Liberalism looms prominent in contemporary debates -- in this journal and elsewhere. But the term, however ubiquitous, remains elusive. By some, it is treated with cruel derision; by others, with breathtaking sanctimoniousness. A few writers, such as Alasdair MacIntyre or Christopher Lasch, finger liberalism as the source of all our miseries; others, such as Milton Friedman, preach that our most painful problems would be solved if we returned to liberalism in a pure and uncorrupted form. Some argue that the United States is a radiant monument to the liberal ideas of its Founders. Others retort that our society has evolved in unexpected ways and that stale eighteenth-century principles have become largely irrelevant to twentieth-century problems. Such postures are exhilarating. But they do not help us understand what liberalism was or how it has changed.
One claim about liberalism, common in textbooks, is that a major discontinuity divides classical from twentieth-century liberals -- James Madison from Franklin Roosevelt, Adam Smith from John Maynard Keynes. Both left and right seem to agree about this purported reversal in theory and practice. Contemporary welfare programs, conservatives assure us, represent a betrayal of the liberal legacy. And progressives in principle agree: we would never have implemented Social Security and a progressive income tax if we had not turned our backs on the devil-take-the-hindmost attitude of eighteenth-century liberals.
But is this true? What is liberalism? What was it in the seventeenth and eighteenth centuries? Was its original promise fulfilled or betrayed? Have American liberals, following Roosevelt, simply misappropriated a term that originally meant the opposite of what it has come to mean today?
"Liberalism" is not a vague Zeitgeist or the outlook of modern man, but a clearly identifiable cluster of principles and institutional choices endorsed by specific politicians and popular movements. The early history of liberalism, in fact, cannot be detached from the political history, in the seventeenth and eighteenth centuries, of England and Scotland, the Netherlands, the United States, and France. Liberal principles were clearly expressed not just in theoretical texts but in the English Habeas Corpus Act, Bill of Rights, and Act of Toleration (1679,1688-89), and the first Ten Amendments to the American Constitution and the Declaration de les droits de I'homme (both of 1789). Some liberal politicians, such as the Federalists, succeeded brilliantly; others, such as their contemporaries, the French Feuillants, just as dramatically failed.
The political theorists who most cogently defended liberal aspirations -- Milton, Spinoza, Locke, Montesquieu, Hume, Voltaire, Blackstone, Smith, Kant, Madison, and J.S. Mill -- were deeply immersed in contemporary controversies. Each spent his life responding to local challenges, agitating for specific reforms, struggling with circumscribed problems. They faced different enemies and allied themselves with different social forces. Their epistemologies and metaphysical beliefs were sometimes diametrically opposed. None can be fully understood if plucked ahistorically from his political and intellectual context and forced to march in a syllabus-like parade of liberal greats.
The positions they defended, nevertheless, tend to converge. Their common liberalism -- for we might as well call it that -- has nothing whatsoever to do (as their critics say) with "atomistic" individualism or a hostility to the common good. So what did it involve? Liberals are sometimes said to advocate "the priority of liberty." While not totally false, this catch-phrase is needlessly telegraphic. A list of the basic components of liberalism would have to include, at a bare minimum, religious toleration, freedom of discussion, personal security, free elections, constitutional government, and economic progress. But much more was and is involved.
Most liberals were both anticlerical and antimilitaristic, for instance. They were also opposed, in varying degrees, to hereditary monopolies, especially to the privileges of a few "great" families who owned large tracts of land. They scorned ties of vassalage and peonage and aimed to universalize the condition of personal independence. Most believed in the value of literacy and secular education for all, a fairer system of taxation, and the legitimacy of social mobility within and across generations. They welcomed immigration and freedom of movement in general. They supported the right to divorce. They opposed legal disabilities on religious minorities (so long as national security was not at stake). They endorsed the freedom to establish churches and to preach.
Legitimate authority, they argued, is based on popular consent, not on divine right or dynastic succession. They therefore defended not merely electoral politics but also the right of rebellion in some form. They advocated political pluralism and government by public discussion among provisionally elected and publicly accountable representatives. They hoped that bloody confrontation between armed factions could be, to some extent, replaced by rational bargaining and debate. They proposed a constant widening of the suffrage, more or less in tandem with the expansion of literacy, the relaxation of religious orthodoxy, and the abatement of religious passions. They also favored an independent judiciary, as well as laws that were clearly framed, publicly proclaimed, and fairly enforced. They prescribed the abolition of torture and savage punishments, legal checks on the police, guarantees both against retroactive legislation and arbitrary imprisonment, and jury trials in criminal cases. They tended to conceive punishment as a means of deterrence rather than a form of revenge. They advocated civilian control of the military. And they admired science or free inquiry as a deepening of human understanding, not merely as an instrument for mastering nature.
They were devoted not only to legal equality, but also to equality of economic opportunity. They were more distressed by poverty and personal dependency, how- ever, than by inequality of income or wealth. Thus, they urged a wide and rapid diffusion of private property. They believed that contracts should be enforced. They favored the abolition of domestic customs barriers, free entry into trades and occupations, and the freedom to exchange goods and services. In other words, they had a generally welcoming attitude toward commercial society. They looked favorably on commercialism because they believed that economic competition would create (among other things) enough general prosperity to improve the lives of even the poorest members of the community. Adam Smith defended free trade, for example, on the grounds that it would increase the welfare of "the lowest ranks of the people" and work "for the benefit of the poor and the indigent."
This cluster of moral principles and most-favored practices provides the best starting-point for an understanding of the liberal tradition. By an inversion and simplification, we can convert this unwieldy catalogue into a shorter list of liberalism's most-disliked institutions and regimes. Four classically illiberal arrangements leap to the eye: autocracy, aristocracy, theocracy, and collective ownership. (Today we might add ethnocracy.)
In an autocratic regime, a single faction, party, or clique monopolizes power, the press is censored or superintended by the government; individuals can be imprisoned for long periods without legal recourse; the secret police are given enormous discretion to liquidate "unreliables" and enforce political subservience; the ordinary police are poorly monitored and controlled; the economy is centrally managed; criticism of political rulers is forbidden and, therefore, government is likely to be capricious, oppressive, corrupt, and grossly misinformed. In an aristocracy, access to privilege is determined almost wholly by pedigree; landownership is the key to life; a closed oligarchy monopolizes political power; and social mobility within and across generations is minimal. In a fundamentalist theocracy or clerical authoritarian regime, bigotry is rewarded, innovation is sacrificed to indoctrination, intellectual exchange is quashed, deviations punished, orthodoxy enforced.
All three regimes are patently illiberal. None stirs much sympathy in the West. Outrage is expressed whenever their vestiges are discovered in liberal societies today. Remedies are proposed and sometimes applied. To the extent that autocracy, aristocracy, and theocracy are decried, liberal rhetoric, at least, has triumphed.
It is different with communism and the principle of economic levelling on which it is purportedly built. The socialist tradition, despite its trumpeted embrace of "progress," has assiduously cultivated and kept alive an archaic inequality taboo, inherited, it seems, from subsistence economies of the distant past. (The archaic roots of communism may partly explain the extraordinary contagiousness of authoritarian socialism in underdeveloped countries where traces of a premodern communalist ethos remain strong.)
What characterizes liberalism, by contrast, is its unembarrassed repudiation of ancient and Christian prohibitions on inequality of resources. While adamantly opposed to any sort of caste system, liberalism is notoriously tolerant of disparities in income and wealth. Liberals are intensely concerned about poverty and economic dependency, about absolute levels of well-being (including a "bottom floor" of decent subsistence) as well as economically entrenched relations of mastery and control. Liberals did not, however, view inequality of wealth itself, apart from problems of dependency and poverty, as an unacceptable social evil.
Critics often assert that liberal acquiescence in economic inequality stems from a profound belief that superior talents "deserve" superior rewards. But that is a dubious claim. Liberals accepted inequality of resources, in fact, because they saw it as an inevitable side effect of a productive economy. They viewed the inequality taboo as an expression of irrational envy, it is true. But they rejected it primarily because they saw it as an infallible formula for reproducing scarcity and exacerbating dependency. As Alexis de Tocqueville said in 1848: "socialism wants equality in poverty and slavery." Collective ownership is not only economically inefficient; it also destroys the independent resources on which political opposition is based.
In these pages, I do not offer a concise definition of liberalism. Without identifying liberalism's essence, I nevertheless sketch out a set of claims that are broadly characteristic of liberal political thought, including what I consider to be its three "core norms." I examine the basic contours of the liberal idea by looking sequentially at five concepts: the state, interests, rights, democracy, and welfare.
Liberalism is classically defined as an attempt to limit the power of the state for the sake of individual freedom. Liberals, it is true, were obsessed with curbing political tyranny. Their driving concern, many historians have argued, was to prevent hypertrophic government from oppressing individuals and groups. The essence of liberalism, from this perspective, lay in techniques for taming absolute power.
There are good reasons for emphasizing the antipower ethos within the liberal tradition. Liberals have a better grasp of economic realities than socialists and Marxists. But the most obvious superiority of liberal over Marxist thought stems from liberalism's persistent concern for -- and Marxism's infamous blindness to -- abuses of accumulated political power.
In fact, the antityrannical strand has always been, and remains today, a vital element within liberal thought. But it is not the whole story. To identify liberalism with a crusade to restrict state power is inadequate. For one thing, liberal states have, since the very beginning, proved breathtakingly powerful. The twentieth century provides some outstanding examples of the superiority of liberalism over autocracy from purely military and administrative perspectives. It is enlightening to reread, after the events of 1989, the speeches that Solzhenitsyn delivered in the United States during the 1970s. There he tells us that the West -- infected by the spirit of liberalism -- is becoming weaker and weaker, while the Soviet Union is moving with doom-like inevitability toward world domination. Many people made the same error about fascism in the 1930s. Liberal states are stronger than those awed by authoritarian power believe.
The dramatic story of nineteenth-century Britain is another case in point. The age of free trade and the industrial revolution, of course, was simultaneously the age of the British Empire. Shockingly enough, a small island off the northwest coast of Europe gained mastery over a third of the globe. The classic country of political liberalism did not display state weakness in any obvious sense. Liberal politics, in fact, seems to have been accompanied by a startling increase in the capacity of the state to mobilize resources for collective purposes.
This was already clear in the early eighteenth century. In the first great wave of liberal propaganda, Voltaire and Montesquieu praised England not only for its liberties, but also for its power -- for the number of ships in its harbors. Oppression, they both argued, weakens a state. Intolerance deepens sectarian conflict and drives useful citizens abroad. Censorship blocks the flow of information vital for the governance of a large nation. Cruel and excessive punishments crush the spirit of ordinary citizens, depriving the government of their active collaboration. Heavy-handed regulations of trade decrease the private wealth that might be tapped for the public treasury. A liberal polity is much better situated than a tyrannical one for enlisting citizen cooperation in the pursuit of common objectives. Voltaire, and even Montesquieu, identified liberalism with a welcome magnification (along some dimensions) of state power.
This line of reasoning makes perfect sense. It is implausible, after all, to view liberal rights as naturally incompatible with political power, as if such rights flourish only when the state withers away. Authority and liberty are interdependent, not simply opposed. As Kant, among others, made dear, rights (including property rights) are defined and enforced by the state. Referring to "natural rights," Emile Durkheim convincingly wrote that "the State creates these rights, gives them an institutional form, and makes them into realities." To violate liberal rights is to disobey the liberal state. In a sovereignless condition, rights can be imagined but not experienced. In a society with a weak state, such as Lebanon for the past decade, rights themselves are weak or underenforced. Statelessness means rightlessness, as the story of migrating Kurds, Vietnamese and Caribbean boat people, and many others should by now have made abundantly clear.
The positive correlation between individual rights and state capacities is an important theme in the history of liberal thought. An emblematic figure in this regard is Pierre Bayle, one of the originators of the liberal defense of religious toleration. Bayle was a theorist of toleration. But he was also an "absolutist," that is, an advocate of increasing the powers of the crown or centralized state. The logic of his position may seem anomalous to those who understand liberalism as vehemently antistatist. But the Baylean linkage of liberal rights to sovereignty is actually quite straightforward. Only a powerful central state can protect individual rights against local strongmen and religious majorities. Only a powerful state can defend the weak against the strong. In France, to be more specific, only a powerful state could resist the pressure of ecclesiastics to persecute the Protestant minority. (Bayle was a Protestant.) Historians of liberalism, as I said, tend to repeat that liberalism was born in protest against state power. This is an accurate but one sided picture. Bayle's tolerationism was born in protest against a lack of state power. His liberalism was most lucidly displayed in his plea to extend the protection of the secular state to a beleaguered sect.
Libertarian rhetoric about "getting the government off our backs" makes the positive correlation between individual rights and state power difficult to comprehend. Better guidance comes from classic liberals, who insisted that, when organized constitutionally, liberty and authority can be mutually reinforcing. Consider David Hume's famous essay, "Of Commerce." In this classic defense of liberal political economy, Hume argues that Britain should deregulate commercial and industrial life and welcome the accumulation of private wealth, because such a system will increase the resources "to which the public may lay claim." An autocratic government, intent upon controlling all economic life, will decrease the stock of private wealth and thereby indirectly undermine its own power. It is a lesson that the leaders of the Soviet Union have absorbed only today.
Not private property alone, but all typically liberal institutions were partly justified because they strengthened the state's capacity to govern and solve collective problems. Consider, by way of illustration, one of the most fundamental institutions of liberal constitutionalism: freedom of discussion. We might justify freedom of speech by arguing that the sphere of individual liberty must be maximally expanded while the sphere of state power must be contracted to a proportional degree. But liberals thought about free discussion in another way. Immanuel Kant, for example, argued that a government could not stifle freedom of the press because, by so doing, it would lose access to vital information and undermine its own capacity to govern.
Early modern theorists of absolutism, such as Jean Bodin, were the first to focus attention clearly on the advantages of liberty for the power of the state. In his Six Books of the Republic (1576), Bodin provided a series of extremely influential raison d'etat arguments for constitutional limits on governmental power. Limited power is more powerful than unlimited power. That is the master thought of his great treatise and his principal legacy to the liberal tradition. Consider the following example. One of the most acute problems any ruler will face is control of his own officials. How will the king know what his agents are doing, especially in remote regions of the kingdom? Are they exceeding his commands? Are they taking bribes to enforce his laws selectively? How can such information travel from the periphery of a large kingdom to the center? How can a ruler monitor the activities of his "staff" without creating another "staff" whose activity would also be suspect and need monitoring? How can the king hear anything his inner circle of advisors do not want him to hear? Bodin's answer to all these questions is that if a king wants to learn promptly about the misdeeds of his own agents, he ought to create an assembly where representatives from the entire kingdom can come together and complain openly under a grant of immunity. Freedom of speech in a national assembly is an indispensable tool in the modern "art of governance."
What can we learn by looking at preliberal arguments for typically liberal political institutions? We can learn, I think, to question the conventional interpretation of classical liberal theory as ardently anti-statist. Liberals were not anarchists. They were opposed to capricious and oppressive authority, not to authority in general. They embraced state power as a means both to prevent anarchy and to enforce impartial laws (against the grain of human partiality). Because they assumed that political rulers will themselves be human, and therefore partial and potentially unjust, they also devised institutional machinery to contain authority within legal channels. The constitutionalizing of authority is anti-authoritarian. But it does not imply a weakening or crippling of the state.
As the countries of Eastern Europe struggle to establish constitutional democracies in difficult circumstances today, we should ask ourselves again how the United States managed to stabilize a liberal-republican regime at the end of the eighteenth century. The endurance of the Constitution written at Philadelphia in 1787 was not foreordained. The members of the Constituent Assembly in Paris in 1791, whose ideals were not radically distant from those of the American Founders, produced a respectable liberal constitution that guttered to a swift and miserable end. Why did the Americans succeed and the French fail? There are many reasons, of course, stemming from the different political, religious, economic, demographic, and military situations of the two countries. But one neglected reason deserves to be pointed out. Unlike their French contemporaries, the American Founders wrote our Constitution after a period of frustration with the weakness of central government. They aimed, therefore, not only to prevent tyranny, but also to create a sturdy government with the capacity to govern effectively and "promote the general Welfare." This devotion to governmental effectiveness was virtually absent at the French Constituent Assembly of 1791. The desire simultaneously to limit and reinforce the state resulted, in the American case, in a constitutional regime that was neither tyrannical nor weak.
When granting powers to the government, the American Founders looked for guidance to the great European liberals. What powers should government be assigned, from a liberal point of view? These powers were not trivial. The power to defend the country from foreign invasion is usually mentioned first. But what should be the domestic powers of a liberal state? Liberals, for one thing, expected the government to provide security from private as well as public violence. In other words, they sought to create not only a police force, but also the mechanisms for monitoring and controlling the police. The liberal state was also expected to define property rights and enforce property law, contract law, and trespass law. No rules for the inheritance or conveyance of property existed in the state of nature; they had to be created by political means. Civil society, therefore, was society "civilized" by the state. The state imposed civilization on society, among other ways, by constantly breaking up spontaneous economic and social monopolies.
But the power of the liberal state did not stop here. Constitutional government also had a significant allocative role. It had to make available judicial institutions for private litigation. It had to deliver fair procedures in criminal cases, allowing a reasonable defense for the accused. It had to provide poor relief. And, of course, it had to provide a whole series of public goods: canals, highways, safe water, street lights, sewers. State-help was conceived as providing the preconditions for self-help. This idea was nowhere more apparent than in liberal advocacy of subsidized education. Adam Smith, for one, favored a publicly financed system of compulsory elementary education aimed to help the indigent. John Stuart Mill, too, held it to be "the duty of the government" to supply "pecuniary support to elementary schools, such as to render them accessible to all the children of the poor." Far from being a road to serfdom, government intervention was meant to enhance individual autonomy. Publicly financed schooling, as Mill wrote, is "help toward doing without help."
From a Marxist perspective, the primary liberal right is the right to economic liberty: the right to own property, to make contracts, to enter into business, to buy and sell, to exchange goods and services. But is it true that liberals viewed economic rights as somehow primary or exemplary? According to Max Weber, freedom of conscience was the first and basic liberal right. More generally, liberals embraced religious toleration, freedom of discussion, the right to criticize government officials, and prohibitions against bodily torture for their own sakes -- not merely because these practices were good for trade. However important, economic liberty was merely one of the core practices valued by liberals.
This is not to deny the emphasis that many liberals placed on market freedoms and economic self-interest. The role of self-interest in liberal theory, however, has been poorly understood. All too often, commentators assume that liberals who adopt a friendly attitude toward self-interest are advocating some sort of hyperegoistical attitude in which nothing matters but the pursuit of personal gain. This is an overly theatrical view of the liberal tradition.
Liberals are often accused of psychological reductionism. They purportedly believed that human beings are propelled by rational self-interest alone, as if benevolence, love of others, and devotion to the common good were wholly unreal motivations. This accusation is reckless. The truth is that, before the nineteenth century, motivational reductionism was virtually unknown. Most human behavior was understood to spring from irrational passions. Rational choice of action was exceptional. Self-destructive and wasteful conduct was rampant. Most individuals were compulsive or impulsive, hide-bound by habit or victimized by passing frenzies. For neo-stoics, calculating and self-interested behavior was a rare moral ideal. It could be achieved only by a few philosophers after a strenuous process of moral discipline, whereby irrational passions were systematically weakened and purged. It could certainly not be expected from everyone.
Given this cultural background, it is implausible to assume that classical European liberals were motivational reductionists. Their focus on calculating self-interest must be understood in a subtler way. It was not a descriptive claim, first of all, but rather a normative recommendation. We can see this most clearly in the work of the greatest of all preliberal theorists, Thomas Hobbes. If human beings were rational pursuers of their own self-interest, Hobbes reasoned, history would not be an endless chronicle of wasteful butchery and self-destruction. Civil wars are so frequent because some individuals are prepared to risk death for the sake of "higher" ideals such as glory and salvation. To eliminate the destructive violence of civil war, it is crucial to discredit all ideals that tempt individuals to defy death. In Hobbes's ideal society, people would rationally pursue their self-preservation, oblivious to the siren-songs of aristocratic glory and religious redemption. While he favored such a purging of irrational motivations, Hobbes did not believe that most people could be wholly rescued from preposterous habits of mind. He cannot be accused of thinking that human beings were self-disciplined enough to be rational and self-interested in an emphatic sense.
The same is true of the liberals who built upon Hobbes's thought. Montesquieu, for instance, believed that economic growth would discourage irrational and self-destructive behavior. It would weaken the vise-grip of xenophobia and bigotry. It would incite forethought and sharpen people's awareness of the remote consequences of their actions. Neither Montesquieu nor any other liberal, however, thought that human beings were programmed at birth to be calculating maximizers of individual well-being.
There is also a moral component to the liberal emphasis on universal self-interest. To say that all individuals are motivated by self-interest is to assert that, from a political perspective, all human beings are fundamentally the same. For political purposes, no individual can claim to have motives that are morally superior to his neighbor's. There are no higher types. Everyone has interests. And one individual's interests are, in principle, as worthy of satisfaction as another's. The right to rule cannot be grounded on natural superiority. All people, including rulers, are driven by self-interest. As a result, constitution-makers must design institutions (such as periodic and competitive elections) to make the interests of rulers coincide with the interests of the ruled. The concept of universal self-interest can thus be said to provide the anthropological foundations for democracy.
The concept of self-interest, in short, contains an implicit reference to some sort of universalistic or egalitarian norm. Moreover, liberals uniformly took an additional step and endorsed the norm of fairness even when it spelled coercive checks on the principle of self-interest.
There is nothing shameful about the pursuit of personal advantage. That concession is a distinctive innovation of the humanism to which all liberals subscribed. Self-interest is not a sign of moral depravity or cowardice (though it does express a novel insouciance toward would-be social "superiors"). Advantage-seeking nevertheless presents an important social problem. Government is necessary, as Locke and the others argued, precisely because individuals are partial to themselves. Liberals worried about self-interest, then, because they were conscious of the damaging effects of human partiality. A self-interested individual will prefer that everyone else obeys the law, while he or she continues to disobey it. Such an arrangement would be in the individual's private interest, but it would also be wrong from a liberal point of view. To benefit from the self-restraint of others, while continuing to benefit from one's own lack of self-restraint, is flagrantly unjust or unfair. Individuals who exempt themselves from otherwise universal constraints implicitly assert, contrary to liberal principles, that they are special, superior, higher types.
Liberalism is a norm-based, not an interest-based, theory. The self-exemption taboo is the first core norm of liberal thought. This norm -- the injunction to play by rules which apply equally to all -- was most systematically expounded by Kant; but it is unambiguously advanced in the works of all liberal theorists. It would be in the interests of each of us to make an exception of ourselves; we would prefer to free ride on the taxes paid by our neighbors or to break the speed limit whenever convenient while benefiting from the well-monitored driving of others. But we cannot be permitted to do this because self-exemption from generally valid laws would be unfair. For liberals, in short, a norm of fairness overrides the motive of self-interest.
Classical liberals uniformly believed that rights are specified and maintained by state power. And they saw economic liberty as only one kind of right, of no greater importance than, say, freedom of speech, the right to a fair trial, freedom from bodily fear, freedom of conscience, the right to vote. In "On the Jewish Question,"the young Marx accused bourgeois rights of destroying community. It has never been dear, however, why limitations on the discretion of armed policemen should be anticommunal. In fact, liberal rights do not protect the atomized individual from society. They protect fragile channels of social communication (such as the press) from being infiltrated, controlled, and destroyed by political authorities. Authoritarian regimes, based on fear, are much more "atomistic" than liberal societies organized around rights.
One of the greatest obstacles to a fresh understanding of liberal rights is the tyranny of false polarities. Political theory lives in thrall to a sequence of binary schemes: individualism versus community, self-interest versus virtue, negative liberty versus positive liberty, limited government versus self-government. Indeed, the history of modern political theory has recently been reconstructed as a running battle between two rival traditions: liberalism versus republicanism. Republicanism, it appears, was everything that liberalism was not. Supposedly, republicans believed in virtue, community, and citizen involvement in politics, whereas liberals were devoted to base self-interest, personal security, and private independence. This stylized contrast between liberalism and republicanism, however, does not provide an accurate picture of the real alternatives that confronted seventeenth- and eighteenth-century political thought or that confront us today.
One source of the misleading antithesis between liberalism and republicanism seems to be Isaiah Berlin's famous essay on negative and positive liberty. The faults of this stimulating essay are well-known. For one thing, Berlin employs the term "positive liberty" in an ambivalent sense. He uses it to mean both the romantic realization of the real self and democratic self-government. This is an unfortunate conflation, since collective self-rule and individual self-fulfillment have no necessary connection with one another.
The second major problem with Berlin's position is his claim that negative and positive liberty are logically, institutionally, and historically unrelated: it is easy to have one without the other. This is a perplexing claim. Political participation has plainly proved to be an indispensable tool for protecting individuals against capricious, corrupt, and tyrannical government. That, after all, is the essential meaning of "no taxation without representation." Similarly, the protection of private rights provides a crucial precondition for "positive liberty" in both of Berlin's senses. If the police can knock down our doors at midnight and drag away our families to unknown dungeons or graves, our chances for "personal fulfillment" will be drastically reduced, as will our desire to participate actively in political life. There is no good reason, again, why controls on police misconduct should be considered either undemocratic or unromantic.
The German liberal Alexander Humboldt argued that limited government makes self-realization possible. His contemporary, James Madison, argued that limited government makes collective self-rule possible. Both claims are reasonable and, taken together, provide good grounds for doubting the adequacy of Berlin's scheme. Madison's argument runs as follows. All attempts to organize popular government in the past have failed miserably. Republican regimes seem doomed to collapse into factionalism and anarchy. To escape from the intolerable chaos, some sort of Caesar or Cromwell is inevitably handed absolute power. The constitutional problem was how to thwart this powerful historical pattern. How could the Americans design a popular government that, unlike all other republics throughout history, would have a decent chance to survive?
Liberal rights are democracy-reinforcing. For self-government to endure, Madison reasoned, it must be limited government in a special sense. For the will of the majority to prevail, outvoted minorities must be willing to comply with electoral results. They must not resort to violence whenever they lose an election. To purchase minority compliance, the electoral majority must assure the electoral minority that its most precious values and rights will not be violated. A baseline of universal security, provided to all citizens, will create a willingness among the outvoted to acquiesce peacefully in the decisions of the majority.
Madison's famous sensitivity toward property rights should be seen in this light. While expecting and encouraging a greater diffusion of property ownership, Madison foresaw the maintenance in the United States of an important distinction between the rich and the poor. For a popular government to endure, the mass of poorer citizens must keep the confidence of the wealthy. Without the willing cooperation of the rich, no system as inherently unstable as collective self-rule could possibly last. If property-holders believe that democratic procedures will lead to confiscatory policies, they will not go along. They will sabotage the workings of popular government. The likely outcome is class warfare, anarchy, and the universal call for a dictator-on-horseback.
This line of reasoning may seem excessively cynical. Are property rights merely concessions that the many make to the few in order to purchase their cooperation in the workings of popular government? The impression of cynicism is mitigated, however, once we recall that Madison also assumed an economic rationale for property rights. They are productive, not merely protective; they contribute to overall prosperity, enhancing the well-being of the poorest members of the community.
Madison's argument suggests not only that positive liberty is a necessary precondition for negative liberty, but that negative liberty is a necessary precondition for positive liberty. A market economy alone cannot guarantee a democratic government or liberal legal institutions. (Germany has notoriously had more or less the same "market system" under the Kaiser, the Weimar Republic, Hitler, and the Federal Republic.) Without decentralized economic power, however, liberal democracy is very unlikely, if not wholly unable to survive. Economic liberty is a necessary but not sufficient condition for the creation and endurance of a liberal-democratic or liberal-republican regime.
Negative liberty does not refer solely to economic freedom. It refers, even more essentially, to personal security. In a liberal state, individuals generally assume that, if they obey the law, they will not be harassed, tortured, or killed by the police. The logical, psychological, and historical connection between negative and positive liberty becomes even more persuasive when we take this into account.
Individuals want their private rights protected because, among other reasons, personal security allows them (1) to exercise their virtues and realize their potentials, and (2) to participate without inhibition or fear in public debate and processes of collective self-rule. Liberalism's critics seldom take the virtue-fostering and democracy-enabling functions of private rights into account. Citizens will not throng voluntarily to the public square, as I said, if their homes can be ravaged at will by the police.
Note, finally, that negative liberty makes another crucial contribution to democratic government. Individual rights of religious conscience and group freedom of worship do not merely protect a non-political (but still social) sphere. They also help keep a divisive issue off the political agenda. By privatizing religion in a multi-denominational society, liberal freedom helps make public discussion and majoritarian decision-making more effective. By securing a "private space" for religious activity, constitutional government encourages citizens to engage in mutual learning and cooperation on a whole range of non-sacred issues. Separation of church and state unclutters the democratic agenda and creates an opportunity for collaboration across sectarian lines. Again, because rights are protective, they can also be productive.
The claim by the critics of classical liberalism that it was inherently hostile to democracy is baffling from a historical point of view. First, the only countries in which the majority of citizens has any chance at all to exercise influence on political decisions are those with liberal-constitutionalist regimes. Second, the democratization of the suffrage in the West did not seriously threaten the primary predemocratic liberal gains: religious toleration, freedom of the press, constraints on police misbehavior, freedom of entry into occupations and trades, and so forth. So from whence derives the myth that liberalism and democracy are mutually exclusive?
For one thing, liberals have always viewed political participation as voluntary and part-time. In a large nation densely populated with busy citizens, collective decision-making can occur only in a representative assembly, informed and stimulated by national discussion conducted by means of the free press. Those who identify democracy with direct full-time, obligatory participation in public life, therefore, have traditionally smeared liberals as anti-democrats. Aspiring to an unrealizable ideal, they condemn liberals for the sin of being practical-minded. To be sure, liberals did not look back with nostalgia to the ancient Greek polis. But was their skepticism, in this regard, antidemocratic? On the contrary. While admiring the extraordinary freedom of discussion in the Greek assemblies, liberals did not want to imitate most of the other characteristics of impoverished, slave-holding, militaristic oligarchies that managed to consume themselves in class warfare.
Another common argument for liberalism's antidemocratic bias relies upon the historical debate over the restricted suffrage. Liberals did have doubts and worries about majoritarian politics, and some were legitimate. Mill, for instance, was naturally distressed at the election of Louis Napoleon to the presidency of the Second French Republic by universal manhood suffrage. Nevertheless, despite practical reservations, liberals provided a strong theoretical basis for democratic politics as it eventually developed.
To be free, for liberals, was to obey laws made by oneself or one's representatives. That, in fact, was the second core norm of liberal theory. "Freedom to choose," liberals argued, includes freedom to choose the laws under which all citizens must jointly live. It is not surprising, therefore, that the two main political institutions of every liberal regime are the suffrage and a representative legislature. Citizens are bound to obey only those laws made by those expressly authorized to do so, laws made by legislators whom the electorate can oust from office if it wills. Locke insisted that the legislative power is "but a delegated Power from the People" and that "the Legislative being only a Fiduciary Power to act for certain ends, there remains still in the People a Supream Power to remove or alter the Legislative, when they find the Legislative act contrary to the trust reposed in them."
That is the classic liberal formulation of a right of rebellion. Democratic politics, as we now know it, is but a routinization of this fundamental liberal right. Our legislators are our trustees whom we remove from office when they violate our trust. Madison was simply following his liberal predecessors when he asserted that "a dependence on the people" is "the primary control on government," more important even than the separation of powers.
The third core norm of liberal theory and one that again reveals the interconnection between liberalism and democracy, is the idea that public disagreement is a creative force. The standard view, among preliberal political theorists was that uniformity of belief is necessary for social order. John Milton was one of the first to reject this traditional idea, scorning what he called "obedient unanimity" and "a grosse and conforming stupidity." The creativity of public disagreement is the theme of Milton's Areopagitica (1644). This improbable idea had an earlier incarnation, among other places, in a few city-states of ancient Greece. But it was only encoded in the political systems of large nations during the liberal period. It is so radical that not even Rousseau, the father of modern radicalism, accepted it.
Liberal rights, as mentioned, are not only protective. They are also productive. The purpose of freedom of speech, from this perspective, is less the protection of individual autonomy than the production of intelligent political decisions. Participation in "the free market of ideas" does not guarantee self-fulfillment or emotional identification "among citizens. Instead, it is a technique designed to enlist the decentralized imagination and knowledge of citizens, to expose errors, and to encourage new proposals. The free market of ideas is an implicitly egalitarian idea, moreover. It assumes that every citizen can, in principle, make a useful contribution to public debate. Milton wrote of "the voice of reason from what quarter soever it be heard speaking." A regime built around free-wheeling debate, finally, can be said to be based on "human nature" in a loose sense. Human beings are animals capable of self-correction. Government-by-discussion is a political embodiment of the elemental human capacity to learn from experience and repair mistakes.
Suppose you wanted to create a political system for a large nation in which the majority would have a chance to influence public policy. What would you do? You would certainly avoid giving excessive power to urban mobs, who never represent more than a slim minority of the population. The only technique available to you would be electoral politics of some sort. Liberal universalism implies that every individual's vote should count the same. The only morally justified decision-making rule in liberal politics, therefore, is majoritarianism.
Liberal polities, however, while based on free and periodic elections, have all instituted a variety of restraints on majority rule. How can these limits be reconciled with a commitment to democracy?
Liberal limits on the power of electoral majorities have basically five justifications. The first three follow directly from the majority principle itself. First, the present majority must not be allowed to deprive future majorities of the right to correct earlier mistakes. Second, the present majority needs the willing cooperation of outvoted minorities, whose personal rights must therefore be protected. Third, without freedom of debate, shielded from majority censorship and bullying, elites will capture power and ensconce themselves beyond criticism, eventually confiscating the majority's own power.
Finally, there are two powerful non-majoritarian principles at work behind liberal limitations on majority rule. One is the norm of fairness. Majorities cannot be allowed to apply laws selectively or unequally. The second is the idea that political decisions will be more intelligent if produced by a process of wide-open debate and subjected (even after they are made) to an ongoing process of criticism. The majority cannot silence its critics, even if it would love to do so. This prohibition insures that its decisions are more thoughtful and informed than they otherwise would be. The latter restrictions are indeed anti-majoritarian. But they are not antidemocratic if "democracy" includes -- as it surely does -- both equality before the law and government by discussion.
One additional point needs to be made about majority rule. Both conservatives and radicals enjoy citing the passage in The Federalist Papers where Madison writes that "the people in its collective capacity" should have no role in political life. This phrase has always seemed oddly dissonant with other passages where Madison insists that the constitution being framed will create a "popular government." The solution to this paradox lies in the nature of majoritarian politics. As I mentioned, the only way to give power to the majority is through elections. To allow "the people" to act collectively, like the Roman mobs, is to disenfranchise the majority. Only when the people act individually as voters, rather than collectively as a mob, can some influence of the majority over the long haul be secured.
The problem with this arrangement, of course, is that the power wielded by voters on election day is relatively feeble. To make popular participation compatible with majority rule, liberalism makes it dangerously weak. What, then, can be done to increase the marginal leverage that a majority exerts through popular elections?
To this problem, liberals provided a classic solution: the separation of powers. The separation of powers was a liberal application of the old maxim: divide and rule. Traditionally, the divide et impera strategy had been employed by tyrants against their restive subjects. Liberals boldly turned this technique into a tool that the people could use against their rulers. By introducing internal divisions within government, liberals did not simply want to prevent tyranny. They were trying to create a regime that was relatively easy to influence from the outside. A "balance" is not stable. On the contrary, it can be upset by a grain of sand. Thus, a multi-branch and multi-level government provides a sensitive barometer for registering changes in public opinion, electorally expressed. This analysis, brilliantly advanced by Hamilton in The Federalist, number 28, suggests again that liberalism and democracy, far from being enemies or rivals, are mutually reinforcing.
Transfer programs presuppose the continued existence of private property. Only a robust market economy, relying on individual incentives, can produce a surplus worth distributing by political means. In fact, welfare measures were originally proposed by liberals to improve liberal economies and enhance their chances of survival. Such policies were not designed to create equality of resources but only a "bottom floor" under which the indigent would not be allowed to fall. This is why communists, who favored collective ownership, were consistently opposed to welfare. They despised the incipient welfare state because they saw it as hostile to collectivism -- as irredeemably liberal.
The historical relation between eighteenth-century rights and modern welfare entitlements is somewhat obscure. But the case for some sort of continuity between the two is quite strong. Spinoza's assertion that "the care of the poor is incumbent on the whole of society" is echoed by every major liberal theorist. One of the key liberal values, moreover, was security. True, this concept originally referred primarily to protection from violence. But as the resources of liberal societies expanded enormously, it was only natural for the concept of security to be gradually stretched to include unemployment insurance and other programs of "social security."
The government should protect citizens from force and fraud, libertarians argue, but it should take no "positive" action. Individuals should not be pampered by the nanny-state. This way of conceiving our constitutional system is inadequate. All liberal rights, including those enshrined in the first Ten Amendments, are exercised on the basis of resources furnished by the state. The government alone is in a position to define and enforce property rights, for example. And what other institution can provide citizens with a sense of physical security? (The social contract, according to Locke, required individuals to surrender the right of violent self-defense to the state.) In a liberal society, self-help always depends upon state help. There is nothing at all illiberal, then, about the idea of an entitlement. Liberal citizens are entitled to a fair trial and to a high school education, for example. Do libertarians doubt this? The widely endorsed plan to give vouchers to every school-age child is illuminating in this regard. In the public discussion about vouchers, all sides on the ideological spectrum plainly accept the liberal state's duty to provide a minimum level of resources to all citizens. Entitlements to affirmative state action, then, are a staple of the liberal tradition. The controversy begins only when we ask: what sort of help, and how much, should the government provide? This is a proper topic for political debate, a debate that cannot be preemptorily dosed by the assertion that our liberal Constitution forbids the government to provide individuals with resources of any kind.
Liberalism is individualistic. Liberals believe that individuals should be rewarded for achievement and merit. In no liberal society, however, are benefits and burdens allocated wholly on the basis of individual desert. Many of society's delights are purchased by means of inherited resources while its headaches fall disproportionately on those who are born without. The current rate of black infant mortality is only the most shocking example of a nonindividualistic pattern in the allocation of social goods. The decisiveness of inherited resources in all modern societies, in fact, presents a huge dilemma for liberals. We cannot justify the vastly unequal distribution of inherited resources (including parental attention) on individualistic grounds. No infant deserves either to be reared in luxury or to shiver undernourished and poorly clothed in a dangerous and drug-ridden tenement. Liberals will argue, of course, that inheritable property is indispensable for maintaining the prestige and cohesion of the family (the best environment we know for the socialization of individuals). They will also point out that the right to bequeath is itself a form of liberty, that it provides an incentive for industriousness and savings, and so forth.
But liberal inheritance law remains a radical concession to an institution that is not truly individualistic at all. Thus, liberals have naturally sought to redeem individualism by providing some life-enabling means to children who are born (through no conceivable fault of their own) without inherited resources. Child nutrition programs, according to this line of reasoning, are redistributive and yet wholly individualistic. They are attempts not to create a society of equals, but simply to compensate in a modest way for a maldistribution of inherited resources, difficult to justify on liberal grounds.
Liberalism is also universalistic. No individual has, by nature, greater entitlements than another. Human equality extends across class, ethnic, racial, and religious lines. It also extends, and for the same reason, across borders. Why should someone starve because he or she happens to live on the wrong side of a political frontier? Can this be "justified" within a liberal framework? No. There is no moral reason for such grim fate. But there is, alas, brute necessity.
Liberals have succeeded in realizing some of their ideals. But they were able to do so only because they willingly compromised with the realities of national sovereignty. Liberal rights are meaningful only within the confines of a liberal state, only where there exists a rights-enforcing power. To the extent that no enforcing power operates between states or across borders, liberal rights are futile.
So, liberals reason in the following manner. It is morally obligatory to secure a "bottom floor" of subsistence to all humanity. Unfortunately, it is unrealistic to attempt domestic-scale redistributions across borders, not only because of scarce resources, but also because of the location of sovereign power. Hence, although we may urge benevolence toward the poor beyond our borders, enforceable welfare rights will remain limited to co-nationals.
Contrast this argument to communitarian thinking about welfare. Communitarians argue that we owe special attention to co-nationals, not on practical considerations, but on grounds of solidarity: We are morally obliged to love our countrymen and to prefer them to foreigners. A liberal would demur.
Being practical, of course, liberals are willing to compromise with human passions. If a sense of "common identity" makes inhabitants of Scarsdale accept income transfers that benefit inhabitants of Harlem, there is nothing objectionable in that. A sense of shared nationality does seem to mobilize support for economic redistributions. (European countries with higher degrees of ethnic homogeneity are more successful in winning electoral support for transfer programs than welfare advocates in the ethnic crazy-quilt of the United States.) Such communitarian considerations may be very useful strategically. They do not, however, provide a moral reason for redistribution from a liberal point of view.
The classical liberals were reformers and social critics. They were not hand-holders and flag-wavers for established regimes. Today, no liberal in the United States would advocate a wholesale remaking of our constitutional, legal, and economic system. But neither can a liberal heir of Locke and Mill ignore the painfully illiberal features of our society. In many places, urban violence makes a mockery of the promise to protect every citizen from physical fear. The homeless are deprived of the elementary security a liberal regime owes to all. Decaying schools represent a national betrayal of liberalism's pledge to the next generation. The steady increase of children living in poverty conflicts rudely with a liberal commitment to equal opportunity. The rising costs of litigation have thrown into doubt the principle of equal access to the law. Rising campaign expenditures suggest that economic inequality is being converted directly into political inequality, against all liberal norms. And how can liberals accept the continuing marginalization of women from positions where political and economic influence is wielded? Finally, black Americans still live to an appalling extent as a stigmatized caste. Infant mortality, poverty, unemployment, school and housing segregation, and reduced access to health care all indicate that social resources are being allocated according to skin color, not along individualistic lines.
A reconstruction of the liberal tradition cannot provide recipes for solving stubborn problems such as these. But it can help us understand why they are problems from a liberal point of view. And it can embolden us to reaffirm today the aims that liberals have traditionally pursued. Those aims are difficult to realize, but not Utopian. A liberal nation is a nation that keeps them steadily in sight.