a fundamental moral right of persons -- the right that Ronald Dworkin calls "the right of moral independence."12

        (2)  Retributivism: Punishment should be inflicted, at least in part, on the basis of desert -- on the basis of the blameworthiness of the individual criminal.  (A retributivist might call punishments that disregard desert "inherently immoral," the phrase "inherently immoral" now meaning "immoral for nonutilitarian or nonconsequentialist reasons.")

        (3)  Fundamental Rights Constitutionalism: Democratic rule is of great importance, but it must sometimes be radically limited if its exercise would compromise vital liberties.  All liberty is important, of course, but only some liberties are important enough to be protected as fundamental rights at the constitutional level.

        In what follows I shall be concerned to argue that (1) -- given certain plausible interpretations -- is in tension with both (2) and (3) -- given certain plausible interpretations.  In short: a person who subscribes (for certain reasons) to The Harm Principle may find it difficult consistently to embrace either Retributivism or Fundamental Rights Constitutionalism.  Thus any form of liberalism that would seek to package them together may indeed risk the kind of deconstruction that radicals like to predict.  I shall first address the tensions involving retributivism and then move to a discussion of the constitutional tensions.

IV. Punishment and the Liberal State

        A liberal theory of the state makes liberty and individual rights the primary political values.  Can a theory of this nature consistently endorse a theory of punishment that approves the use of state power, not in pursuit of the defensive social goals of crime control and harm prevention (protection of rights and liberty), but rather in pursuit of what may seem a purely moral objective -- namely, the assumed good of retribution, of having wrongdoers suffer in proper proportion to their desert in some rich and interesting sense of "desert"?  Can it even consistently constrain or limit punishment on the basis of such values?

        What do I mean by a rich and interesting sense of desert?  For the moment, let me simply tell you what I regard as an impoverished and uninteresting sense of the term.  If all one means by the claim that a person deserves to suffer punishment is that the person has been convicted of a crime (i.e., is legally guilty), then of course taking account of desert will be consistent with the harm principle.  Indeed, taking account of desert in this sense is no doubt required by the kind of rule-utilitarianism that is often used to defend the harm principle itself.   If this is all that one means by desert, however, then one is not offering the kind of retributivism that I seek to examine in this essay.  Although some philosophers once maintained that utilitarian and retributive theories could be reconciled by a showing that a practice of punishing (or telishing) the innocent would be disutilitarian, I am here interested in versions of retributivism that involve more than the simple demand that the legally innocent should not be intentionally punished.

        Kant was the kind of retributivist I have in mind.  He was one of the great philosophical defenders of the liberal state, and yet he claimed that criminal wrongdoers should be made to suffer in proper proportion to their personal moral blameworthiness -- what he called their "inner wickedness" (innern Bösartigkeit).  Can liberals consistently be retributivists of this sort?13

        One might think that this question, though perhaps of historical and abstract theoretical interest, has little relevance to our own system of criminal liability.  This thought, however, would be mistaken; for concerns with punishing "inner wickedness" are by no means absent in American law.  In many American states, for example, capital murder's mens rea requirement of "malice aforethought" may be implied from recklessness if a killer is said to have the mental state or character defect variously described as "an abandoned and wicked heart," "a depraved heart," "a depraved mind," "wickedness of disposition, hardness of heart, recklessness of consequences and a mind regardless of social duty," "wickedness of heart or cruelty," or (in the Model Penal Code) "extreme indifference to the value of human life."14   Even when a concern with inner wickedness does not find its way into the definition of the crime, it often arises dramatically when character is considered for purposes of sentencing -- particularly state level capital sentencing where such adjectives as "cruel," "heinous," and "depraved" loom large in characterizing aggravating factors.  Prosecuting attorneys, of course, will also attempt to nudge juries toward conviction by attempting to convince them that the defendant is a very wicked person.   Also worth noting here are those criminal statutes -- e.g., hate or bias crime statutes -- that (on at least one interpretation) seek to increase the penalty for certain crimes if those crimes were the product of evil motives -- e.g., racial hatred.  In short: our society, supposedly a liberal society, sometimes does in fact administer punishment in part on the basis of beliefs about the defendant's inner wickedness.   The focus upon such wickedness is no doubt sometimes simply as a predictor of dangerousness -- and is thus consistent with the harm principle -- but sometimes it operates with an independent life of its own.  The criminal may be punished primarily for breaking the rules, but we often feel free to give him some additional punishment if we think he broke the rules because of traits of character we find loathsome.  Is this kind of retributivism consistent with liberalism or does it represent a departure from liberal values?

        I think that this is a genuine and important worry.  Although some retributivists (e.g., Hegel) are not liberals, a great many -- both classic and contemporary -- are.  I have already mentioned Kant, and many contemporary defenders of retributivism (e.g., Herbert Morris, Michael Moore, and some of my previous selves) argue from a political perspective that strikes me as mainly liberal in nature.  Are such writers consistent in their endorsement of retributivism?

        Lord Devlin, following in the footsteps of James Fitzjames Stephen, thinks not.  Mill-inspired liberals condemn, for example, the use of state power to punish the "vice" of homosexuality and tend to ground their condemnation on the claim that it is not the state's business to address matters of personal vice and virtue.  But when these same liberals begin to discuss such matters as the grading of punishments, their own views of vice and virtue clearly enter the picture.  In his book Harmless Wrongdoing, Joel Feinberg summarizes the Devlin-Stephen challenge in this way:

The liberal allows no legitimate role to "the enforcement of morality as such" in the making of criminal law.   The only legitimate function of criminal law for him is to prevent private and public harms and nuisances.  In all consistency then, he should not permit any considerations other than the prevention of harm (and offense) to enter into decisions about the degree of punishment to be assigned to different categories of crime, and to commissions of the same crime by different offenders under different circumstances.   And yet it is our traditional practice, which not even the liberal would wish to alter, to treat greater moral blameworthiness (Stephen's term was "wickedness") as an aggravating factor and lessor moral blameworthiness as a mitigating factor in assigning punishment, a practice impossible to justify on the assumption that the aim of punishment, as of criminal law generally, is simply to prevent harmful behavior.  If the makers of criminal law can have no legitimate concern with wrongdoing as such, then neither should judges deciding punishment have any concern with morality independent of harmfulness.15

        The problem here will be particularly dramatic, I think, for those liberals who base their defense of the harm principle (or something like it), not on its supposed social utility, but upon some more general deontological principle.  Ronald Dworkin, for example, at one time argued that the state's refusal to use its coercive power to enforce judgments about the moral merits or demerits of individuals is definitive of the liberal state.16  Liberal governments may of course make value judgments about rights and interests worthy of protection.  Otherwise liberal governments could not even seek to prevent harm and indeed could have no legal systems at all.  A certain kind of neutrality is mandatory for them, however, if they are to respect the "right to moral independence" that is crucial to the general value of treating all persons with equal concern and respect.   Such governments must, Dworkin claims, "be neutral on what might be called the question of the good life [and thus] political decisions must be, so far as possible, independent of any particular conception of the good life, or of what gives value to life."17  The more robust versions of retributivism, however, seem (to put it mildly) to rest on pretty firm judgments about what makes one person more virtuous or vicious than another, and thus one might well wonder if liberal societies may consistently act on such judgments.   Liberals such as Mill and Hart do not think that it is legitimate to criminalize homosexuality simply because it is perceived as a vice, and yet they see no problem in punishing particularly vicious criminals more severely than those of greater virtue.   Is this not simply an inconsistency?  Are not Mill and Hart and Dworkin simply being unprincipled when they claim that the criminal law should be concerned only with harm and not with personal vice and virtue?  Is not their so-called principle simply a slogan that they trot out only when it serves their own sexually permissive ideology?

        In Harmless Wrongdoing, Joel Feinberg explores this charge of inconsistency and argues that, once a few important distinctions are drawn it can easily be met.18  In addition to arguing that the liberal may consistently limit criminal harms to wrongful harms (i.e., harms involving grievances or rights violations), he also offers an argument more directly relevant to our present inquiry.

        Enriching some suggestions of Hart, Feinberg argues that the liberal theory of punishment is a theory about the general justifying aim of punishment -- the primary version of the theory being that the general justifying aim is (roughly) the prevention of harm to others.  He then argues that it is perfectly consistent to suggest, as the liberal does suggest, that this institutional practice of harm prevention should be governed by fair procedural rules.  Thus, if it is unfair to punish people without regard to their moral desert, their personal blameworthiness, then the liberal may consistently preach against this.

        Feinberg's move here is very creative and goes a long way toward overcoming the claimed inconsistency.  Does it totally do the job, however?  I must confess that I retain some doubts. It certainly works for those factors (e.g., some mens rea requirements) that Feinberg calls conditions of responsibility.   I find it odd, however, even to use the label "procedural fairness" to refer to a concern with blameworthiness -- a concern that Feinberg quite rightly sees as involving motive and character and as thus distinct from responsibility.19   Also, I suspect that one's sense of "procedural fairness" is more affected by one's general justifying theory than Feinberg suggests.   Fairness involves respecting the rights of the accused, but one's general justifying theory may play a large role in defining just what those rights will be.   Of course if the harm-reducing system of criminal law is to function as a system of rules -- as a kind of price system -- it will require taking seriously such issues as whether the actor caused the harm intentionally or knowingly.  Thus a preference for a mens rea system over a strict liability system is, I think, not only consistent with liberalism but required by it.  Liberalism will indeed require this minimal level of procedural fairness (if one wants to call it that), for a system without this level of fairness would hardly be a system of rules at all and would be at least as scary and unpredictable as the state of nature.

        A mens rea requirement of intention or knowledge as a condition of responsibility, however, is a far cry from a requirement for moral culpability or blameworthiness in a rich retributive sense; and thus if the liberal has a fondness for the latter he may indeed be bordering on inconsistency,.  It is of course analytic that if one aim of punishment is to express moral blame of those who are truly blameworthy, then a procedure will not be deemed fair if it fails to capture moral blameworthiness.  But Feinberg is far more confident than I am that his well-known "expressive theory of punishment" can consistently be embraced in full by liberals -- that they can happily embrace his claim in Harmless Wrongdoing that "the criminal law is a great moral machine stamping stigmata on its products."20  Liberals can no doubt consistently approve of different degrees of stigma for different degrees of harm, but it is by no means clear that they can consistently approve of different degrees of stigma for different degrees of blameworthiness with respect to motive and character.  Feinberg does not always seem to realize that his endorsement of the expressive theory of punishment may not simply be a constraint on a harm-based theory but rather an endorsement of a new general justifying aim of punishment -- one whose consistency with the harm principle is not so obvious.21  Thus I am inclined to think that the Stephen-Devlin challenge remains to some degree unanswered.

        Feinberg himself makes it very clear that he has no sympathy with the view that the criminal law may be used to aggravate punishment on the basis of vices of character that are unrelated to any actions that might be a legitimate object of criminalization in a harm-preventing system.  "The liberal would [indeed] be inconsistent," he argues,

if he defended a rule making lascivious motivation an aggravating condition in the commission of crimes while staunchly opposing legislation creating independent crimes of lasciviousness..... [He] must not permit the types of blameworthiness which he excludes at the legislative level to sneak in the back door at the sentencing level [and thus] may use the criminal law to promote ... only those virtues that consist in the disposition to respect and promote the rights of others.22

        I agree with Feinberg that punishing for character defects unrelated to possible harm is inconsistent with liberalism; but, unlike Feinberg, I am not yet persuaded that liberalism may consistently target character even it it limits itself in the way Feinberg suggests.  If such character defects as a callous disregard for human life are a legitimate target at the sentencing level, then it would seem that there would be nothing wrong in principle with targeting them before they are realized in action; and yet most liberals, I think, would regard such early intervention as wrong in principle and not simply as unwise policy.  Also, suppose people manifesting a certain kind of stupidity or carelessness posed exactly the same threat of harm as people manifesting a certain kind of malice.  Most retributivists would, I think, argue for punishing persons in the second category more severely because they are worse people.   Is this kind of retributivism consistent with liberalism?

        I have argued elsewhere that versions of retributivism that are based on the Principle of Fairness (e.g., the version presented by Herbert Morris in Persons and Punishment23) face many serious problems but at least are consistent with a liberal theory of the state.24   Those robust versions of retributivism that involve deep notions of blameworthiness, however, probably are not.   Michael Moore, for example, wishes to base punishment in part on the offender's attitude toward his own wrongdoing -- increasing punishment if the offender is unrepentant or unremorseful, lacking (as Moore puts it) the "virtue of feeling guilty."25  This account, however, will be hard to reconcile with certain forms of liberalism, since it is not implausible to argue that a right of moral independence would include a right to be morally shallow, hard of heart, or unrepentant.  It seems, then, that the liberal must either abandon robust retributivist commitments (which some liberals will no doubt be happy to do but which many will not) or modify liberalism itself -- hoping that the modifications made in liberalism will not make it unrecognizable as liberalism.

        Let me put the point another way.  It is now common to distinguish between Dworkinian neutrality liberalism and Razian perfectionist liberalism -- the latter, unlike the former, employing some idea of the good or virtuous life in order to generate its theory of liberty.26  Neutrality liberals cannot consistently embrace strong forms of retributivism.  Perfectionist liberals perhaps can.27  But then perfectionist liberals would seem to be barred from arguing against the criminalization of sodomy on the grounds that the state has no business concerning itself with matters of personal vice and virtue.  Thus these liberals may have to consider abandoning strong versions of The Harm Principle -- a principle that many would regard as a cornerstone of liberal doctrine.  This is an issue I will explore more fully as I now move to a discussion of legal moralism and the liberal theory of fundamental rights.

V. Popular Sovereignty, Fundamental Rights, and the Human Good

        If one is committed to democracy as a form of government (as both Devlin and Mill are) then one seems to be committed to the ratification of majoritarian preference absent some compelling reason why it should not be ratified.  Thus why should not a majority preference against sodomy (if there is one) be ratified?

        The only way to block this move is, I think, to show that some fundamental (perhaps constitutional) right would be encumbered by letting the majority have its way on this matter -- as it would, for example, if the majority wanted to preserve its shared morality by silencing free speech on sodomy and jailing those who publicly sing the praises of these practices.  Surely the public advocacy of sodomy tends to undermine society's shared convictions on these matters far more effectively than does a private act of sodomy, and yet even Devlin is reluctant to prohibit the former because he seems to believe that a fundamental right of speech is involved.28  He must thus be assuming that when people are forbidden by the state from acting out their sexual preferences that this is comparatively trivial (perhaps like being forbidden to collect stamps or pursue some other form of recreation) and thus does not raise important constitutional issues or issues of fundamental rights in the way that curtailment of speech does.

        But this is controversial and one would like to see it argued.  It is perhaps true that liberals like to talk more than they like to immerse themselves in the high seas of erotic adventure, and this perhaps explains in part why the liberal's favorite sexual topic has been obscenity (which is largely talking about or otherwise portraying sex) and why On Liberty is a great free speech classic but not, except perhaps indirectly, a classic of sexual liberation.  But surely the issues deserve more profound reflection than this.  No one would think of having theoretical views on the rights of free speech without having developed accounts of the social and personal nature and importance of speech.  And yet nobody involved in the Hart-Devlin debate has seemed to think it important to bring to bear on the discussion any theories of sexuality -- its social and personal nature and importance.  (Even given a reasonable fear about AIDS, one should not recommend measures against homosexual activity without a clear conception of just what a person is being asked to give up if he is in effect being asked to give up the expression, even in private, of his sexuality.  Is this like giving up the freedom to pursue a hobby?   Or is it more like giving up free speech or free exercise of one's religion?)   The core issue is the criminalization of sodomy and everyone manages to discuss this without discussing sodomy, or even sexuality, in any detail.

        Well, actually, not everyone.  Here the justices of the United States Supreme Court have managed to do a better job than the professional philosophers.  In Bowers v. Hardwick,29 the leading homosexual sodomy case, the justices on both sides in effect write philosophical essays on the nature and value of homosexual sodomy as an example of sexual liberty.  They all seem to recognize that if sexual freedom is to be recognized as a fundamental right, it will not be because of its social or political value.  Free speech may be necessary for the proper workings of representative democracy, but sexual freedom is not.  Thus if sexual freedom is to have fundamental rights status, it will have to be primarily because of its value or meaning to the individual, not to society.  In this it will be like religious freedom, protected in large measure because of the role that religion plays in giving genuine meaning to the lives of those persons who are religious.30  

        To oversimplify a complex case, the essence of the philosophical disagreement between the majority and dissenting justices in Bowers v. Hardwick was this: the majority refused to regard general sexual freedom as a fundamental right because it saw no social or political value in sexual freedom outside of marriage or reproductive possibility and could not see such freedom playing a role in the lives of individuals comparable to the role of, say, religious practice.  They saw it essentially as (at best) recreational amusement or (at worst) perversion, and they refused to force Georgia to show a compelling interest before encumbering what is at best someone's recreational amusements.  The dissenting justices, on the other hand, while making no effort to argue for sexual freedom's social or political value, seemed to see sexual freedom as analogous to religious freedom in at least this way: it often forms a part of the very fabric of meaning and worth in an individual human life.  Often it is not simply a recreational amusement but is rather a part of what is involved in participating in those relationships that are among the crowning glories of human existence -- relationships of love and personal intimacy.  To ask one to give this up should, they argued, require a compelling state interest.

        Liberals tended to celebrate the opinions written by the dissenting justices, but was this celebration really consistent with their liberalism?  The dissenting justices did not reason in the manner of Mill.  They did not argue that sexual freedom is simply freedom and that it, like any other freedom or liberty, deserves fundamental rights status so long as it does no harm to others.   Rather they suggested (whether rightly or wrongly) that sexual freedom is special, a liberty worthy of special protection as a fundamental right, because it often forms a part of meaningful, virtuous, or -- in short -- good human lives.31  This approach would not seem to sit very well with neutrality liberals, and thus only perfectionist liberals would have any tendency to find it congenial.

        Even perfectionist liberals will have to pay a price, of course, if they accept this pattern of reasoning as a part of their liberalism.   The price will be that of having to listen to and take seriously arguments that certain liberties -- recreational drug use or pornography consumption, perhaps -- are not worthy of protection because they involve tendencies toward vice, tendencies toward bad or degraded or insufficiently human ways of living.  They will even have to consider such arguments on the issue of homosexual behavior if those who offer them can put them (as perhaps they can) in terms that are sufficiently secular to avoid constitutional problems.32  And once the virtue/vice arguments are weighed fairly, it is not obvious that all of the liberties that liberals traditionally have wished to protect will survive intact.

        On the other hand, immediate rejoicing from conservatives would be quite premature here.  Simply agreeing with conservatives that one might have to argue about issues of virtue and vice in developing a theory of fundamental rights in no sense commits one to the particular list of virtues that conservatives enjoy celebrating.  It is possible that some of the pagan virtues will win out over some of the Christian ones.  It is even possible that homosexuality may plausibly be conceptualized, not merely as an instance of a freedom to be tolerated, but as a form of virtuous living to be respected and even admired.  At the moment, the political theory of virtue is owned mainly by conservatives and religious sectarians -- an ownership that may liberals have been happy to concede while they have fought under the banners of neutrality and The Harm Principle.  Once they enter this unaccustomed arena, however, and start to hold up as models of virtue such persons as (say) Walt Whitman, it is hard to predict how the discussion might go.  My only point here is that we need to have this discussion and that liberals, if they want to protect a list of rights as truly fundamental, should stop resisting it.

        In summary: Many liberals like to rank liberties on a scale of importance -- some of only minor importance (and thus worthy of only minimal protection) and others of such fundamental importance that the state should be required to show a compelling interest before encumbering them.  Can an acceptable liberal list of fundamental liberties be constructed that does not draw on an account of human good or virtue -- or, at the very least, on some idea of what is truly important in a human life?  I suspect that the answer to this question is no.  Some liberties may be defended as fundamental simply because they are instrumental to free and democratic political institutions -- e.g., freedom of political speech may be defended in this way.  Other liberties, however -- non-political speech, free exercise of religion, sexual liberty -- cannot be defended in this way but must instead draw on some account of the good.33   (Here, at least, the right is not prior to the good.)   Once liberals acknowledge the dependence of certain fundamental liberties on an account of the good, however, they must be prepared to argue in the arena of the good -- not their accustomed arena.  Once in this arena, they might be able to give some of their favorite liberties a better defense than ever before.  They should also prepare, however, for the surprise of seeing the demise of some of their previously cherished commitments.

        One cherished commitment that might be vulnerable is The Harm Principle itself.   Although there may be practical reasons that will often weigh against the state seeking to promote virtue, what reason of principle would require society to limit its coercive powers to preventing rights violations when those rights themselves get their nature and importance from some human good they seek to protect?  I see no clear reason.

VI. Conclusion

        Devlin and Stephen have many positive conservative views about the proper organization of society, and it has been no part of my project here to defend any of those views.  I certainly have no interest here in defending the criminalization of homosexual sodomy.  I have rather been concerned with the challenges of inconsistency or hypocrisy that Devlin and Stephen raise against liberalism -- the challenge that certain principles invoked by liberals against them, particularly The Harm Principle, would, if applied fairly and uniformly, undermine some of the cherished doctrines of liberalism itself.  I have particularly been concerned with the tensions between The Harm Principle and two other principles: the Retributive Principle that moral blameworthiness is relevant to setting degrees of punishment and the principle (Fundamental Rights Constitutionalism) that some few liberties are so much more important than others that they deserve special protection at the constitutional level.   Both of these principles, I have argued, require that the state regard it as part of its legitimate business to concern itself with the human good -- with matters of personal vice and virtue.  Finally, I have argued that once the state recognizes such a concern as legitimate, there are no longer any clear reasons of principle why the state should accept the constraints upon its powers imposed by The Harm Principle.  It might be argued that the state should accept The Harm Principle as a matter of prudence or because of rule-utilitarian considerations.  But who has the relevant evidence here, and -- if we did have it -- would it really support Mill's strong form of The Harm Principle?  I am by no means confident that it would.

        Liberals may thus be faced with some unpleasant choices, for it looks as though some important liberal principles may have to be abandoned or modified if consistency is to be preserved.  Not all liberals will agree on which principles to sacrifice first, of course, and thus different forms of liberalism will no doubt emerge as a result of the choices made.34

        Exploring the details and relative plausibility of these various forms of liberalism would require another paper, and thus I shall here conclude the present essay simply by siding with Devlin and Stephen on one modest but important point: on any social issue as complicated as determining the proper scope of the criminal law, it is probably irrational to hope that The Harm Principle, or any other "simple principle," will allow us to draw a bright line between what is and what is not the state's legitimate business.  Skepticism over simplicity is usually in order, and I see no grounds for an exception here.

Appendix: Rejoinder to Professors Hampton and Morris35

        I am grateful to Professors Hampton and Morris for sending their comments to me in advance of the conference.  This gave me, alas, a chance to lament large portions of my paper; but it could have been much worse.  The last time that Professors Hampton and Morris subjected my work to public commentary (in that case a lecture on forgiveness and resentment), they tacitly suggested -- quite rightly I fear -- that my philosophical views probably grew out of certain defects in my character, specifically a lack of trust and generosity of spirit.  This time they charge me solely with intellectual errors; and, although I do not much like being charged with intellectual errors, I am very glad that they note no signs of further moral decay on my part.

        Speaking more seriously: Both Hampton and Morris have offered a rich and complex set of comments, far more than I could adequately address in the short space allowed.  I will simply try to indicate very briefly why, inspite of all their insights, I am still not ready to concede total defeat in my attempt to expose certain tensions within liberalism.  Both my commentators focus solely on the section of my paper dealing with retribution, assuming no doubt that the flaws in the final section speak for themselves.  Thus I too will speak here only about retribution.

        Let me sketch again the nature of my project: For most of my career I have been perplexed about the moral justification of criminal punishment.  A few years ago, however, I began to worry about punishment, not simply as a moral issue, but as an issue in political philosophy.  Specifically, I began to worry that the retributive account of punishment to which I was drawn on moral grounds might not be consistent with the generally liberal theory of the state to which I was also drawn.  The same tension I was feeling was, I thought, present in -- if not consciously experienced by -- Kant.  For Kant defended a general theory of the state that seemed to limit its proper role to the protection of rights -- what he called the domain of external freedom -- and yet advocated (at least sometimes) an account of punishment that aimed, like God's very own cosmic justice, at giving wrongdoers the level of suffering merited by what he called their inner wickedness, making sure that they received the pain that they deserved and no happiness of which they were unworthy.   That Kant was pulled in both of these directions kept me from being too embarrassed to feel comparable pulls (what better philosophical company could one possible have, after all?); and I even thought that I had a kindred spirit in some of Professor Michael Moore's writings on punishment -- for Moore had argued at the close of his important essay "The Moral Worth of Retribution" that, if he believed in God, he might not demand that the state administer retributive punishment -- thereby (I think) revealing that he too, like Kant and Murphy, saw some affinities between secular punishment and the moral objective hoped for in the old theological idea of divine retribution.

        Why am I sharing with you this piece of my own autobiography?  The answer is that I am assuming that others may share this tension in their own outlooks on punishment -- this tension between the moral and the political -- and might find my explorations of this tension of some interest and use.  My paper is thus written for kindred spirits feeling kindred tensions.

        Professor Hampton is manifestly not a kindred spirit here.  She seems to think that all that I have shown is that if one subscribes to an uncommonly silly view of retribution and an uncommonly silly view of liberalism, then a tension is present.  The solution to a tension generated by silly views, of course, as she rightly points out, is not to labor to resolve the tension, but simply to give up the silly views that generate it.

        Let me first focus on what she has to say about liberalism as I characterize it.  (I agree with Hampton, by the way, that the label "liberalism" has perhaps outlived its philosophical usefulness, but I continue to use the term here because I am commenting on writers -- Dworkin, Devlin, Hart, and Feinberg -- who make its use central to their own expositions).  I am a latecomer to contemporary political philosophy and am happy to be instructed in the discipline by those more expert than I, as Professor Hampton surely is.  On this particular issue, however, I think that Hampton has failed to instruct me because she has failed to understand me -- has even failed to understand one of the crucial liberal doctrines that she criticizes as unworthy of serious consideration.  This is Dworkin's doctrine of neutrality.

        Hampton is quite right to focus on this doctrine as essential to the development of my argument.  Having little sympathy with utilitarianism, I have little interest in defenses of The Harm Principle that are -- like Mill's -- utilitarian in nature.  Thus I want something of a Kantian nature -- a defense grounded in principle rather than policy -- and Dworkin's principle of liberal neutrality, generated from what he calls "the right of moral independence," struck me as a plausible ground to consider.

        Wrong, says Hampton.  Liberalism cannot accept the neutrality principle; indeed no sane political theory can accept it.   Why? Because, as Hampton argues at some length, the liberal state (like any state) must make value judgments about what rights and interests are worthy of protection, is thus deeply engaged in a moral endeavor, and thus cannot subscribe to a principle of value neutrality.  From these claims, with which I would not quarrel since they strike me as falling within that category John Cleese once called "the bleeding obvious," she draws the following conclusion: Dworkin's neutrality principle is too silly to take seriously as a foundation of liberalism or any other viable political theory and thus I should never have been tempted by it in the first place.

        But is this not simply a misunderstanding of Dworkin's neutrality principle and the use to which I put that principle in my paper?   Dworkin manifestly does not say that liberalism must be neutral about all values but only about questions of personal virtue or the good life.   Let me quote from my essay:

Liberal governments may of course make value judgments about rights and interests worthy of protection.  Otherwise liberal governments could not even seek to prevent harm and indeed could have no legal systems at all.  A certain kind of neutrality is [according to Dworkin] mandatory for them, however, if they are to respect the "right of moral independence" that is crucial to the general value of treating all persons with equal concern and respect.   Such governments must, Dworkin claims, "Be neutral on what might be called the question of the good life [and thus] political decisions must be, so far as possible, independent of any particular conception of the good life, or of what gives value to life."  The more robust versions of retributivism, however, seem (to put it mildly) to rest on pretty firm judgments about what makes one person more virtuous than another, and thus one might well wonder if liberal societies may consistently act on such judgments.36

        Dworkin's own version of the neutrality principle faces some serious and perhaps fatal problems -- some of them raised in the final section of my paper -- but the principle is not just silly.  The neutrality principle rejected by Hampton is silly, but it is not Dworkin's and it is not mine.  Hampton has extremely rich and interesting things to say about Rawls' version of a neutrality principle, but that also is not Dworkin's and not mine.  Perhaps there are too many senses of the term "neutrality" around, and thus perhaps it -- like "liberalism" -- is another term that may have outlived its philosophical usefulness.

        What about my version of retributivism?   Is it also too silly to take seriously?  Part of my project was to explore various senses of retributive desert in order to find out what sense or senses of retribution my liberalism would allow me to retain.  I noted five senses of desert: desert as legal guilt, desert as involving mens rea, desert as involving responsibility (capacity to conform conduct to the rules), desert as a debt owed to annul wrongful gains from freeriding (the Morris theory), and, finally, desert as involving ultimate character -- "wickedness" in some deep sense.  I argued that the first four are compatible with liberalism (some indeed are vital to avoiding what Morris calls the "terror" that would be present if we abandoned all notions of retributive desert), but I also argued that the fifth may not be compatible with liberalism.  But is the fifth, the idea of targeting deep character in sentencing, just silly?  I do not think that it is; and indeed I am inclined to join Kant in thinking that one tempting (though hardly conclusive) moral reason for believing in God is the hope for a being who will eventually apportion out pain and happiness in accord with true merit or worth.  I have, however, recently expressed the view (in a forthcoming essay) that such strong retributive objectives may be morally unacceptable for human beings to pursue.  Thus I am now tempted to abandon them, insist that sentencing judges do likewise, and thereby save my view of liberalism from this particular tension.37   It took me a lot of reflection to get to this point, however, and so I hope that the view I now reject was not just obviously silly in the first place.

        Let me now move briefly to Professor Morris's characteristically profound remarks.  He believes that my paper proceeds more by free-association than logical development, that I tend to collapse together claims that are actually quite distinct and with radically different degrees of plausibility, and that my ultimate concerns are far afield of the Hart-Devlin debate in which I try to locate them.  I must confess that I think that he is to some degree right about this, so what can I say but mea culpa?

        Where I remain unpersuaded is where Morris joins with Hampton in arguing that I fail to see that Joel Feinberg's fairness-based attempt to resolve the apparent inconsistency between liberalism and retribution does in fact succeed -- succeeds not just to the substantial degree that I admit but succeeds all the way down.

        Both Morris and Hampton remind us that Feinberg limits the scope of the criminal law to what he calls grievance morality and that, within the constraints of such morality, fairness is not only consistent with but demands that wrongdoers be punished in proportion to their blameworthy character so long as blameworthy character is understood in such a way that it is limited to "dispositions to feel and act in ways condemned by grievance morality" or "dispositions to violate the rights of others."  Morris writes: "Two individuals, both first-degree murderers, might differ significantly in their degree of blameworthiness -- say a contract killer and a mercy killer.  It confounds moral judgment and offends our notions of fairness to  punish them equally."  He then says this about remorse and repentance: "A repentant person at the time of punishment is less blameworthy than an unrepentant one" and this should "earn the criminal points, for a downward departure [in severity of sentence]."  And finally he writes: "Degrees of blameworthiness, as taken into account by the law and treated as relevant in sentencing determinations, reflect degrees of commitment to the legal norms."

        I find all of this problematic.  I grant the importance of grievance morality at the core of the criminal law, and see no problem in analyzing The Harm Principle in terms of grievances or rights violations.  I also grant the importance of punishment as a kind of debt owed to one's fellow citizens -- a model brilliantly explicated in Morris's essay "Persons and Punishment"38 in terms of annulling the wrongful gain that the criminal gets by freeriding on a mutually beneficial scheme of social cooperation.  What I still do not see, however, is how this helps in justifying the targeting of deep character, even at the sentencing level.   If I am maimed unjustly, I do indeed have a grievance.  But is my grievance greater if I am maimed by a botched attempt to kill me for profit instead of by a botched attempt to meddle in my life for more altruistic reasons?  Is my grievance any less if my assaulter finds Jesus, repents, and there improves his moral character?  (He may be less evil, but is my own grievance any less?)  And in what way is the altruistic meddler, to use Morris's language, "more committed to the relevant legal norms" than the contract killer?  And why does the altruistic meddler owe less of a debt to society for violating its norms than the killer motivated by greed?  And if the concept of blameworthiness demanded is, as Morris says, a "disposition to violate the rights of others" or a "disposition to violate the rules of grievance morality," do not both of these individuals have exactly that very same disposition (although for different motives) and thus merit the same punishment?

        Hampton, defending Feinberg in a manner similar to that developed by Morris, insists that "what makes a state liberal ... is its rejection of the idea that any enforcement of moral behavior should include punishment of immoral behavior which nonetheless has no victim other than the offender himself."   But if a sentencing judge, after giving the criminal a punishment properly proportional to the injury he has inflicted, adds on a little extra punishment because of the defendant's vicious character -- his smug unrepentance, say -- is that not simply to punish him for an aspect of his character that has no victim?  If so, is not the liberal who approves this inconsistent when he crusades against using the criminal law against victimless immorality?

        The fact that I have the formidable forces of Feinberg, Hampton and Morris all aligned against me on this issue makes me feel a deep anxiety that there is something important that I am missing.  One thing I certainly am not missing, however, is the platitude that we must distinguish between issues of justifying aim and issues of procedural distribution with respect to punishment.  Of course I see that and did not need my commentators to remind me of it.  What I do not see is how that distinction along with some simple notion of fairness can be used to overcome all of the tensions I find in the sentencing practices current in liberal societies and apparently approved of by many liberal theorists.

        Imagine a system of punishment that exists to protect rights by enforcing a system of rules that prohibits rights violations.  To be fair, this system must punish only those who have a fair opportunity to conform their conduct to those rules and thus must operate according to procedures that require legal guilt, mens rea (at least negligence), and such conditions of responsibility as voluntary behavior and sanity.  If it does all this, it seems to me that it has satisfied all demands that could reasonably be conceptualized under the heading of "procedural fairness."  As presently specified, the system fails to target certain aspects of character that some liberals (Moore and Morris, for example) think it appropriate for judges to target at the level of sentencing -- e.g., an absence of remorse and repentance.  A failure to target these aspects of character may be deplorable for many reasons, but what is unfair about it?  Of course it is always unfair to ignore a relevant differentiating characteristic, but in what sense is this a relevant differentiating characteristic in the liberal system here specified?  It would, of course, be unfair to ignore such factors in a system that exists to target them in addition to targeting the responsible creation of grievances.  The legitimacy of such targeting is the very point of the debate, however -- a debate obscured rather than illuminated by the suggestion that its resolution will emerge from a consideration of the value of fairness in procedure.

NOTES

*Regents' Professor of Law and Philosophy, Arizona State University.  B.A. 1962, John Hopkins University; Ph.D. 1966, University of Rochester.  This essay was prepared for presentation at a conference honoring Joel Feinberg held at the University of Arizona College of Law on September 30-October 1, 1994, and I dedicate the essay to him with esteem and affection.   My commentators were Jean Hampton and Herbert Morris.  I had hoped to revise the paper in the light of their comments and the comments of others who were kind enough to read and discuss the paper with me.  Unfortunately, I can see no way to make extensive changes in the paper that would not undercut the essays by my two commentators -- essays that will also appear in this volume.  Thus, I see no fair course except to publish here the paper pretty much as it was presented at the conference, making only those changes that, in my judgment, could not have a bearing on the essays by Professors Hampton and Morris.  I received very helpful comments from the members of the Arizona State University Moral, Political and Legal Philosophy Discussion Group and from Professor Robert P. George of Princeton University.  Some of these persons made comments that deserve treatment in the paper, and I hope that they will realize that my failure to deal with them is not based on a failure to appreciate their merits but rather on the logistical problems outlined above.  They will no doubt see their influence in my future work.   Even in the present version, the influence of Robert George is obvious.  His fine book, Robert P. George, Making Men Moral: Civil Liberties and Public Morality (1993), and the comments he made on this essay have, I must admit, made me very nervous about my commitments to liberalism.   [BACK]

1.  James Fitzjames Stephen, Liberty, Equality, Fraternity (University of Chicago Press 1991)(1874); Patrick Devlin, The Enforcement of Morals (1965)(containing a revised version of Devlin's 1959 Maccabaean Lecture "The Enforcement of Morals" and his response to Hart and other critics.); H.L.A. Hart, Law, Liberty and Morality (1963). [BACK]

2.  I have no intention of citing this essay, since I wish for it to remain in the obscurity to which it has justifiably been consigned.  [BACK]

3.  Joel Feinberg, "Some Unswept Debris from the Hart-Devlin Debate," 72 Synthese 249-275 (1987); Joel Feinberg, "Harmless Wrongdoing," 4 The Moral Limits of the Criminal Law 124-175 (1988).  [BACK]

4.  Joel Feinberg, "Some Unswept Debris from the Hart-Devlin Debate," 72 Synthese 249 (1987). [BACK]

5.  Id.  [BACK]

6.  As Feinberg points out in "Harmless Wrongdoing," Devlin's exposition of his view is often unclear -- leaving the reader uncertain of the exact nature of his "legal moralism."  Feinberg, "Harmless Wrongdoing," supra note 3, at 124-25.  [BACK]

7.  This worry applies for U.S. law, of course.  Devlin, however, lives under a legal system that does not have a constitutional prohibition against the establishment of religion.  I do not, by the way, mean to suggest here that it is always easy to draw the secular/religious distinction.  The very existence of natural law theory, for example, suggests that any simple way of attempting to draw the distinction will probably be inadequate.   [BACK]

8.  This issue of sodomy and virtue is actually of some complexity.  I will explore it in a bit more detail later in the paper.  [BACK]

9.  Suppose that many or even most people find that they are repelled or disgusted by homosexuality.  It does not follow from this that their repugnance or disgust is moral in nature.  That would require a separate argument.  One could argue that a majority of citizens in a democracy have a right to enforce their strong moral convictions but that they do not have a right to enforce all of their strong convictions (regardless of their basis).   Suppose it is true that shared strong moral preferences bind society together, and that for that reason the law may enforce them.  Do shared strong aesthetic preferences, for example, bind society together in a similar way?  I do not, of course, mean to suggest that the moral/aesthetic distinction is always easy to draw.   [BACK]

10.  Homosexuality may be a particularly unpersuasive example of sexual conduct that, if tolerated, would threaten family and thus social stability.  If homosexual inclinations are mainly a matter of genetics, and if such inclinations tend to occur only in a minority of the population, then toleration here would presumably have less extensive ramifications than if directed toward inclinations more widely shared.  It does not strike me as absurd to suggest that the sexual revolution of the sixties -- and the resulting freedom men felt to abandon family responsibilities for "self-fulfillment" -- generated considerable social harm, particularly to women and children.  It is much harder to link up homosexuality with this particular piece of social decay, however -- unless one thinks that the problem was simply excessive sexual freedom in general and that homosexuality falls under this description.  [BACK]

11.  The first is explicitly discussed in Devlin's text, supra note 1, at 124-39.  I am building a discussion of the second from some remarks he makes in passing.  [BACK]

12.  Ronald Dworkin, A Matter of Principle 181-204 and 335-372 (1985).  [BACK]

13.  Kant's views on punishment are actually more complex and ambiguous than this.  Here I am simply picking up on the one thread from his various discussions of the topic.  See, Jeffrie G. Murphy, "Does Kant Have a Theory of Punishment?," 87 Colum. L.Rev. 509 (1987).  [BACK]

14.  Joshua Dressler, Understanding Criminal Law 461 (1991).  [BACK]

15.  Joel Feinberg, Harmless Wrongdoing, supra note 3, at 145.  [BACK]

16.  Ronald Dworkin, A Matter of Principle 181-204 and 335-372 (Harvard University Press 1985).  Some of Dworkin's more recent writings suggest that he has to some degree changed his views on these matters.  [BACK]

17.  Id. at 191.   [BACK]

18.  Joel Feinberg, Harmless Wrongdoing, supra note 3, at 147-51.  [BACK]

19.  Even using the phrase "procedural fairness" to refer to conditions of responsibility strikes me as a bit odd.  Strict liability is certainly unfair, but procedurally unfair?   The unfairness of strict liability, even in a harm preventing system, strikes me as involving the unfairness of holding people to rules to which they lack substantial capacity to conform -- something to be avoided, through mens rea requirements, in any just system of substantive criminal law.  A system that fails to take account of motives or character, however, may also be criticized -- but on grounds of unfairness?   It will, of course, be unfair -- perhaps even procedurally unfair -- to ignore these factors in a system that claims to target them rather than mere harm.  The legitimacy of such targeting is the very point of the debate; however, a debate obscured rather than illuminated by the suggestion that its resolution will emerge from a consideration of the value of fairness.  [BACK]

20.  Joel Feinberg, Harmless Wrongdoing, supra note 3, at 150.  [BACK]

21.  Hart, in his discussion of the grading of punishments and the moral judgments of denunciation expressed by such grading, trades on an ambiguity that may tempt Feinberg as well.  Hart, supra note 1, at 36-37.  He does not seem to see the difference between adjusting punishment in accord with the gravity of the offense and adjusting punishment in accord with the wickedness of the offender.  the former surely is compatible with liberalism, but it is not at all clear that the latter is.  [BACK]

22.  Feinberg, "Some Unswept Debris from the Hart-Devlin Debate," supra note 3, at 259-260.  [BACK]

23.  Herbert Morris, "Persons and Punishment," 52 The Monist 475-501 (1968).  [BACK]

24.  See my forthcoming article, Jeffrie G. Murphy, "The State's Interest in Retribution," J. Contemp. Legal Issues (forthcoming), an essay upon which I have drawn in writing the present section of this essay.  In recent years, I have come to have some serious doubts about the strong versions of retributivism that I was once inclined to defend.  I have expressed some of these doubts in the forthcoming essay noted above and also in my "Cognitive and Moral Obstacles to Imputation," Jahrbuch für Recht und Ethik 67-79, Band 2 (1994).  See also my "Retributivism, Moral Education and the Liberal State," 4 Crim. Just. Ethics 3-11 (1985), reprinted in Jeffrie G. Murphy, Retribution Reconsidered 15 (1992).  [BACK]

25.  Michael Moore, "The Moral Worth of Retribution," in Responsibility, Character and the Emotions 179-219 (Ferdinand Schoeman ed., 1987).  [BACK]

26.  For a rich discussion of various forms of liberalism and their bearing on theories of punishment, see Jean Hampton's forthcoming essay "Retribution and the Liberal State," J. Contemp. Legal Issues (forthcoming).  See also Joseph Raz, The Morality of Freedom (1986); and Vinit Haksar, Equality, Liberty and Perfectionism (1979).  [BACK]

27.  Moore is probably best interpreted as a perfectionist liberal.  [BACK]

28.  Devlin never talks explicitly about fundamental rights as limiting democratic rule.  I draw the conclusion that he holds that there are such rights, however, from some things he says about the free exercise of religion and from his claim that an environment of open discussion will give those who think of themselves as advanced thinkers on such issues as homosexuality an opportunity to change the shared morality that, before it is changed, may justify criminal prohibitions of what it condemns.  [BACK]

29.  Bowers v. Hardwick, 478 U.S. 186 (1986). [BACK]

30.  The analogy with free exercise of religion is, for better or worse, mine.  It does not occur in any of the opinions in Bowers v. Hardwick.  I use it simply as an example of a liberty that is worth protecting, not primarily because of its social and political benefits, but because of its genuine value to those individuals who would exercise it.  I use the phrases "genuine meaning" and "genuine value" because I have in mind an objective standard here.  To protect religion as a fundamental right requires, in my view, that religion actually is an important human good -- not merely that many people take a kind of personal satisfaction in being religious.  (Religion is not protected because it is a great hobby.)  I would argue in a similar way about sexual freedom -- that those forms of it worth protecting at the level of fundamental rights must be defended as a part of a genuinely good, and not merely satisfying, human life.  I think that it is a serious business for the state to interfere with human satisfaction in any form, and thus I would want to require that the state have a good reason to justify such interference.  I would not, however, wish to impose upon the state the extreme burdens of compelling state interest/least restrictive means unless a right identified as fundamental is being encumbered.  I would save this test for liberties of the deepest importance.
        For an argument that free exercise of religion -- and other important constitutional liberties -- require an objective account of the human good served by those liberties, see George, supra note*.   [BACK]

31.  Note the difference with Texas v. Johnson, 491 U.S. 397 (1989).  In this flag burning case, several justices who upheld Johnson's right to burn a United States flag made it quite clear that they believed that Johnson's behavior was loathsome.  They certainly did not conceptualize it as a part of a good life.  [BACK]

32.  In Sexual Desire, Roger Scruton argues that male homosexuality is perverted (in the sense of incomplete or deficient) because it fails to embrace the mystery and risk of coming to terms with someone who, since of a different gender, is totally other.  Roger Scruton, Sexual Desire (1986).  There is, of course, no necessary disutility or rights violation in conduct that is perverted in this sense, and thus theories of rights or justice or utility would see such perversions as morally irrelevant.  Theories of virtue, however, might regard it as highly relevant.  I mention Scruton here not because I agree with him (I am by no means sure I even understand him) but simply to illustrate the possibility of having a debate on the virtue of homosexuality that is not based on religious dogma.  [BACK]

33.  I have argued that Kant, in his essay "Theory and Practice," held this view about the free exercise of religion.  See my forthcoming "Kant on Theory and Practice," 36 Nomos (forthcoming 1995).  [BACK]

34.  Hart, for example, wants to supplement the harm principle with a principle of justified paternalism.  Hart, supra note 1, at 30-34.  Paternalism, of course, is coercing a person for his own good, and it is difficult to see (as Devlin points out) how Hart could develop a concept of a person's "own good" that did not draw upon the very non-neutral conceptions of the morally good or virtuous life that Hart seeks to oppose in other contexts.  See also the rich accounts of punishment developed in Herbert Morris, "A Paternalistic Theory of Punishment," 18 Am. Phil. Q. 263 (1981), and Jean Hampton, "The Moral Education Theory of Punishment," 13 Phil. & Pub. Affairs 208 (1984).  As I have argued in "Retributivism, Moral Education and the Liberal State," supra note 24, I think that Morris and Hampton are on to some very important insights in these essays.  What is not so clear to me, however, is the degree to which these insights are compatible with political liberalism.  [BACK]

35.  This rejoinder is responsive to the comments as presented at the conference by Professors Hampton and Morris.  I have no idea what changes, if any, they may have made in the versions of their comments to be published in this volume.  (Professor Hampton has suggested to me that she may make some major changes in her comments.)  I hope, however, that my rejoinder is still to some degree responsive or, if not, that it at least helps to clarify my views.   [BACK]

36.  Supra p. 81.   [BACK]

37.  The tension I cannot resolve is that explored in the section of the paper dealing with fundamental rights.  [BACK]

38.  Herbert Morris, supra note 23.  [BACK]