Perspectives in Political Science

Fall 1996

pp. 183 -191


by Peter Berkowitz

Virtue is back in fashion, and not only in the academy where liberals and communitarians, feminists and civic republicans, neo-Aristotelians and natural law theorists are busy expounding rival and in many ways incompatible catalogues of virtue. Popular culture, too, is exhibiting a lively new interest. 1995 saw the release of no fewer than three film adaptations of Jane Austen novels, novels whose charm consists in Austen's remarkable ability to display the complexities of individual character, and the dependence of happiness on the unending contest between virtue and the vicissitudes of fortune. A certain hunger for virtue could be seen as well during the autumn of 1995 in the groundswell of bipartisan enthusiasm for a Colin Powell presidency; for what appeared to make Powell preeminently well-qualified in the eyes of so many to hold the highest office in the land was his manifest decency and integrity. And since the modern president always has his finger on the pulse of the people, it is significant that in his 1996 State of the Union Address, President Clinton kicked off his campaign for a second term by returning to the New Democrat themes that served him so well during the campaign of 1992: the political importance of personal responsibility and the nonpolitical institutions such as family, religion, and civic association that foster it.

Virtue's new found popularity of course reflects a certain anxiety and is a telltale sign that all is not well with liberal democracy in America. The growing interest in virtue is in part a response to the widespread perception that public life in the United States has coarsened and is steadily getting worse. It is also a reaction to the increasingly common fear that the institutions and associations in America responsible for making men and women moral are unraveling before our eyes. The moral life, in short, seems to be under assault. And what has complicated efforts to defend it is that liberalism, once the pride of the American political tradition, has come to be seen as not only one of the prime culprits but as aggressively leading the charge.

In its contemporary form-the form characteristic of the left wing of the Democratic Party-liberalism is associated with the goals of securing the social and economic bases of equality and protecting an expansive domain of personal freedom. The fundamental premise of liberal thought-the natural freedom and equality of all human beings-need not issue, as it has in contemporary liberalism, in hostility to the political claims of moral and intellectual virtue. Still, the fact remains that in promoting equality, contemporary liberalism has carelessly cast aspersions on the very idea of human excellence; and in seeking to expand the domain of personal freedom, contemporary liberalism has recklessly gone beyond the principle that it is not government's business to declare what freedom is good for to declaring through government action that in principle what freedom is good for is not only unknown but unknowable. In the climate of opinion created by these excesses, the very concept of virtue has come to be seen as inconsistent with the recognition of equality and the protection of personal freedom.

Oddly, some of liberalism's proponents have made common cause with its critics to insist on a fatal or at least bitter antagonism between liberalism and virtue. (1) But this is a serious mistake, one that prevents liberalism from recognizing the conditions that preserve it. Contrary to much conventional wisdom, the liberal tradition not only makes room for virtue but shows that the exercise of virtue is indispensable to a political regime seeking to establish equality and protect freedom. Of course I do not mean to say that it is a simple matter to protect or promote virtue in a liberal society; nor do I wish to deny that peculiar features of liberal thought may put the very intelligibility of virtue at risk. (2) Rather, I wish to suggest that one can begin to grasp the genuine complexity of the matter and start to see the real risk by appreciating the rich and illuminating set of opinions advanced by the makers of modern liberalism about the dependence of freedom and equality on virtue.


I shall follow Judith Shklar in understanding liberalism as a political doctrine, the primary goal of which is "to secure the political conditions that are necessary for the exercise of personal freedom."(3) I add to Shklar's definition what she left implicit, namely, that the exercise of personal freedom is a right that liberalism seeks to extend equally to all. To establish and secure the personal freedom of all, the liberal tradition has elaborated a characteristic set of practices and institutions including toleration, liberty of thought and discussion, representation, and the separation of governmental powers. John Locke, Immanuel Kant, and John Stuart Mill are among the liberal tradition's leading spokesman. But many others--including Thomas Hobbes, Adam Smith, the authors of The Federalist, Montesquieu, and Tocqueville--shared its fundamental premise and in various ways elucidated its strengths and weaknesses.

A new appreciation has recently emerged that one of the conditions necessary for the preservation of a political society that protects the personal freedom of all is a certain virtue in citizens. (4)By "virtue" I mean a functional excellence. Virtues, in this view, which has its roots in the political philosophy of Aristotle, are qualities of mind and character that not only are exercised in the pursuit of man's highest end or human excellence, but also in the achievement of intermediate and lesser ends-ends, for instance, such as the responsibilities of citizenship, cooperation for mutual advantage, and the preservation of political life. (5) On examination, one discovers that it is common for the makers of modern liberalism, even where the very idea of human excellence is not only put in question but plainly repudiated, to expound a catalogue of virtues, to distinguish the end or ends which the virtues serve, and to specify the means by which virtue is to be fostered in political society.

The thought of Thomas Hobbes may seem like a peculiar place to begin a brief survey of the varieties of opinion about virtue in the liberal tradition. After all, in his masterwork Leviathan (1651) Hobbes defends an absolute sovereign who commands the fear of his subjects and who maintains peace and order by such decidedly illiberal measures as the regulation of Church doctrine and oversight of the university curriculum. Yet crucial elements of Hobbes's political theory exhibit a striking family resemblance to distinguished liberal ideas. Hobbes argued that human beings are fundamentally equal and endowed with certain natural and inalienable rights;(6) defended the idea of a state based on the rule of law; (7) maintained a basic distinction between the public and private; (8) held that a primary task of good government was to secure a minimum welfare for all citizens; (9) affirmed that civil laws govern actions, not thoughts or conscience; (10) insisted on the utility of toleration; (11) and advanced a form of representative government based on the idea that the source of subjects' obligation to obey the civil law stems from the fact that each subject, in obeying the sovereign's command, is obeying a power that the subject himself could be seen as having consented to and authorized. (12)

Hobbes then is "a kind of liberal. (13) And despite his famous rejection of the idea of an ultimate end or greatest good (14) and notwithstanding his scathing attack on the Christian Aristotelianism of the Schoolmen, which he mocked as "Aristotelity,"(15) virtue of a kind is at the very center of Hobbes's political theory. (16)To be sure, that honor is usually accorded to Hobbes's doctrine of the laws of nature. But in chapter 15 of Leviathan, at the end of his enumeration of nineteen laws of nature, Hobbes acknowledges that, properly speaking, the laws of nature are not really laws at all: "These dictates of Reason, men use to call by the names of Lawes, but improperly: for they are but Conclusions, or Theoremes concerning what conduceth to the conservation and defence of themselves." (17)

If the laws of nature are not really laws, what then are they? They are, as Hobbes says, conclusions or theorems; and, among other things, they proclaim the dependence of politics on certain qualities of mind and character that enable individuals to overcome restless and unruly passion so as to perform the actions that reason dictates. Indeed, the laws of nature in Hobbes's view are best thought of as moral virtues. So states Hobbes without apology or embarrassment:

all men agree on this, that Peace is Good, and therefore, also the way, or means of Peace, which (as I have shewed before) are Justice, Gratitude, Modesty, Equity, Mercy, & the rest of the Laws of Nature, are good; that is to say, Morall Vertues; and their contrarie Vices, Evill. Now the science of Vertue and Vice, is Morall Philosophie; and therefore the true Doctrine of the Lawes of Nature, is the true Morall Philosophie. (18)

Far from erecting a theory of the state that dispenses with virtue, Hobbes's Leviathan is firmly grounded in it.

To be sure, virtue may seem to lose much of its luster as Hobbes restricts its meaning to the qualities of mind and character that serve "sociability" or the maintenance of peaceful and cooperative social relations. But moral virtue so understood is neither easy nor abundant. Moral virtue is hard because men tend to fear the wrong things, especially religious things, "Powers Invisible" and punishments in a world to come,(19) instead of visible threats such as the sovereign's sword and the anarchy that civil war unleashes. Absent the fear of the right things, violent or reckless passion gains the upper hand, causing men to fail to do what is right and reasonable: to keep promises, to honor those who have benefited them, to ask no more for themselves than others receive, to treat all equally, and to perform other habitual actions that reason declares are necessary for the preservation of social and political life. Ultimately, failure to fear the right things in the right way brings about the most fearful of things: the dissolution of political society and the advent of a war of all against all, a condition which makes the life of each "solitary, poore, nasty, brutish, and short."(20) In sum, Hobbes teaches that cooperation for mutual advantage is an art, requiring the cultivation of specific beliefs, the restraint of powerful natural passions, and a range of habitual dispositions to perform certain general acts prescribed by reason. But how is this art learned? Where are such habits of the heart and mind acquired?

When the laws of nature are understood as Hobbes explicitly and repeatedly said they ought to be-as moral virtues-the question of how the moral virtues arise and what can be done to foster them comes into focus as more critical to his thought than the questions which have recently preoccupied so many scholars concerning political obligation or how the laws of nature can be binding. In fact, much that the liberal mind finds noxious in Hobbes has to do with the means Hobbes identifies for fostering the moral virtues that conduce to a good affirmed as much by liberalism as by Hobbes, namely, the primary good of peace. Control of what opinions are taught by the Church and at the universities (but not interrogation to determine what is believed) is a crucial task for the sovereign because "the Actions of men proceed from their Opinions; and in the wel governing of Opinions, consisteth the well governing of mens Actions, in order to their Peace, and Concord." (21)

Although the means Hobbes identifies for fostering moral virtue may be obtrusive and clumsy, and while he no doubt exaggerated the dependence of peace on an absolute and indivisible sovereign, it is of the essence of his argument that a politics that no longer aims at leading human beings to salvation or perfecting souls is not for that reason a politics that can do without virtue. To the contrary. The logic of politics, as Hobbes expounds it, reveals that a state freed from what Hobbes viewed as the false and pernicious idea that it is the task of government to save souls or perfect persons-for example, a liberal state-is still very much in need of "Equity, Justice, Gratitude, and other morall Vertues. (22)

John Locke would not have disagreed with the importance that Hobbes attached to virtue, yet, along with Hobbes, Locke is often blamed for devising an approach to politics that excludes or neglects virtue. Sometimes Locke's indifference or downright hostility to virtue is inferred from his critique of innate ideas inAn Essay Concerning Human Understanding (1689). Yet in seeking to identify the limits of human understanding Locke aims to set virtue on a rational and solid foundation by loosening the hold of bad arguments from custom and tradition made on virtue's behalf. (23) The virtue Locke is primarily concerned to defend in theEssay is not in the first place directed at human perfection or eternal salvation. Like Hobbes, Locke focuses attention on the qualities of mind and character that keep society together, or what he calls "social Virtue." (24)

The foundation of social virtue is a rule: "That one should do as he would be done unto.(25) Again like Hobbes, Locke argued that this moral rule was fair and reasonable but neither self-enacting nor self-enforcing and therefore in need of virtue for its support. Moved by desire for happiness and aversion to misery, but prone to "wrong Judgements" about what conduces to happiness, (26)human beings require particular qualities of mind and character to restrain desire so as to perform the actions and comply with the rules that reason prescribes. (27) Reason can show, Locke believes, that the practice of the social virtues is useful or consistent with the promotion of public happiness because private advantages accrue to those who live in a society where the rules of morality are honored.(28)

Sometimes Locke's indifference or hostility to virtue is inferred from the fact that he does not have very much to say about it in his most famous contribution to political theory, theSecond Treatise (1689). But to make such an inference is to look for virtue in the wrong way and in the wrong place. For in the Second Treatise, Locke explicitly focuses on a circumscribed part of politics, the part, Locke's title page explains, "concerning the True Original, Extent, and End of Civil Government." (29)

Nevertheless, although his primary concern in theSecond Treatise is the form of legitimate government, Locke makes the presence of virtue felt at crucial junctures in his analysis, and the very logic of his argument calls attention to the necessity for virtue in citizens and office holders. For example, in Locke's account the state of nature deteriorates into a state of war because of a shortage of the qualities of mind and character that conduce to self-preservation; (30) the right to private property is grounded in the exercise of the virtues of rationality and industry; (31) the powers of government must be separated because virtue is always in short supply, but prerogative, which depends on virtue in judgment, must be retained by the executive because of the necessary imperfections of the rule of law;(32) and the right of resistance to illegitimate government presupposes the exercise of restraint and rational judgment by the people. (33) Indeed, in the Second Treatise, Locke even goes so far as to identify the family as the institution crucially responsible for fostering virtue in a political society that respects natural freedom and equality. (34)

It is in another work, though,Some Thoughts Concerning Education (1693)-an eighteenth-century best seller on how to rear children from infancy through young adulthood-that Locke spells out how education in the family can support liberty by fostering virtue. (35) Indeed, education, in Locke's view, is essentially formation of character or training in "the Principles and Practice of Vertue." (36) As in An Essay Concerning Human Understanding, Locke's Some Thoughts Concerning Education does not deny the importance of, but also does not deal very much with, virtues oriented toward human perfection or eternal salvation. (37) In Education, though, Locke goes beyond the social virtues to focus on what might be called "gentleman's virtues," the qualities of mind and character that produce useful and able citizens capable of maintaining their possessions, prospering in commercial affairs, and governing well. (38)

The largest part of Lockean education consists in making desire subject to the rules and restraints of reason (39) Children must be taught to take pleasure in the approbation that comes from acting rightly or in accordance with reason, and to suffer the pain of shame when they act badly or contrary to reason. While Locke himself points out that love of reputation is not the "true Principle and Measure of Vertue," it is appropriate, he insists, for the education of gentlemen. (40) And while elsewhere he credits Christianity with making virtue "the most enriching purchase," (41)in the Education Locke argues that "Shame of doing amiss, and deserving of Chastisement, is the only true Restraint belonging to Vertue." (42) Locke praises liberality, justice, courage, and civility, and other moral virtues in the Education because they place reasonable restraints on desire and contribute to happiness by enabling naturally free and equal men to live together in peace and prosperity.

Locke's scheme for moral education-and more generally his solution to the problem of the source of virtue in regimes dedicated to the protection of individual liberty-presupposes stable and prosperous families, depends on parents with a generous endowment of moral virtue, and calls for an immense concentration of time, energy, and financial resources on the moral education of the young. It does not quite follow that if parents no longer praise and blame as Locke recommends, or if the two-parent family dissolves as a basic unit for the rearing of children, or if it becomes infeasible to devote to each child the extraordinary care and attention that Locke believed was necessary to form adults capable of self-government, that Lockean liberalism would cease to be a viable form of political organization. Such developments, however, would mean that a substitute source would have to be found to foster the virtues for which, on Locke's account, there is no substitute.

Striking features of the philosophy of Immanuel Kant suggest, contrary to Locke, that concerns about virtue and its sources are in fact unimportant for liberal morality and politics. The distinction fundamental to Kant'sCritique of Pure Reason (1st ed. 1781; 2nd ed. 1787) between a phenomenal or empirical world strictly governed by the laws of nature and a noumenal or intelligible world governed by the laws of freedom seems to deprive of moral worth the everyday world of sense and experience in which virtue-understood as qualities of mind and character-operates. (43) The contrast basic to Kant's Groundwork of the Metaphysic of Morals(1785) between non-moral inclination-which includes wants, needs, and desires-and moral motivation-the duty to obey the moral law out of respect for the moral law-appears to rob our particular attachments and passions of moral worth and dignity. (44) And the famous assertion in Perpetual Peace (1795)-central to Kant's thinking on politics-that the problem of the state can be solved even for a nation of intelligent devils apparently implies that citizens do not even need to possess a minimum level of virtue to make a liberal republic work. (45) It can easily seem that if Kant is to give virtue any place at all in his practical philosophy, then it must be a narrowly defined place occupied by a desiccated notion of virtue.

This appearance is not altogether misleading, but it is far from adequate. In the preface to the Groundwork of the Metaphysic of Morals (1785), Kant explains that ethics consists of two parts. One part is concerned with what is contingent and empirical; Kant calls it practical anthropology (it includes moral psychology, comparative politics, and sociology) and it deals with the actual traits of flesh and blood human beings and the historically diverse forms of social and political life. The other part of ethics is concerned with what is formal and rational; Kant calls it morals or moral philosophy and it deals with a priori practical principles that can be known objectively and that bind universally. Moral philosophy is of substantially higher dignity in Kant's understanding of ethics than practical anthropology. Nevertheless, it is not disdain or disinterest, but a kind of philosophical modesty, an appreciation of the limits of theoretical reason, that compels Kant largely to refrain from investigating issues of practical anthropology in his writings on moral philosophy, even while his examination of theoretical reason itself shows that what belongs to practical anthropology is an ineliminable dimension of ethics.(46)

Indeed, critics and many Kantians frequently overlook that while Kant insists that the laws of morality must be formulated without reference to the facts of empirical human nature or the circumstances of particular human beings, he knew full well and did not hesitate to affirm that it is only when assisted by the empirical part of ethics that the laws of morality can be made effective in particular lives and actual circumstances:

These laws [of morality] admittedly require in addition a power of judgement sharpened by experience, partly in order to distinguish the cases to which they apply, partly to procure for them admittance to the will of man and influence over practice; for man affected as he is by so many inclinations, is capable of the Idea of a pure practical reason, but he has not so easily the power to realize the Idea in concreto in his conduct of life. (47)

Kant then is committed to the view that while qualities such as critical reflection, prudent judgment, and self-discipline are not good without qualification-and hence strictly speaking are without "genuine moral worth"-they nevertheless play an indispensable role in supporting the moral life (48) Although in the Groundwork Kant restricts the meaning of virtue to "true virtue," by which he means purity of the will, (49) he also affirms that such impure and empirically tinged traits as "talents of the mind" and "qualities of temperament" are not only necessary but good in a qualified way because they enable individuals to actualize the moral law in life. (50)

In theDoctrine of Virtue, part 2 of the Metaphysic of Morals (1797), Kant does not so much change his theoretical position as loosen his definition of virtue by understanding it as "moral strength of the will." (51) Such strength of will is essential because as rational beings who are also finite and situated in the natural world, human beings are constantly tempted to violate the moral law even while recognizing its authority. (52) Intellectual virtue or understanding is necessary to make respect for the moral law one's incentive for obeying it. (53)Fortitude, or "the capacity and considered resolve to withstand" strong but unjust natural impulses, and moderation, or control of physical desire, support the disciplined exercise of the understanding.(54) Lying, avarice, and servility are vices that must be avoided at all costs because uttering falsehoods, lusting for possessions, and disavowing one's dignity as a creature endowed with a rational nature always reflect a maxim that contradicts the idea that humans are moral beings. (55) Man has a duty to cultivate his natural powers of spirit, mind, and body because they may be needed in the pursuit of the various ends that reason prescribes. (56) Beneficence is a virtue that expresses a general respect for the humanity in other persons; gratitude a respect for the humanity in benefactors; and sympathy a respect for the humanity in the poor and downtrodden. (57)

Although he sees virtue-both in the technical sense of purity of will and in the looser sense of qualities of mind and character that support the moral life-as part of morality, Kant appears in his writings on politics to marginalize virtue by arguing that morality is unnecessary for political life. Yet he does not, as it is sometimes said, reduce politics to legality, or reduce the science of politics to the science of law or jurisprudence. Rather, Kant limits thephilosophical investigation of politics to what reason, independent of experience, can clarify about the principles according to which human beings ought to organize their collective lives. In articulating these principles Kant's philosophy in fact affirms the inescapableness of virtue--or at least what Aristotle, Hobbes, and Locke could all agree was a form of virtue--and it does so precisely where Kant is often said to show how the escape from virtue in politics is necessary and possible.

Kant held that a liberal republic, one in which the freedom of each coexists with the freedom of all, was "the most difficult to establish, and even more so to preserve." (58) But he famously added that reason teaches that to maintain such a state men need not be angels. By this Kant meant that men need not be morally upright and genuinely virtuous, or as he memorably put it: "As hard as it may sound, the problem of setting up a state can be solved even by a nation of devils (so long as they possess understanding)." (59) What is easy to overlook is that Kant's claim that liberal republics can be preserved by unangelic, self-seeking men does not imply that such men need not possess particular qualities of mind and character that do not arise spontaneously but come from discipline and education. One reason that it is so easy to overlook this is because Kant is reluctant to give such qualities the name virtue.

Readers also misinterpret Kant because they overlook the important qualification he inconspicuously places inside parentheses, namely, that the self-serving devils for whom the problem of the state can be solved must possess understanding. (60) Understanding, as Kant uses the term, includes more than the ability to determine what is in one's self-interest however understood. Kant's devils with understanding grasp long term or enlightened self-interest; they also possess the capacity to overcome unruly inclinations so as to comply with the counsels of enlightened self-interest. Therefore, just as it was necessary, in Kant's view, for individuals to acquire and cultivate qualities of mind and character relative to moral perfection, so too it will be necessary for a nation of intelligent devils, if they wish to establish and preserve a liberal republic, to cultivate and acquire the qualities of mind and character relative to the lesser end of respecting the coercive public laws of the state. Whether one gives these qualities the name virtue is much less important than appreciating that on Kant's own account liberal republics require them, and that since they do not arise spontaneously, particular beliefs, practices and associations must be instituted and sustained to foster them.

Kant himself seems to recognize this in Perpetual Peace when he argues that before such time as perpetual peace is firmly established among the nations of the world-an achievement which does depend upon "the moral improvement of man" and is advanced by "good political constitution[s]" (61)-much knowledge of experience and human nature will be necessary to determine what policies will best secure the obedience and prosperity of the people. Although he consigns such questions to the realm of "political expediency," and while he disparagingly characterizes such undertakings as "mere technical tasks," Kant sees government as essential to insuring that citizens possess the qualities necessary to maintain the conditions necessary for personal freedom. (62)

In contrast to Kant, John Stuart Mill argued that moral improvement or virtue, grounded in the "permanent interests of man as a progressive being, (63) was not only a precondition for but also an aim of good government. (64) In On Liberty (1859) Mill celebrates individual liberty, and especially the liberty of thought and discussion, for the service it renders to the formation of strong, energetic, and upright individuals. (65) Individuality, for Mill, is not a quality men and women are born with, but a moral standard for judging the quality of the lives men and women lead. (66) Human excellence understood as individuality depends on a rigorous education and stern self-discipline. It culminates in a free and complete mind that is skeptical, rational, self-critical, solicitous of the needs of the heart, and capable of seeing not only what is foolish and false in inherited beliefs and practices but also of discerning what is rational in custom and tradition and in need of preservation.(67) The liberal spirit is, accordingly, characterized by "many-sidededness," a notion Mill borrowed from Goethe and a virtue he associated with Socrates. (68)

Mill did not think that politics would be the primary source in modern democracies of the virtue that liberty depended on-or the virtue that it made possible. This was not because he underestimated the importance of self-government or denied that law can play some role in making men and women moral. Rather, politics could not be the chief source of virtue in the modern age because, as Mill observed inThe Subjection of Women (1869), "citizenship fills only a small place in modern life, and does not come near the daily habits or inmost sentiments." (69) Like Tocqueville, about whom he wrote with perception and admiration,(70) Mill saw associational life as rescuing individuals from the isolation and self-absorption fostered by modern democracy and the commercial spirit. Voluntary associations teach habits of cooperation and instill an enlightened concern for the public good. (71)

But the most important institution, according to Mill, in preparing individuals for the rigors of liberty in the age of modern democracy was the family, or more accurately the family reformed in accordance with the fundamental premise of liberalism, the natural freedom and equality of all human beings. InThe Subjection of Women, Mill argued that the family as it was still constituted by law was "a school of despotism in which the virtues of despotism, but also its vices, are largely nourished." (72) But he also insisted that "the family, justly constituted"-that is a family based on the legal equality of men and women-"would be the real school of the virtues of freedom." (73)

The State too had its role in fostering virtue. Active involvement of the State was necessary to correct the neglect of "one of the most sacred duties of parents," that of providing one's child with "an education fitting him to perform his part well in life towards others and towards himself. (74)It was "almost a self-evident axiom, that the State should require the education, up to a certain standard, of every human being who is born its citizen." (75) Parents who failed to cultivate the moral and intellectual capacities of their child committed a "moral crime" that obliged the State to step in. (76) Mill did not want the State itself to be in the business of providing a universal education: he feared intractable controversies about the content of the curriculum; and in the event of agreement, he feared a uniform education that cultivated nothing so much as uniformity of opinion. But Mill did want the State to enforce a universal standard of education through the administration of public examinations. Parents would be held legally responsible for ensuring that their children acquired a certain minimum of general knowledge. Payments from the State would be provided to parents who could not otherwise afford basic education for their children. In addition, the State would provide certification through examination in the higher branches of knowledge. To prevent the State from improperly influencing the formation of opinion, such examinations-in particular in the fields of morality, politics, and religion-would be confined to facts and opinions on great intellectual controversies that had been held rather than to the truth or falsity of those opinions. (77)

InConsiderations on Representative Government (1861), his most systematic treatise on politics, Mill argued that virtue is a standard for judging actual regimes; (78) that the regime that in practice best takes advantage of and promotes virtue is popular government; (79)and that in a popular government suitable to modern conditions, representative institutions must be fashioned so as to bring to the fore individuals outstanding in terms of moral stature and intellectual competence, while increasing the supply of virtue among the multitude of citizens. (80)

To counteract the destructive tendency to level differences and promote a uniform mediocrity that he discerned at work in modern democracy, Mill proposed the devices of proportional representation, plural voting, and publicity in voting. The intended benefit of proportional representation was to help secure a voice in government for a very particular minority, one that tends to be neglected, Mill thought, in free elections: "the minority of instructed minds. (81) Plural voting, while respecting the just demand of popular government that all equally be given the opportunity to participate in government, gives the more competent more votes in order to respect the need in government for individuals of enlarged and refined capacities.(82) And publicity in voting, Mill argued, would promote a certain kind of civic virtue by compelling voters to justify their choice to others; or, at least, facing the possibility of having to justify their choice to others, voters would be impelled, in casting their ballot, to look beyond private advantage to considerations of the public good. (83) Obvious and even decisive practical objections to his proposals should not be allowed to obscure the key supposition underlying Mill's ideas for institutional reform of popular government: democratic institutions should be designed with an eye to the fact that in modern democracies virtue is always necessary and, because of democracy's destructive tendencies, frequently in short supply.

To summarize: If one rejects the simple equation of virtue with human perfection and understands virtue also as those qualities of mind and character that support the attainment of a range of ends and the performance of a variety of tasks, then such makers of modern liberalism as Hobbes, Locke, Kant, and Mill come into view as assigning an essential place to virtue in moral and political life. Their differences of opinion about virtue as well as underlying continuities can be brought out by examining in the case of each thinker the specific catalogue of virtues put forward, the end or ends virtue is asked to serve, and the means proposed for fostering virtue. As I remarked at the outset, I do not wish to deny that the very idea of virtue in the liberal tradition is marked by basic and destabilizing tensions. What I do wish to affirm, though, is that the liberal tradition provides an illuminating and underappreciated source of instruction about the necessity of virtue where the natural freedom and equality of all is a principle upon which the legitimacy of government is thought to rest.


Setting the historical record straight is a good and sufficient reason for reexamining the opinions about virtue in the liberal tradition. But it is not the only reason and perhaps today not the most compelling one. The liberal tradition is also worth reexamining because it continues to articulate tensions between competing goods in our political life, goods whose importance is often obscured by contemporary scholars who, taking them for granted, suppress the difficulties that surround their reasoned defense.

For example, in contrast to those who oppose to contemporary liberalism and its concern for individual rights and fair procedures a civic republicanism devoted to the goods of democratic participation and the energetic practice of civic virtue, (84) the liberal tradition teaches how to affirm the importance of virtue and the associational life to which it is intimately connected without losing sight of the good reasons for protecting individuals against the authority of community and protecting communities as well as individuals, when necessary, against overbearing state power. In contrast to those who analyze the weaknesses of American democracy in terms of disappearing stocks of social capital and a declining civil society, (85)the liberal tradition reminds that social capital depends on moral capital-that is, on energetic and self-reliant individuals capable of forming and maintaining the voluntary associations that sustain the habits of cooperation and self-restraint that are so useful to liberal democracies. In short, in contrast to today's democratic theorists who typically see only the need to restore some single element of democracy in America, the makers of modern liberalism teach the permanent necessity-at least for states based on the freedom and equality of all-of weaving together moral and political principles that must be made to support one another although they often pull in opposing directions.

Today this weaving must involve the state. Yet, for a liberal state to take some responsibility for virtue without violating liberal principles or stepping beyond its limited competence, it will have to both show vigor and exercise forebearance. (86) In particular cases, whether government should exercise restraint or in what way it should intervene will of course depend on the answers to complicated empirical questions concerning the actual effects on character of existing institutions and the likely consequence of proposed reforms on individual conduct. Theory has its uses but its uses for politics are limited. One use of the sorts of theoretical considerations I have sketched is to direct attention to questions about the proper ends that liberal regimes may pursue, the legitimate means that liberal regimes can employ to foster or safeguard virtue, the catalogue of virtues that serve liberal purposes, and, not least, the vices that are bred by the protection of freedom and the establishment of equality.

It is, finally, no flighty metaphysical impulse or self-indulgent hankering for the transcendent, no sentimental yearning for times past, no proud preoccupation with human perfection that compels liberals and their friends to return to questions about the virtues necessary to liberalism's preservation. It is the logic of politics that makes virtue a permanent issue for every regime; it is the logic of liberalism that insures that the care for the necessary virtues in liberal democracies must be a delicate balancing act; and it is the peculiar situation of U.S. liberal democracy today that makes the recovery of the old sources of virtue or the invention of new ones an urgent matter. Urgent, that is, for those who honor the claims of democratic equality, for those who cherish political liberty, and for those who think regimes wise that, out of respect for both what is good and bad in human beings, push from the center of politics the enduring question of human excellence.

Peter Berkowitz is associate professor of government at Harvard University. He is the author ofNietzsche: The Ethics of an Immoralist.

1. For example, Michael Sandel's criticism of liberalism as a chief source of the ills that beset American political life is based on a sharp distinction between a liberalism that is devoted to individual rights and fair procedures, and a civic republicanism that is concerned with freedom through self-government, democratic participation, and civic virtue; see his Democracy's Discontent: America in Search of a Public Philosophy(Cambridge, Mass.: Harvard University Press, 1996), 3-54, 321-23. At the same time, Stephen Holmes, in a vigorous defense of liberalism, goes out of his way to ridicule those elements of Mill's political theory that stress the significance of moral and intellectual virtue and implies that they are incompatible with Mill's liberalism. See, "The Positive Constitutionalism of John Stuart Mill," in Passions and Constraint, (Chicago: University of Chicago Press, 1995), 188-93, 194-96. Ironically, Holmes joins forces with Sandel in advancing the dubious thesis that concern for virtue is foreign to or incompatible with the political theory of liberalism.

2. For works that illuminate the problem of virtue by examining the spirit and intellectual framework of liberal modernity see Charles Taylor, Sources of the Self: The Making of the Modern Identity(Cambridge, Mass.: Harvard University Press, 1989); Alasdair MacIntyre, After Virtue, 2nd ed. (Notre Dame: University of Notre Dame Press, 1981, 1984); and Leo Strauss, Natural Right and History (Chicago: University of Chicago Press, 1950).

3. "The Liberalism of Fear," in Liberalism and the Moral Life, ed. Nancy Rosenblum (Cambridge, Mass.: Harvard University Press, 1989).

4. For a summary, by a leading revisionist, of recent revisionist scholarship that has stressed liberalism's dependence on virtue see William Galston, "Liberal Virtues and the Formation of Civic Character," in Seedbeds of Virtue, ed. Mary Ann Glendon and David Blankenhorn (New York: Madison Books, 1995), 37-39.

5. In the Ethics Aristotle is largely concerned with "human virtue" or "virtue of the soul," that is, the qualities of mind and character that conduce to human excellence. See, for example, Nicomachean Ethics 1102a5-26. But in the Politics he pays more attention to the virtues or qualities that support lesser ends, undertakings, and responsibilities--lesser but not unimportant ends such as physical health, managing the household, and citizenship. See, for example, Politics 1260a1-25, 1276b15-1278b5.

6. Leviathan, ed. Richard Tuck (Cambridge: Cambridge University Press, 1991), chap. 13, pp. 86-90; chap. 14, pp. 93, 96, 98; chap. 21, p. 153; chap. 27, pp. 202, 208.

7. Leviathan, chap. 15, pp. 108-09; chap. 18, pp. 125-126; and in general chapters 26-30.

8. Leviathan, chap. 19, p. 131.

9. Leviathan, chap. 30, p. 239; see also chap. 24, pp. 172-73.

10. Leviathan, chap. 42, p. 343; chap 46, p. 471.

11. Leviathan, chap. 47, pp. 479-80.

12. Leviathan, chaps. 16-18, pp. 111-129; chap. 21, pp. 145-48. See also The Elements of Law in Human Nature and De Corpore Politico, ed. J. C. A. Gaskin (Oxford: Oxford University Press, 1994), 2.xxv.12, p. 153.

13. See Richard Tuck, Hobbes (Oxford: Oxford University Press, 1989), p. 97; see also pp. 72-74.

14. Leviathan, chap. 11, p. 70.

15. Leviathan, chap. 46, p. 462.

16. An incisive analysis can be found in R.E. Ewin, Virtue and Rights: The Moral Philosophy of Thomas Hobbes (Boulder: Westview Press, 1991). See also Michael Oakeshott, "The Moral Life in the Writings of Thomas Hobbes," in Rationalism in Politics and other Essays, foreword by Timothy Fuller, New and Expanded Edition (Indianapolis: Liberty Press, 1991), pp. 295-350; and Leo Strauss, The Political Philosophy of Hobbes, trans. Elsa M. Sinclair (Chicago: University of Chicago Press, 1936, 1952), esp. 44-58, 108-28.

17. See also Elements of Law, 1.xvii.12, p. 97; and De Cive, in Man and Citizen, ed. Bernard Gert (Indianapolis: Hackett, 1991), 3.33, p. 152.

18. Leviathan, chap. 15, p. 111. This forthright statement, in which the laws of nature are equated with the moral virtues, far from being an aberration in Hobbes's thought, is affirmed elsewhere in Leviathan and expounded in the Elements of Law and De Cive. See Leviathan, chap. 26, pp. 185, 191, and chap. 31, p. 248. See also Elements of Law, 1.xvii.10-15, pp. 96-99; and De Cive, 3.29-33, pp. 149-52, and 15.8, p. 295.

19. Leviathan, chaps. 11, 46.

20. Leviathan, chap. 13, p. 89.

21. Leviathan, chap. 18, p. 124.

22. Leviathan, chap. 26, p. 185.

23. John Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch (Oxford: Oxford University Press, 1975), 1.i.5-6, pp. 45-46, and 1.iv.23-24, pp. 100-02.

24. Essay, 1.iii.4, p. 68.

25. Essay, 1.ii.4, p. 68. Cf. Hobbes, Leviathan, chap. 14, p. 92, chap. 15, pp. 109-10.

26. Essay, 2.xxi.62-63, pp. 274-76.

27. Essay, 2.xxi.42-70, pp. 258-82.

28. Essay, 1.iii.6, p. 69.

29. Nathan Tarcov has called attention to a suggestive passage in a minor work on education in which Locke himself articulates a fundamental distinction between two kinds of political inquiries: "Politics contains two parts very different the one from the other, the one containing the original of societies and the rise and extent of political power, the other, the art of governing men in society" ("Some Thoughts Concerning Reading and Study for a Gentleman," in The Educational Writings of John Locke, ed. James L. Axtell [Cambridge: Cambridge University Press, 1968], p. 400). For an extended discussion of the significance of this passage for understanding the division of labor among Locke's works on morality and politics see Tarcov, Locke's Education for Liberty (Chicago: University of Chicago Press, 1984) 4-8; and Ruth Grant, John Locke's Liberalism(Chicago: University of Chicago Pres, 1987), 21-26.

30. Second Treatise, sec. 128, p. 352.

31. Second Treatise, sec. 34, p. 291.

32. Second Treatise, secs. 143-144, pp. 364-65; sec 153, pp. 369-70; secs. 158-68, pp. 373-80.

33. Second Treatise, sec. 225, p. 415; sec. 240, pp. 426-27.

34. Second Treatise, secs. 54-76, pp. 303-18.

35. The leading study of Locke's Some Thoughts Concerning Education is Tarcov,Locke's Education for Liberty. For an interpretation that argues that Locke's education represents a kind of deformation of character see James Tully, "Governing Conduct: Locke on the Reform of Thought and Behaviour," in An Approach to Political Philosophy: Locke in Contexts (Cambridge: Cambridge University Press, 1993), pp. 179-241. For a critique of Tully's interpretation see J. M. Dunn, "'Bright Enough for All Our Purposes': John Locke's Conception of a Civilized Society," in Notes and Records of the Royal Society of London, vol. 43, no.2, July 1989, pp. 133-53.

36. Education, in The Educational Writings of John Locke, sec. 70, p. 167.

37. In the Education Locke rarely mentions Christianity, and although he alludes to the idea of God as the true ground of virtue, he indicates that probing such matters does not belong to the business at hand. See, for example, Education, 61, pp. 155-156; 116, p. 226; 136-137, pp. 241-242.

38. It is right to recognize that the Education was designed for members of a class that comprised only about 4-5 per cent of the English population. It should also be appreciated that it was the members of this class who, in the seventeenth century, governed England and managed her commercial affairs. Thus, Locke's prescriptions for education address both a very narrow audience and a general or universal readership: while his Education is a guide to the education of an elite, it is at the same time a basic education for all those who exercise the rights and responsibilities of citizenship. Interestingly, Locke goes out of his way to point out that much of what he says about the education of "a young Gentleman" applies as well to the education of daughters. Education, sec. 6, p. 117.

39. Education, sec. 36, p. 140.

40. Education, sec. 61, pp. 155-156.

41. The Reasonableness of Christianity, ed. I. T. Ramsey (Stanford: Stanford University Press, 1958), sec. 245, p.70.

42. Education, sec. 78, p. 177.

43. Critique of Pure Reason, trans. Norman Kemp Smith (New York: St. Martin's Press, 1965), A540/B568, A550/B578.

44. Groundwork of the Metaphysic of Morals (hereafter cited as Groundwork), trans. H.J. Paton (New York: Harper & Row, 1964), 397-99. The page numbers are those found in the margins of Paton's translation and refer to the standard edition issued by the Prussian Academy of the Sciences.

45. Perpetual Peace, in Kant's Political Writings, ed. Hans Reiss (Cambridge: Cambridge University Press, 1970), 112-13.

46. According to a thriving tradition dating back to Hegel and sustained today in large measure by critics of liberalism, Kant's philosophy distorts the moral life with a rigorism that defines morality as a set of universal and necessary rules and a formalism that is indifferent to circumstances and lacking in substantive moral content. For Hegel's classic criticism see The Philosophy of Right, trans. T. M. Knox (Oxford: Oxford University Press, 1980), secs. 105-40, pp. 75-103; and Phenomonology of Spirit, trans. A.V. Miller (Oxford: Oxford University Press, 1980), secs. 419-37, pp. 252-62. For contemporary versions of the classic criticism see Bernard Williams, Ethics and the Limits of Philosophy (Cambridge, Mass.: Harvard University Press, 1985), 54-70; Alasdair MacIntyre, After Virtue, 2d ed. (Notre Dame: University of Notre Dame Press, 1984), 43-47; Michael Sandel, Liberalism and the Limits of Justice (Cambridge: Cambridge university Press, 1982); Charles Taylor, Hegel and Modern Society (Cambridge: Cambridge University Press, 1979), 76-84; and Roberto Unger,Knowledge and Politics (New York: Free Press, 1975), 50-51, 53-55. However, a wave of revisionist scholarship in recent years has constructively challenged this venerable view, arguing that there is greater subtlety and flexibility in Kant's moral philosophy than his Hegel-inspired critics have acknowledged, and rich, untapped resources for his proponents to develop. See, for example, Barbara Herman, The Practice of Moral Judgment (Cambridge, Mass.: Harvard University Press, 1993); Onora O'Neill, Constructions of Reason: Explorations of Kant's Practical Philosophy (Cambridge: Cambridge University Press, 1989), 145-64; Henry E. Allison, Kant's Theory of Freedom(Cambridge: Cambridge University Press, 1990), 180-98. See also Jürgen Habermas, "Morality and Ethical Life: Does Hegel's Critique of Kant Apply to Discourse Ethics?" in Kant & Political Philosophy(New Haven: Yale University Press, 1993), 320-336. For special emphasis on Kant's political theory see Patrick Riley, Kant's Political Philosophy (Totowa, N.J.: Rowman and Littlefield, 1982).

47. Groundwork, 389.

48. Groundwork, 398. See also Doctrine of Right, in The Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University Press, 1991), 217; page numbers are those found in the margins of Gregor's translation and refer to the standard edition issued by the Prussian Academy of the Sciences).

49. Groundwork, 407.

50. Groundwork, 393. For a more relaxed usage of the term virtue see Religion within the Limits of Reason Alone, trans. Theodore M. Greene and Hoyt H. Hudson (New York: Harper & Row, 1960), 12-13, 42-43.

51. Doctrine of Virtue (hereafter cited as Virtue), in The Metaphysics of Morals, 405.

52. Virtue, 379, 380, 394, 405.

53. Virtue, 386, 392.

54. Virtue, 380, 425-28. In a related discussion, Kant argues that the failure to achieve enlightenment in "a large proportion of men" has its sources in a lack of moral virtue. It is laziness, lack of resolution, and cowardice that keep so many in "self incurred immaturity." "The motto of enlightenment" expresses the dependence of intellectual virtue on moral virtue: "Sapere Aude! Have courage to use your own understanding!" See "An Answer to the Question: 'What is Enlightenment?'" in Kant's Political Writings, 54.

55. Virtue, 420, 429-37.

56. Virtue, 444-45; see also Groundwork, 421-23.

57. Virtue, pp. 452-61.

58. Perpetual Peace, 112.

59. Perpetual Peace, 112-13.

60. See, for example, Sandel, Democracy's Discontent, 321-22.

61. Perpetual Peace, 113, 124-25.

62. Perpetual Peace, 122.

63. On Liberty, in Essays on Politics and Society, ed. J. Robson (Toronto: University of Toronto Press, 1977), chap. 1, p. 224 (hereafter cited as On Liberty).

64. For an overview of the centrality of virtue in Mill's life and thought see Bernard Semmel,John Stuart Mill and the Pursuit of Virtue (New Haven: Yale University Press, 1984).

65. On Liberty, chap. 2, esp. pp. 231-32, 241-48, 257-58.

66. On Liberty, chap. 3, esp. pp. 261-64.

67. On Liberty, chap. 2, pp. 252-57. See also "Bentham" and "Coleridge" in Essays on Ethics, Religion, and Society, ed. J. Robson (Toronto: University of Toronto Press, 1969).

68. On Liberty, chap. 2, p. 252; Autobiography in Autobiography and Literary Essays, ed. J. Robson and J. Stillinger (Toronto: University of Toronto Press, 1981), chap. 5, p. 171.

69. The Subjection of Women, in Essays on Equality, Law, and Education, ed. J. Robson (Toronto: University of Toronto Press, 1984), chap. 2, pp. 294-95 (hereafter cited as The Subjection).

70. See "Tocqueville on Democracy in America, Vol. 1," and "Tocqueville on Democracy in America, Vol.2," in Essays on Politics and Society, ed. J. Robson (Toronto: University of Toronto Press, 1977).

71. On Liberty, chap. 5, pp. 305-06.

72. The Subjection of Women, chap. 2, p. 295.

73. The Subjection of Women, chap. 2, p. 295. See also "Inaugural Address Delivered to the University of St. Andrews," in Essays on Equality, Law, and Education, ed. J. Robson (Toronto: University of Toronto Press, 1984), 248.

74. On Liberty, chap. 5, 301.

75. On Liberty, chap. 5, 301.

76. On Liberty, chap. 5, 301-02.

77. On Liberty, chap. 5, 302-05.

78. Considerations on Representative Government, in Essays on Politics and Society, ed. J. Robson (Toronto: University of Toronto Press, 1977), chap. 2, pp. 385-86, 390-92 (hereafter cited as Considerations).

79. Consideration, chap. 3, pp. 403-04, 406-12.

80. Considerations, chap. 2, p. 392.

81. Considerations, chap. 7, pp. 455-60.

82. Considerations, chap.8, pp. 474-75.

83. Considerations, chap. 10, pp. 488-91.

84. See, for example, Sandel, Democracy's Discontent, pp. 3-54, 317-51.

85. See, for example, Robert Putnam, "Bowling Alone," in Journal of Democracy(January 1995): 65-78.

86. Cf. Benjamin Constant, "The Liberty of the Ancients Compared with that of the Moderns," in Political Writings, ed. Biancamaria Fontana (Cambridge: Cambridge University Press, 1988), 324.