Civil Rights

The Libertarian Quartet


The Structure of Liberty: Justice and the Rule of Law, by Randy E. Barnett, New York: Clarendon Press, 347 pages, $29.95

Randy Barnett's new book, The Structure of Liberty, weaves together the two main strands of its distinguished author's career. In the realm of the practical, Barnett has drawn on his extensive experience as a state's prosecutor in Cook County, Illinois. As a legal theorist, Barnett (now a law professor at Boston University) builds on the great writers of the liberal tradition–Hobbes, Locke, Hume, Hayek, and Nozick–for his own theoretical defense of the rights and duties that all individuals owe each other as a matter of natural law. He then uses his judgments on rights and duties to define the province of a properly limited government's activities. Barnett's instincts should be more widespread today, when lawyers, philosophers, and policy makers automatically posit a government solution for any perceived social failure. His interest in basic theory as it relates to the uses and abuses of political power makes his views on a wide range of state policy issues, from taxation to criminal law, worthy of careful attention.

In natural law tradition, the purpose of government is to address the downside of human nature, which in Barnett's view requires a proper response to the problems of knowledge, interest, and power. The problem of knowledge is the difficulty of understanding other people's subjective preferences and predicting the complex forms of behavior to which those preferences will lead. The problem of interest is the natural partiality that persons give to their own concerns relative to those of others. Finally, the problem of power concerns the dangers of excessive or inadequate enforcement of legal norms.

Answering these three challenges defines the three major parts of this book. The solution requires the rule of law: coherent, fixed, intelligible, predictable, prospective, public, and stable rules of conduct with which rational human beings can comply. Only a system that embraces the rule of law can allow individuals to plan and organize their lives in an intelligent fashion.

Once these concerns are identified, Barnett outlines the basic rules of proper conduct to facilitate the flourishing of human beings in a society where scarcity of resources requires the limitation of individual freedom. To his critics on the left, much of what he has to say will seem to come from another century, or perhaps from another planet. I shall not tarry over their objections. But I too have major disagreements with Barnett's theory, because I also think that he tries to get along with a state that is too small, one which cannot discharge the essential functions needed to advance human flourishing or social welfare. The challenge is to point out the weaknesses in Barnett's theory without throwing us into the deadly, all-consuming embrace of the welfare state.

The dispute here goes to the heart of the question of what it means to be a libertarian. Like all great terms, libertarian evokes powerful emotions because it contains deep-seated ambiguities. It could refer to a philosophical system, or it could be identified with a political party whose views are imperfectly aligned with that system. But even if we put politics to one side, the term should be understood in opposition not only to socialism and welfare state liberalism but also to social conservatism of all stripes. Viewed in that context, Barnett and I are as one.

My disagreement with Barnett is not over the primacy of individual liberty as the end which government serves. Rather, it is over the seeming paradox of whether liberty must be limited so that it may be preserved. Barnett sees little place for any such limitations within a system of liberty, and by implication in a system of libertarian thought. My view is that, however indispensable liberty is for the advancement of human welfare, it must, like all great principles, be hedged in by other principles that flesh out a more complete legal system. I see a limited but irreducible function of government in responding to the problems of monopoly and public goods, while Barnett thinks these concerns require no concessions toward a larger state.

Barnett begins by explaining why his view of human nature makes him a natural lawyer. Unlike many of his ilk, he is quite relaxed about this characterization, using homey examples (for which he has a real gift) to illustrate his position. Natural law has gotten something of a bad name in instrumentalist and policy-oriented quarters. It is condemned for appealing to (take your pick) deductive, immanent, necessary, or self-evident truth (or worse, Truth) that retains its plausibility only at a high level of abstraction. Such a charge can be lodged against the abstruse writings of such political theorists as the late Leo Strauss and such modern jurists as Ernest Weinrib and Lloyd Weinreb. Barnett, commendably, has less lofty goals in mind. For him, the phrase natural law helps remind us that salient features of human nature are not easily manipulable. He rightly cautions against the oft-heard claim that human nature is "socially constructed" (it is never quite said by whom) and therefore can be reconstructed in ways that fit contemporary ideals of human (read, gender or racial) equality.

Yet Barnett does not press the point on his skeptical readers as hard as he might, for his main goal is to persuade us that natural law should be understood in a conditional "if-then" sense. Thus, if you want to achieve human flourishing, then these are the rules of conduct that you have to obey. His asserted parallel is to engineering principles, which say that if you want your building to stand, then you had better put your center of gravity over your base; if you want your building to fall down, then by all means have a lopsided overhang. The parallel is instructive, but not in the sense that Barnett intends. There are no laws of engineering; there are only laws of physics that indicate the relationships between distance, mass, time, force, and so on. Those laws cannot be couched as "if-then" instructions; they are the constraints that must be respected for any human project, whether noble or nefarious, to go forward.

The moment, therefore, that we couch human laws in these terms, we are not working parallel to natural (i.e., scientific) laws. Rather, we are asking how we can maximize certain outputs, such as human flourishing, given the constraints that we face. Many of these constraints are imposed by physical laws; others are biological imperatives that have to do with caloric intake, heat retention, and reproduction. But no matter what their source, Barnett's version of natural law, like that of the most successful of the classical liberal writers, becomes in practice nothing more or less than a sensible, constrained form of utilitarianism which measures the success of rules of conduct by the way in which they allow individuals to order their own lives, and groups to order their collective existence.

In this framework, the distinction between the language of human flourishing and that of social welfare becomes important. The former talks about the individual in relative isolation and treats self-realization as the highest goal. But that approach tends to miss the question of conflicts between individuals, which are more squarely addressed under theories of social welfare. These theories compare social states–claiming, for example, that social state A should be preferred to social state B if one person is better off in state A than in state B, and everyone else is at least as well off. Such calculations are not easy. But Barnett does not explain why the individualistic account provides us with better traction for social problems than the more comprehensive accounts, such as the test of Pareto superiority just mentioned, which is commonly used in economic theory.

By stressing the personal account of human flourishing, Barnett fails to discuss with sufficient fullness the key question of which rules of conduct should be individually chosen and which should be legally imposed–that is, backed with the coercive power of the state. Eating three square meals a day, avoiding smoking, and getting enough sleep sound like fine rules of conduct, but only a dangerously authoritarian state (such as ours is now becoming, on the first two points at least) would make adherence to such dietary, tobacco, and sleeping laws subject to collective control.

What, then, should the objects of public force be? Here Barnett draws on the work of the philosopher Hillel Steiner to insist that state power should be directed toward the articulation and formation of "compossible rights." These are the rights which it is possible for individuals whose personal interests sometimes diverge to assert for themselves and to recognize in others in ways that maximize their respective spheres of freedom. At this point in the book, the discussion of natural law recedes into the background, for Barnett is rightly concerned with getting people to buy into his substantive regime, even if they reject his legal metaphysics. He adheres (as do I) to the classical liberal tradition that starts with four simple rules as the keys to organizing social behavior.

The first of these is the principle of individual liberty, which gives to each individual a sphere of control over those matters closest to him. Liberty allows individuals to pick courses of action that advance their own flourishing. Since individuals live not in a void but as physical entities bound in time and space, the rule of first possession allows each person to choose and defend some part of the earth's surface on which he can carry out his own plans.

As social beings, humans understand the mutual benefit that comes from cooperation and exchange; therefore they must have a law of contract that permits them to deploy and redeploy their labor and property in ways that work to their own advantage. The freedom to contract, moreover, covers the right to determine with whom one will contract; rightly understood, therefore, it embraces a freedom from contract as well. And to make sure that liberty, property, and contractual interests are respected, a law of tort (or crime) has to be invoked to punish those who seek to gain advantage by deviating from the accepted rules–that is, who violate the fundamental liberal prohibition against force and fraud.

The case for this libertarian quartet is so powerful that it would be foolish for anyone to mount a frontal assault on it. But that is just the point. The opponents of freedom of contract never take so silly a position as to urge the prohibition of all contracts. Rather, they make selective claims of market failure that are said to justify various forms of state intervention. Barnett's defense of this libertarian quartet is underdeveloped because he does not explicitly address and reject the focused attacks on freedom of (and freedom from) contract that form today's conventional wisdom.

On the top of that list stand anti-discrimination laws, which forbid the use of race, sex, age, disability, and sexual orientation (for starters) as reasons for refusing or failing to do business with someone. I have already climbed to the top of the rafters to denounce these interventions in competitive labor markets as mischievous. We would have more vibrant labor markets by scrapping the entire government apparatus in favor of the 19th-century common law regime that allows people to refuse to deal for good reason, bad reason, or no reason at all. Barnett has to agree with this conclusion, so it would be nice to see him call outright for the repeal of Title VII of the 1964 Civil Rights Act and similar legislation.

That position is not necessarily conservative, since (as I argue in Forbidden Grounds) repealing anti-discrimination laws would undermine the color-blind norm in private competitive industries–a norm that is supported by such prominent conservative thinkers as Abigail and Stephan Thernstrom–and pave the way for private affirmative action programs, however foolish they may seem or be. The decentralization of the affirmative action problem would allow individual firms to take advantage of local knowledge. Research universities, for example, might have different thresholds for affirmative action than regional colleges. Or affirmative action might be introduced at lower cost in the social sciences than in the physical sciences. Allowing different levels of affirmative action for different institutions would help defuse the chronic political tension from pointless presidential commissions that search for a misguided social solidarity on matters best left to the multiple judgments of separate firms.

But on this question Barnett gives us only silence. Ditto for wrongful dismissal suits (detailed in such gruesome particularity by Walter Olson's The Excuse Factory), collective bargaining laws, minimum wage laws, equal pay, family leave, and mandatory pension laws–all mistakes in his view and mine. Opponents of contractual freedom trot out arguments of exploitation and market failure to justify their schemes. To make good Barnett's claim that freedom of and from contract represents sound social policy requires a close, patient analysis of the proposed reforms and the institutional dislocations they cause. Barnett's Olympian detachment sounds ecumenical, but it does not take the fight to where his political opposition lives. His elegant theoretical defense can be airily dismissed by people who assert that this finespun theoretical structure could not survive the pounding it would receive in the real world.

Barnett's adversaries are not solely on the left. One commonly cited source of market failure is the inability of firms to take into account safety risks when they set working conditions for their employees. Enter the Occupational Safety and Health Administration, with its bureaucratic imperatives, nosy inspectors, and mind-numbing regulations. Philip Howard achieved massive and deserved publicity with his book The Death of Common Sense by pointing out the absurdities this system has wrought. One unintended but powerful consequence has been to disrupt local safety regimes with external standards imposed by people who mistakenly think that capital improvements are more effective than worker cooperation in achieving safety objectives.

Yet Howard's book fizzles in the end, because he is unable to rid himself of the thought that OSHA has a useful role to play in workplace safety. He therefore veers away from the evils of inflexibility by embracing a common-sense discretion that allows officials to waive OSHA's rules based on local conditions. But in so doing Howard substitutes one set of vices for another. Which firms get the waivers, and why? Discretion is often exercised in arbitrary and capricious ways that cannot be ferreted out by the crude devices available to the legal system. One connected firm gets a waiver, and it then lobbies fiercely to keep its key competitors from getting the same relief. Barnett could have advanced his case mightily by pointing out that a system based on freedom of contract avoids these multiple embarrassments by reducing tout court the opportunities for political intrigue. And he could have referred to economist W. Kip Viscusi's solid empirical studies, which show that firms face the lash of a hefty wage premium (even with OSHA in place) if they do not tend to worker safety.

My first criticism of Barnett, then, is that he does not tackle head on his chief adversaries in those arenas where the quartet of libertarian principles is able to ward them off. My second criticism cuts in the opposite direction. Barnett does not satisfactorily defend the quartet as the outer limit of state intervention–though, to his credit, he is well aware of the challenges that are raised in both moral and economic theory to this four-part approach.

The first challenge involves the so-called cases of necessity, wherein a person in imminent peril wishes to override the absolute right of an owner of private property to exclude other individuals. The basic virtues of a rule of exclusion for land and chattels is well understood. Without exclusion, cultivation, development, and conservation of natural resources are curtailed, because B can always reap or destroy the crops that A has planted. The necessity exception, endorsed by virtually every legal system that proceeds by the common law method Barnett defends, does not allow B to enter A's land willy nilly. But it sometimes does permit entrance if B must take refuge from an ocean storm, subject to his obligation to compensate A for any loss thereby inflicted. It might not let B enter A's cabin if A is present, but it would allow him to break into A's unoccupied cabin in order to forestall death from starvation or cold, just as it would allow B to enter A's land to save the life of C.

As these examples show, the exact scope of the privilege is hard to define because so much depends on the messy balance between B's need and the strength of A's property interest. But we have to fight through the examples to reach rough cloture. To say that B may be "morally" entitled to take A's resources while denying that B could be "legally" entitled is not very helpful. To say that a slippery slope could lead to the unacceptable expansion of the privilege (which has not been the case) is to ignore the dangers of a rigid rule that allows no privilege at all. Some limitation on liberty and property is recognized in practice. Barnett is mistaken to conclude that the principle of absolute exclusion is deductively necessary. A more nuanced concern with the demands for social utility under extreme conditions would lead to a better result.

The necessity cases represent only the thin edge of the wedge. The greater challenge to the libertarian quartet arises when the stubbornness of a single individual can prevent the implementation of a collective goal that advances the subjective interests of all parties, especially maximizing the output of common-pool resources. A group of landowners sits atop a valuable pool of oil and gas. Many people own property along a river or stream. Many fishermen or hunters have access to wild fish or birds. In each case, the rule of first possession is inadequate to stop the premature consumption of valuable resources. From the earliest times, legal systems have responded to these risks by adopting systems of common property that operate side by side with systems of separate property. Often these common systems are customary in origin, as with the rules that govern rivers and beaches. Sometimes they are developed by painful experience, as with statutory restrictions on groundwater use, fishing, and oil and gas exploration.

In each of these cases, the pattern runs roughly as follows. When resource utilization is low, people see great merit in a rule of capture, which is cheap to enforce and gives clear title to the first possessor. But when utilization rates increase, the first possession rule leads to premature and wasteful efforts to gobble up resources before others do. The capture rule for groundwater yields to some mushy regime of correlative rights and duties. The fishery is subject to catch limits. Oil and gas exploration is governed by spacing and unitization rules that prevent overexploitation. In each of these cases there are smart ways and dumb ways to proceed after the first possession rule is abandoned. But the strengths and weaknesses of the various approaches have to be examined directly. It simply will not do to assume that the capture rule, with its enormous defects, always comes out on top simply because the alternatives are subject to error.

The same difficulty with the libertarian quartet arises with the provision of classic public goods: common carriers, street lights, lighthouses, and the like. Barnett is very skeptical of taxation and regulation because they run smack into the problems of knowledge, interest, and power with which he is rightly concerned. But as with natural resources, it is wrong to assume that the imperfections of the libertarian quartet are smaller than those of some collective solution.

The requirement that common carriers serve all comers (explicitly overriding freedom from contract) is of very old origin, and it represents one possible response to the question of monopoly power. Even today, the strongest defenders of competition in the telecommunications market recognize that some form of regulation is necessary to secure interconnection (with appropriate access fees) between the local exchange carrier that controls the last mile of wire and the long-distance carriers that compete with each other. Volumes could be written about mistakes in the design and implementation of the 1996 Telecommunications Act. But the one objection that does not hold water is the idea that businesses with a legal or de facto monopoly should have the absolute power to exclude.

Barnett does not deal with these examples directly. Rather, he notes, correctly, that the mere presence of a coordination problem does not justify the use of government intervention to whip recalcitrant participants into line. But that skepticism can be incorporated into a more sensible view that weighs the gain from overcoming the monopoly problem against the administrative costs of putting some particular solution into place.

Barnett engages in a calculation of that sort when he relies on economist Ronald Coase's famous 1974 article, "The Lighthouse in Economics," which attacked the orthodox view denying the ability of private firms to provide lighthouses. The beacon so provided has been called a public good that no one will pay for if others will provide it anyway. Since all self-interested parties can take the position of free-riders, the lighthouse will never get built. But it turns out that it does get built, and the fees for it are collected not as ships sail by on the open sea but when they dock at nearby ports.

That is undeniably a clever solution, but it is not one that operates solely within the confines of the libertarian quartet. The only way the fee can be charged is through the exercise of state monopoly power at the port. Indeed, in many cases the task of collection was left to customs officials. No competitive firm could collect that charge from its customers if they were free to go elsewhere. And the port fee system is easily subject to monopoly abuse if the revenues exceed the cost of providing the lighthouse and are diverted to other ends. The private owners of the lighthouse have a monopoly position and, unless some common carrier obligation is imposed, can charge what the traffic will bear. Indeed, the British parliamentary reports from the 1830s that recommended the shift to public ownership of lighthouses noted that the revenues far exceeded operation costs.

For Coase, this problem did not matter, since his main mission was to show that economics textbooks which treated lighthouses as the classic government-provided public good erred because they did not study the actual practices of lighthouses. But for Barnett, these observations are much more damaging, for they show the intractable second-best nature of every public good problem. The only way to avoid the evident dangers and inefficiencies of government funding is to take the risk of private monopolistic behavior. No a priori answer tells us which loss is greater.

More generally, holdout and monopoly problems are very tough, and sometimes they require all sorts of government coercion. To analyze when and how that coercion can be used requires a far more detailed examination of the intricate takings and just compensation problem than Barnett offers in this book. By failing to undertake this task, Barnett misses the opportunity to weigh in against those forms of redistributive regulation–whether through assigned risk pools in insurance or the grotesque universal service obligation that is part of the 1996 Telecommunications Act–that have much more to do with subsidy than with efficient deployment of resources in a network industry. The dangers of misguided regulation are hard to attack credibly simply by declaring every deviation from the libertarian quartet unacceptable. What is needed now is not a blanket rejection but a differential analysis whereby some systems fail and other succeed.

Last we come to the question of the criminal law, on which I shall comment only briefly even though it occupies much of Barnett's attention. As might be expected, he is a strong defender of privatizing the criminal law function, and he relies on economist Bruce Benson's useful work to show the extent to which the private sector in law enforcement has grown faster and performed better than the public sector in recent years. Although that point is surely true, Barnett could have spent at least some time trying to explain whether, and to what extent, improvements in public law enforcement account for some of the rapid decline in the crime rates that we have observed in recent years.

Even if we put that question aside, Barnett makes the same mistake of proportion that occurs elsewhere in the book. It is easy to show that increasing the level of private enforcement in the overall mix can produce substantial improvements, especially against a baseline of a generation ago. But it hardly follows that we should go the whole nine yards and do away with public enforcement altogether.

Indeed, much of the success of private enforcement depends on its ability to rely on a public enforcement backup. Private police may be good at preventing crimes and nudging suspicious types along their way (something the regular police could do if the Supreme Court came to its senses and eased up on its hostility toward anti-loitering laws). But when murders and robberies occur, when individuals cross state lines, when the use of force may be required in a standoff, the private police rely on public enforcement.

The University of Chicago has had extensive policing of Hyde Park for the better part of 50 years, but no one supposes that we would love to see the Chicago police leave the beat. Quite the opposite: Local arrangements are designed to make sure that each investment by the university does not lead to a corresponding diminution of public enforcement in the neighborhood, given the obvious temptations involved. What has to be done is to work on the proper mix of public and private. Law enforcement, like the construction and maintenance of infrastructure, has a necessary public component.

In sum, Barnett's book represents much of what is sound and much of what is suspect in traditional libertarian thought. Its basic instincts on the importance of liberty, property, contract, and tort are undeniably correct. The difficulties start when defenders of this system exaggerate its ability to address the wide range of problems that require some degree of government taking and regulation.

In refusing to address those problems, Barnett reminds me of physicians who defend therapeutic nihilism, claiming that most medical treatments have negative results, so patients are better off with no treatment at all. That position became untenable probably by the end of the 19th century, and certainly after the end of World War I. Yet it holds a lesson for us today: Regulation can kill, just as bad medicine can kill, but it can also serve useful ends.

Today the knowledge required for sensible regulation is available. Incremental improvements are possible; bad schemes can be denounced, good ones improved. That debate is where the action is, but Barnett will have to sit on the sidelines. Regulatory nihilism will not cut it as we enter the 21st century. Regulation beyond the libertarian norms is a necessity. The only question is whether we shall do the task wisely or poorly.

Richard A. Epstein ( is the James Parker Hall Distinguished Service Professor of Law at the University of Chicago. His latest book is Principles for a Free Society: Reconciling Individual Liberty with the Common Good (Perseus Press).