Dean Rostow’s The Sovereign Prerogative is a collection of essays on legal topics as apparently unrelated as the doctrine of “judicial restraint” and the government’s security program. Although Dean Rostow displays a certain hostility to “academic” philosophers—in particular, it would appear, to Professor Hart and to Mr. Richard Wollheim—the unity of his book derives from the unity of his own, fairly familiar “philosophy.” Dean Rostow is an American legal realist and, I take it, an American liberal. As a realist Dean Rostow deplores legal formalism and legal positivism—for he takes them to encourage an undesirable separation of law and morals, and a faineant judicial philosophy. Judges, he insists, cannot simply apply rules; they must exercise that “sovereign prerogative of choice” of which Holmes wrote, and in so doing they will inevitably assert moral and political preferences. As an American liberal Dean Rostow would have the judiciary generally, and the Supreme Court in particular, assert what he takes to be liberal values, and his attitude toward the Court reflects his judgment of its fidelity to them. In what is surely the ablest essay in his collection Dean Rostow indicts the Court for its war-time acquiescence in the relocation of Japanese American citizens. Normally, however, he is complacent about the performance of the Court, and will defend the adequacy of its reasoning in the White Primary, the Restrictive Covenant and the School Segregation cases against all comers—Northern liberals as well as Southern bigots. The legalistic scruples expressed by liberal critics such as Professor Wechsler are traced directly to inadequacies in their legal philosophy.

Although the legal realists have long since convinced everyone that judges or at least judges in courts of last resort—cannot simply apply rules, Dean Rostow continues to suppose that judges and jurisprudential thinkers who disagree with his doctrines (and they include some of the most sophisticated legal thinkers of our time) have failed to learn their realistic lessons. Consequently, he never tires of invoking the ancient realist texts, and he appears to accept the alternatives of yesterday’s polemics as a fair formulation of the issues in today’s philosophy of law. While he claims that the Yale Law School, over which he presides, is the hottest bed of legal realism, he draws his arguments against opposing conceptions of the judicial process from those Harvard oracles, Oliver Wendell Holmes and Roscoe Pound. In The Common Law (1881) Holmes opined that “the life of the law has not been logic: it has been experience,” and in “Mechanical Jurisprudence” (1908) Dean Pound concurred: Law must be judged “by the results it achieves…, not by the beauty of its logical processes or the strictness with which its rules proceed from the dogmas it takes as its foundation.” If the judge cannot decide on the basis of “logic” (however beautiful) he must decide on the basis of what the realists are given to calling “values.” In their early debunking phase the realists—they have never been very clear about the distinction between causes and justifications—held that the judge’s breakfast, or what Dean Rostow calls his “private value preferences,” might determine his decision. (Indeed, as Yosal Rogat has said, “Anxious to go as far back from the facade as possible, some inevitably reached the back stairs.”) But the contemporary realist is no debunker, and Dean Rostow espouses an “affirmative” version of the doctrine. He would have the judge govern his decisions—and I offer a partial list—by the “reasons” for existing rules, immanent social ends or desires, the ideals projected by the Constitution, inescapable ethical ideals, the needs of the time, the ends (justice and liberty) or the end (justice) of the law. It is supposedly this commitment to values, conscious or unconscious, which permits the judge (since he cannot simply follow rules) to exercise his sovereign prerogative of choice. But Dean Rostow has very little to say about how the judge is to choose among these possibly conflicting, and often political, values. And this is a choice he must make, if he is to choose. In the end, Dean Rostow suggests an appeal—as dubious as intellectual history as it is unhelpful as legal philosophy—to “Bergsonian” intuition.

It is, then, not because they are affected by a “positivism” Dean Rostow irrelevantly imputes to them, but because they believe that the Constitution requires political conflicts to be settled by political means (and not by judicial intuitions) that the late Judge Hand and Justice Frankfurter adopted the “monkish” doctrine of judicial restraint. To be sure, judges “make law” and “legislate”; but it does not follow that they must make those particular laws which are more appropriately made by other branches of government. Nor is it the case, as Dean Rostow avers, that each judicial opinion “derives its validity [from the judge’s] conscientious effort to make certain that it represents not only law, but ‘good law.”‘ Surely, the distinctive task of the judiciary is, as Judge Hand thought, to interpret and apply the law; it should act neither as a third legislative body nor as a company of Platonic guardians decreeing “good law.” It is precisely the sort of suggestion Dean Rostow makes here that ignited Holmes’ famous explosion: “I hate justice.” And if Judge Hand, in the Dennis case, and Justice Frankfurter, in the Flag Salute cases, failed to legislate values to which they were personally committed, it is doubtless because they had taken the point of Holmes’ “great dissent” in the Lochner case. They do not see the ground—and I cannot see that Dean Rostow has provided one—for objecting when “conservative” judges legislate Herbert Spencer’s Social Statics into the Constitution if “liberal” judges are to perform the same service for the preferences of the ACLU, the NAACP and the CIO. It is not necessary to believe that the Court is an inherently undemocratic institution—some of Dean Rostow’s best pages are devoted to the refutation of that notion—in order to reject his “affirmative” conception of the judge’s role. For Dean Rostow contends that the law (as opposed to the family?) “is the central institution of a changing society”; and he urges that the Court (the deus ex machina of the American tragedy) become “the proponent and protector of values which are premises, goals, needs, and ambitions of our culture, as they have been expressed in our living constitution.”


It is precisely because he thinks that the Supreme Court has too often heeded Dean Rostow’s summons that Professor Wechsler delivered a controversial protest in his Oliver Wendelt Holmes Lecture, “Toward Neutral Principles of Constitutional Law,” at Harvard in 1959. Dean Rostow calls for a “result-oriented, sociological jurisprudence,” and Professor Wechsler has protested that the Supreme Court is “result-oriented” to the point of rendering decisions on ad hoc and politically interested bases which are incompatible with such standards of neutrality and generality as are implied in the very notion of law. It is possible to think that Professor Wechsler has not properly articulated these standards (their articulation is a primary and difficult task of jurisprudence) and yet to agree with his judgment that the Supreme Court’s decisions in the White Primary, the Restrictive Covenant and the School Segregation cases do not satisfy the standards he has properly in mind. (How, for instance, can the Court hold to the principle it exploits in the case of Shelley v. Kraemer? If the Supreme Court prohibits judicial enforcement of discriminatory restrictive covenants solely because a state court acts, must it not also prohibit the judicial enforcement of discriminatory wills, and of the rights of private property against trespassers if the grounds of exclusion are discriminatory?) Dean Rostow supposes that Professor Wechsler’s ideals, far from being implied by the very idea of law, are incompatible with the very nature of the judicial process. But Professor Wechsler is not in fact committed to the kind of mechanistic and formalistic philosophy Dean Rostow habitually assigns to his opponents. Professor Wechsler is willing to admit that within limits the judge must choose between values if he is to choose at all. But he quite rightly insists that such choices must be justified by principles susceptible of application “across the board.” And it is surely curious that while Dean Rostow proclaims Professor Wechsler’s standards impossible to fulfill, he suggests how the decision in Shelley might have been written so as to meet those very standards. If such standards can be met they ought to be met, and Professor Wechsler’s concern about the recent performance of the Supreme Court ought not complacently to be dismissed or assigned to the supposed inadequacies of his legal philosophy. Indeed, two of Dean Rostow’s colleagues in that hot-bed of legal realism have themselves written that “The Court’s product has shown an increasing incidence of sweeping dogmatic statement, of the formulation of results accompanied by little or no effort to support them in reason, in sum, of opinions that do not opine and of per curiam orders that quite frankly fail to build a bridge between the authorities they cite and the results they decree.” These realities cannot be hidden forever behind the fig leaf of Dean Rostow’s legal realism.

So long as one grants a distinction between the role of the judge and that of the legislator, the question of the extent to which the judge ought to “legislate” what he takes to be the moral aspirations of society must be distinguished from the question of the extent to which the moral convictions of society quite generally ought to be given the force of law. In his Stanford lectures on Law, Liberty, and Morality H. L. A. Hart, who is Professor of Jurisprudence in the University of Oxford, presents an acute critique of the persistent doctrine that morality “as such” ought to be given the force of law. These lectures are in part an expansion of his earlier examination of the views of Sir Patrick Devlin (views which have been defended by Dean Rostow), and they serve to reinforce my opinion that Professor Hart is quite without peer among writers on jurisprudence in the English-speaking world.


In On Liberty Mill argued that “the only purpose for which power can rightfully be exercised over any member of a civilized community against his will is to prevent harm to others.” And to identify the many different purposes he meant to exclude he added, “His own good either physical or moral is not sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because in the opinions of others, to do so would be wise or even right.” In 1957, the Wolfenden Committee, appointed to consider the criminal law in relation to homosexuality and prostitution, recommended to the English government—on grounds strikingly similar to those urged in On ‘Liberty—that homosexual practices between consenting adults in private should no longer be a crime and that prostitution should not itself be made illegal, although legislation should be passed “to drive it off the streets” on the ground that public soliciting is an offensive nuisance to ordinary citizens. In Liberty, Equality, Fraternity, the great Victorian judge, James Fitzjames Stephen, took issue with the principles of On Liberty, and, a century later, in his Maccabaean Lecture on The Enforcement of Morals, Lord Devlin, who sits on England’s highest court, challenged the Report’s contention that “there must be a realm of morality and immorality which is not the law’s business.” Stephen thought he had established against Mill that the law should “be a persecutor of the grosser forms of vice”; and Lord Devlin argues against the Report that “the suppression of vice is as much the law’s business as the suppression of subversive activities.” (Dean Rostow, who thinks Lord Devlin made a rhetorical error in introducing his analogy between homosexuality and treason, nevertheless acknowledges his own indebtedness to Professor Helen Silving “for the observation that in psychoanalytic terms sex, crime and treason are related.”)

In an argument whose power and finesse can hardly be suggested in a review, Professor Hart undertakes to vindicate the tradition of Mill and the Wolfenden Report against the “legal moralism” of Stephen and Devlin. He does not, to be sure, share Mill’s view that only “harm to others” can justify the law’s intervention. He recognizes with his opponents that the criminal law has never (except in the case of rape) regarded the victim’s consent as a defense. Thus, euthanasia is still regarded as murder. But it is one thing to justify legal paternalism and another to justify legal moralism. Because one grants that the law may under certain circumstances protect a person against himself, it has not been conceded, as Lord Devlin seems to think, that the function of such law is “to enforce a moral principle and nothing else.” An analogous distinction is required in the case of bigamy. A law against bigamy might be justified on the ground that bigamy is a public nuisance without conceding that it is being punished as either irreligious or immoral. (When such distinctions are drawn it will be plain that the case of homosexuality is utterly different from that of bigamy to which Dean Rostow assimilates it. For in the case of homosexuality an urgent instinct is suppressed, and it is more than implausible to argue that private sexual relations constitute a public nuisance.)

Having shown that the existing criminal law on euthanasia and bigamy does not imply that the law enforces morality as such, Professor Hart turns to the positive arguments offered for the view that the law ought to enforce it. He distinguishes the moderate thesis that the enforcement of morals is necessary to society’s existence from the extreme thesis that the enforcement of morality, and its preservation, are valuable apart from this beneficial consequence. He argues against the moderate thesis, which he associates with Lord Devlin, that it is either “Utilitarianism without benefit of facts” or that it is based on a confused definition of society (which identifies it with its morality and thereby guarantees the thesis as a necessary truth). He urges against the extreme thesis, which he associates with Stephen, that it asks us to restrict freedom, and to inflict the misery of punishment on human beings for the sake of values which seem to belong “to the pre-history of morals and to be quite hostile to its general spirit.” These values include, he says, a mere conformity to moral rules induced by fear; the gratification of feelings of hatred for the wrongdoer; retributory punishment, even where (as in the case of adult homosexuality) there is no victim to be avenged; and the infliction of punishment as a symbol or expression of moral condemnation. These, and the mere insulation from any change in social morality, however repressive or barbarous, are the high purposes of the extreme version of the doctrine that the law ought to enforce morality “as such.”

It is not surprising to discover that Dean Rostow, who regards the law “as an instrument for carrying out the moral purposes of our tradition,” regards Professor Hart’s restraint as a manifestation of the same misguided attempt to separate law and morals that he discerns in Judge Hand and Justice Frankfurter. Nor, in view of his cavalier attitude toward Professor Wechsler’s demand for the rationally satisfactory justification of judicial decisions, is it wholly surprising to discover his indifference to Professor Hart’s concern lest the sheer fact that attitudes are enforced by “morality” be taken as a justification for their enforcement by law. (Professor Hart’s devotion to rational principles strikes Dean Rostow in any case as an anachronism in a period which has—as he somewhat incredibly and quite irrelevantly tells us—been taught the power of the irrational in human affairs “by Montesquieu and Sir Henry Maine, Max Weber, Durkheim, Ehrlich, Spencer and Buckle.”) What is genuinely surprising, and seriously disturbing, is Dean Rostow’s assumption that a thinker such as Lord Devlin, who says that “it is not possible to set theoretical limits to the power of the State to legislate against immorality,” can stand in the “main line of English liberal political theory.” For Dean Rostow must know that it is precisely against Devlin’s view that Mill took up his pen. Indeed, the juxtaposition of these books by Dean Rostow and Professor Hart puts in the gravest doubt the common American assumption that legal realism is the natural ally of political liberalism, and legal “positivism” its natural antagonist. For Professor Hart displays a comprehension of liberal principles, and a commitment to liberal purposes quite beyond anything to be found in The Sovereign Prerogative. And Professor Hart’s defense of the Wolfenden Committee’s Report serves as a melancholy reminder of one of the inexcusable defections of the American liberal movement, even as his qualities of mind and style put to shame the prevailing standards of jurisprudential writing in this country.

This Issue

June 1, 1963