Coercion vs. Consent

The following exchange got underway at the Reason 35th anniversary banquet, which was held last November in Los Angeles. Keynote speaker Richard Epstein, the eminent legal theorist and author of the new Skepticism and Freedom, delivered a provocative talk about the foundations of libertarianism titled "The Ambiguities of Reason: Of Large and Small 'r's." For the spirited dialogue below, we adapted Epstein's comments and invited three responses, which are followed by a final comment by Epstein.

The Limits of Liberty
Why we need taxation and eminent domain
Richard A. Epstein

Perhaps the most fundamental question we face is how to think about liberty. Some libertarians stress the formal power of logic to resolve hard questions. They insist that all rights and duties flow from a necessary conception of individual autonomy or self-rule that allows all individuals to do whatever they wish with their own lives so long as they do not interfere with the like liberties of other individuals. No person may use force or deception against other people, either for his own advantage or for the advantage of third persons.

This moral imperative holds seemingly without regard for its social and economic consequences. Political organizations should adapt to this strong conception of rights and duties, the argument goes, and not yield to whim or fashion. Given this simple premise, individuals may use their own labor to acquire property, to exchange their labor or property with others, or to form complex business, social, and charitable organizations.

This strong intuitive conception of rights and duties tightly corresponds to our ordinary concept of right and wrong behavior. Most people do not seek to order their daily lives by discerning the origins of property in the mists of history; nor do they typically ponder the larger questions of public finance and political organization. But they are taught from birth to be neither bullies nor cheats. In dealing with life-size events, they adhere unswervingly to these simple basic principles. Why then resist their universal application? The content of the rules is clear, and any effort to switch to some calculus that weighs consequences case by case would at best yield indeterminate results, which would in turn heighten overall social insecurity. Better not to scratch beneath the surface. Even if these rules are not necessary truths, we should still treat them as such. Deductive principles order practical affairs well.

Unfortunately, this principle of personal guidance does not supply us with a comprehensive theory of social organization. First, there is the question of philosophical foundations. Can we really support any kind of political order that pays no conscious attention to the consequences it generates? On this point, the ostensibly deductive view is right to shun judging individual actions one case at a time. But this detached form of analysis really should be regarded as a form of closet consequentialism. Setting up public institutions to pass on all individual actions becomes so costly and intrusive that it flunks the standard of good government in just those consequentalist terms. But it is possible to moor this judgment of political structure in a keen appreciation of the mainsprings of human nature, which yields a decidedly mixed picture of the best and worst in human behavior.

We start with the biological observation that no individual could survive in a world of scarce resources without a strong measure of self-interest, one that includes at the very least his own family and close associates. That self-interest can manifest itself in one of two ways when dealing with strangers; through either aggression or cooperation.

The overall social consequences of these two approaches are massively different. With force, one person wins while the other person loses. With cooperation, both persons win. This simple observation underlies the consequentialist explanation for the libertarian preference for agreement over coercion: Take that arrangement that leaves both parties better off than they are under the alternative legal order. Contracts result in joint improvements, such that the greater the ease of contracting, the greater the gains from cooperation. Coercion creates at least one loser for every winner, where the losses (e.g., death, rape, or theft) can be huge relative to the gains on any intuitive interpersonal comparison of utility. When the odds are right, any individual may find it in his interest to use force or deception, but from a social point of view this conduct merits strong condemnation. The basic libertarian imperatives are well-grounded in human nature.

Yet just how far does this insight go in a practical sense? To hang your hat on empirical regularities is to retreat from the language of absolutes and to invite exceptions to general rules. If our ultimate criterion asks what arrangement leaves all parties better off than they are under the next best alternative, there may be cases where the dominance of agreement over coercion should be displaced. In fact, there are: Some contracts are suspect, and some force is justified.

In dealing with ordinary contracts of sale or partnership, we tend to ignore the consequences of these joint efforts on third parties, which any comprehensive social theory should take into account. In most cases, happily, these external effects are positive. Two or more traders not only increase their own wealth and happiness but also expand the opportunities for trade and advancement for others. But a contract to kill a third person has the opposite effect. Indeed, it is precisely because contract yields gains from trade to the participants only that we worry about such agreements, now called conspiracies, because of the threat they pose to the basic rights of liberty and property of others. We are now faced with the difficult practical question of how to identify these rights-threatening arrangements, to punish them if they achieve their object, and, more important, to nip them in the bud.

Concerns about bad contracts are not limited to such situations. Contracts that seek to bribe individuals to violate previous agreements are similarly dangerous. More controversially, contracts that operate in restraint of trade are also possible candidates for special treatment once it is accepted that overall levels of social output are higher under competition than under monopoly. Exactly what should be done with these arrangements, assuming that they can be properly identified, is no easy task, given the political risk that perfectly sensible business arrangements will be attacked by government action -- as is often the case when aggressive competition is branded "predatory" pricing, a dubious appeal to the libertarian norm against predation by the use of force.

Sometimes the strong libertarian synthesis breaks down in the opposite direction. The most conspicuous illustrations are condemnation and taxation, each of which contemplates the use of force against ordinary persons who have neither committed any wrong nor breached any promise. Yet it is in my view impossible to maintain any per se rejection of these two venerable if dangerous institutions, both of which are not only consistent with limited government but required by it.

The libertarian prohibition against force does not take into account the possibility that successful cooperation in key situations can be thwarted by individual holdouts. It will not be possible to build a railroad from point A to point B solely by getting the cooperation of 99 out of 100 private landowners along the way. The last one (indeed all) must be brought into line, and the way to do it is to compel the purchase by paying them the highest value of the land in any alternative use whose value is not dependent on the railroad that is about to be built. The public, including those whose property is condemned, gain the benefit of the railroad, but if compensation is correctly calculated -- a big if -- no individual suffers financial deprivation in the process. State coercion is used to create the win/win situations found in private contracts.

What works in condemnation cases helps explain taxation as well. The public enforcement of private rights and the creation of infrastructure through condemnation both need money that only compulsory exactions can supply. But once the coordination and holdout problems are overcome, much work has to be done to prevent massive abuses from working their way into the system. The flat tax is one sensible limitation on the power of taxation (others can be devised as well), for it allows state funding to vary in amount without picking on one segment of the population.

In sum, the central challenge to any political theory is to devise a set of institutions that first allows and then controls the use of coercion against individual citizens for their own benefit. In light of the justifications that have been put forward here, one could ask the question whether these concessions to state power amount to a backhanded capitulation to the modern welfare state, where any claim of the government to action in the public interest is sufficient to justify state intervention.

The short answer to that question is no. Indeed, there is a strong sense in which exactly the opposite happens. The traditional deductive form of libertarianism allows for state force to protect against aggression and fraud. The more complex version recognizes that state power is also appropriate to overcome holdout problems by the limited use of force. This two-tier inquiry clearly legitimates some forms of government action, but by the same token it makes the case against state intervention stronger in those settings where none of these justifications are available. It is easy to see why the state should keep its hands off the substantive terms of labor contracts in a deductive libertarian world. Hence we should get rid of minimum wage, antidiscrimination law, collective bargaining statutes, and mandatory pension and insurance regulation. Systems of price and rent control similarly go by the boards, as does the full range of tariffs, anti-dumping laws, and other impediments to international trade.

We arrive at the same results even after we recognize the legitimate state role in condemnation and taxation. Unregulated labor and product markets present no coordination and no monopoly problem. The basic libertarian position in favor of competitive markets is thus strengthened by allowing in principle a broader range of state justifications, none of which works in these cases. Removing these ordinary activities from the thrall of government regulation should increase the tax base and thus reduce the need for taxation, while simultaneously increasing the liberty and prosperity of all. The greater level of wealth should in turn reduce the calls for redistribution of wealth by state action, which in turn will reduce if not eliminate much of the welfare state. Government will still be larger than deductive libertarians might want, but it will be far smaller than the current bloated state.

Our limited use of coercion is done with the paradoxical intention of expanding the scope of individual freedom. It is always dangerous business, but it is only with a conscious awareness of how we must both use and limit government power that we shall find the intellectual tools to resist a descent into the all-powerful welfare state. The practical success of our endeavors depends on the ability to avoid not only the dangers of the all-powerful welfare state but also any categorical reluctance to use coercion to initiate forced exchanges that benefit us all.

Richard A. Epstein, the James Parker Hall Distinguished Service Professor of Law at the University of Chicago, is the author, most recently, of Skepticism and Freedom: A Modern Case for Classical Liberalism (University of Chicago Press).

The Lesser Evil
Sometimes the cure is worse than the disease.
Randy Barnett

For most of his career, Richard Epstein has been urging libertarians to expand the exceptions to the prohibition on the use of force beyond self-defense, detention of rights violators, and restitution to include using forced transfers to solve the economic problems of "holdouts" and "free riders." Holdouts would be addressed by eminent domain, free riders by taxation. He justifies both of these expansions on consequentialist grounds.

In his writings on contract law, however, Epstein has insisted that we do not want to make categorical exceptions to contractual enforcement that are so malleable that people can define themselves into them at will. He allows that some agreements can be voided, such as contracts signed under duress or as a result of material misrepresentation, but each justified exception amounts to a direct or indirect way to police force or fraud. As he has written: "The rules in question should not create artificial incentives for parties to lower the level of competence they bring into the marketplace. It is dangerous to allow people to plead their own incompetence in any transaction that they wish, with the benefit of hindsight, to repudiate."

I am not sure he has adequately appreciated how his instincts about contract law defenses apply as well to the categories of free riders and holdouts. It is far too easy to assert the existence of these barriers to trade, and far too difficult to defeat unjust expansions of the use of force when these exceptions are allowed, even if he is right that some such problems are otherwise insoluble. If so, then the cure could well be worse than the disease, and no amount of fancy economic analysis can establish otherwise. The consequence of rent-seeking -- of interest groups using the coercion of the state to acquire unearned benefits for themselves -- matters as much as the consequence of failing to build a road. Ask the residents of Poletown, whose Detroit neighborhood was destroyed by eminent domain to build a General Motors assembly plant. Ask the Atlantic City client of the Institute for Justice who successfully resisted the condemnation of her house to erect a parking lot for Trump Towers.

I know that Epstein's theory of the Takings Clause would restore the requirement of "public use" that would prevent using eminent domain to transfer rights from A to B, and I agree his approach is better than what we now have. But who's to say that these takings for the "public good," as opposed to public use, do not increase aggregate welfare? Who's to say that the welfare created by General Motors remaining in Detroit is not greater than the welfare of the families who must leave their homes? Who's to say that the parking lot to be used by thousands does not create greater welfare than a house used by just one woman?

Due to limitations on our knowledge, we have little choice but to rely on the principle of freedom of contract to answer these questions, however imperfectly. Unlike self-defense and restitution, exceptions for free riders and holdouts cannot be justified as the enforcement of the rights of others. Authorizing force in defense of individual rights is a necessary evil to address the problem of compliance when persons put their own interests ahead of respect for the rights of others -- rights that are themselves necessary, on consequentialist grounds, to solve the pervasive problems of knowledge and interest. Caution should be our guide in pursuit of better consequences than properly defined individual rights provide.

Of course, to some extent this debate is moot. If we ever get to a libertarian world in which these are the only forms of coercion still existing beyond self-defense, etc.,we will know a lot more about how liberty actually works and how to achieve it politically than we do now. We will be in a much better position to decide whether to abolish these practices along with all the other vestiges of the welfare state. I should live so long.

Why then debate them now? For the same reason Epstein has been harping on these points for decades. We debate the form of the ideal end stage as part of the debate over whether to take any further steps in its direction. Epstein clearly believes that a more sympathetic and defensible end state is one in which these additional exceptions for free riders and holdouts exist. On this, I have always had my doubts.

I do not think that Epstein has ever seriously addressed the alternative ways of solving these problems that have developed historically on the market, and which libertarians propose be extended to address the problem of so-callled "public goods." "Public goods" are more a construct than an artifact of the world. It is often only a lack of imagination by academic economists that prevents them from seeing a solution. Indeed, one of the functions of entrepreneurs, as opposed to academics, is to figure out how to make a public good into an excludable private good.

Such techniques include creating the fencing technology needed to exclude free riders (think of the walls around movie screens and theater stages), tying nonexcludable goods to excludable goods, and such legal devices as conditioned contracts that go into legal effect only when a sufficient number of persons have agreed to pay for a particular project. Holdouts are dealt with by real estate developers assembling parcels of land in a variety of ways. I am not sure anyone can prove that these alternatives to takings and taxation will definitely increase aggregate social welfare. But I am certain no one can prove the opposite either.

This is not to deny that consequences matter, a point on which Epstein and I agree. Indeed, I think there are very few libertarians today for whom consequences are not ultimately the reason why they believe in liberty. The issue is always how best to achieve good consequences. As Epstein notes, making no exception to a general prohibition on the use of force is not an option. Self-defense is an exception, as is forcible compensation, and anyone who studies the common law of torts, contracts, and property knows that other exceptions are built right in to the doctrines that define the liberal conception of several property and freedom of contract.

The remaining dispute is over whether we should expand the exceptions to include holdouts and free riders simply because economic theory seems to suggest that only coercion can deal with them effectively. Epstein is convinced. I am not. I would prefer to jump off that bridge if and when we ever come to it, and only after the alternatives are thoroughly explored. I see no reason, whether tactical or principled, to let economic theory trump liberal rights when experience shows these problems are so often solved by entrepreneurs without benefit of any special license to expropriate the property of others without their consent.

In the end, we must never forget that permitting self-defense, restitution, and preventive actions against standing threats gives rise to the problems of power: enforcement error and abuse. Every additional exception legitimating the use of force, such as taxation and takings, further aggravates these serious social problems. The fact that we must take three steps down a dangerous road does not, by itself, justify taking two more, as Epstein seems to imply. Especially when, unlike force that responds to previous violations of rights, the problems of knowledge and interest surrounding these additional exceptions permit enormous opportunities for rent-seeking by those who can credibly claim to be increasing welfare by pursuing the "public good."

Randy Barnett is the Austin B. Fletcher Professor at Boston University School of Law, and the author of The Structure of Liberty: Justice and the Rule of Law (Oxford). His latest book is Restoring the Lost Constitution: The Presumption of Liberty (Princeton).

Swallowing the Rule
Epstein overestimates the power of his proposed limits on state action.
David Friedman

Like Richard Epstein, I find versions of libertarianism that claim to deduce it by straightforward arguments from fundamental principles unsatisfactory. One reason is that libertarians, like other people, have no convincing arguments to show that their principles are true. Another is that concepts such as rights, property, and coercion are very much more complicated, and less susceptible to simple rules and sharp definitions, then such versions of libertarianism generally suppose.

While we cannot logically derive our values, we have them. So do other people. Fortunately, human values vary a good deal less than one might suppose from reading political philosophers. Few egalitarians would prefer a society where everyone had a real income of $1,000 to one where incomes ranged from $90,000 to $100,000. Few Rawlsians would choose to improve the lot of the world's worst-off person by one dollar at the cost of massively reducing the welfare of everyone else in the world. And few libertarians, however hard-core in theory, would choose a perfectly free society of desperate poverty over one slightly less free and very much wealthier. Almost everyone, in my experience, values most of the same things, although not with identical weights. It is easy for both libertarians and socialists to claim to support their principles whatever the consequences -- when each group believes the consequences would be, on very nearly all dimensions, the most attractive society the world has ever seen.

If most people have at least roughly similar values, and if libertarians are correct about what sort of society libertarianism would produce, we need not justify our own values in order to argue for libertarianism. All we need do is to show that a libertarian society would be more attractive, by widely shared standards, than any alternative -- wealthier, wiser, freer, more just, better for poor as well as rich. That is, after all, what most libertarians believe.

Having adopted that strategy, I am sympathetic to Epstein's approach: Derive a legal and political system from the practical requirements for achieving the things humans want to achieve. My disagreement is with his conclusions.

To begin with, he is too pessimistic about the possibility of achieving important objectives without the state. Consider his claim that "the public enforcement of private rights and the creation of infrastructure through condemnation both need money that only compulsory exactions can supply."

The rights half of that claim assumes that private rights must be publicly enforced, despite a considerable number of societies where rights enforcement was produced privately. The infrastructure half assumes that the sort of coordination problems associated with building roads cannot be solved, imperfectly but adequately, by private mechanisms -- despite real-world examples where they have been.

Even if we accept those assumptions, the conclusion still does not follow. To justify taxation we need the additional assumption that rights enforcement cannot be done by the state at a profit, despite historical examples of societies where the right to enforce the law and collect the resulting fines was a marketable asset, and that the government cannot charge enough for the use of its roads to compensate the owners whose land was condemned.

A second general problem with Epstein's argument is that he overestimates the ability of his proposed rules to constrain state action. He writes: "It is easy to see why the state should keep its hands off the substantive terms of labor contracts in a deductive libertarian world….Systems of price and rent control similarly go by the boards, as does the full range of tariffs, anti-dumping laws, and other impediments to international trade."

In a world with income taxes, the state cannot keep its hands off the substantive terms of labor contracts because it has to define which terms do or do not count as income. And it requires only a moderate degree of economic ingenuity to create coordination arguments for state-imposed restrictions on labor contracts designed to reduce the deadweight burden from taxation or solve subtle problems of adverse selection.

Tariffs are an even clearer case. The infant industry argument for tariffs -- the idea that trade barriers are needed to help a potentially competitive industry get started -- can be, and has been, recast as a coordination argument, in which one's firms activities in a new industry are alleged to produce external benefits for other firms in the industry. And it is straightforward to show that a country in a monopoly position as either a producer or consumer can use a tariff or export tax to extract monopoly returns from its trading partners.

There are many other examples of government policies that Epstein does not like but that could be defended on his principles, including government involvement in education, in research, and in the production and regulation of information. His exceptions swallow his rule, leaving us with everything up for grabs -- and familiar public choice reasons to expect that far too much of it will be grabbed.

Epstein hopes to prevent this outcome by better institutional design. Perhaps that is the best we can do. But there are at least two other alternatives worth serious consideration.

The first is the extreme version of the libertarian state: no coercion beyond a monopoly on retaliatory force. Such a state will do less well for us than a state that initiates coercion in the rare circumstances where doing so produces large benefits. But it might do considerably better than the realistic alternative: Epstein's society as we can expect to see it actually implemented, in a world with plentiful arguments for extensive uses of state power and strong incentives to act on them.

The second alternative is to eliminate state coercion by eliminating the state. In that world, some coordination problems will go unsolved, making the result worse than the world that would be produced by a state run by perfectly wise and virtuous libertarians. But eliminating the state also eliminates the largest coordination problem of all: the problem of controlling the state. Given the record so far, that is a more serious problem than how to build roads without the power of eminent domain.

David Friedman is a professor in both the law school and the economics department of Santa Clara University. His first book was The Machinery of Freedom: A Guide to a Radical Capitalism (Open Court). His most recent is Law's Order: What Economics Has to Do With Law and Why It Matters (Princeton University Press). A draft of his next book, Future Imperfect, can be found on his Web page.

Beyond Economics
Freedom is more than dollars and cents.
James P. Pinkerton

Richard Epstein makes the useful point that libertarianism should be embedded in a practical philosophy, and he offers an elegant two-tier approach to deciding when and where to work toward the laudable goal of "expanding the scope of human freedom." I can't quibble with his approach to the issues that fall within his purview, but I also can't help but observe that the most important issues of the day seem to fall outside of that purview. Epstein's circumscribed approach to libertarian philosophy will, I am afraid, also circumscribe libertarianism's appeal and influence.

On the biggest issues of the day, Epstein is silent. I looked in vain for words such as drugs, pollution, immigration, foreign policy, terror, Iraq, or even Bush. That, to me, is the definition of a narrow piece. Not that there's anything wrong with that, but I believe libertarians have an important contribution to make on the hottest of the hot-button issues: drug laws, immigration controls, environmental regulation (including the reality that the United States is involved in a host of international agreements that affect America, no matter what we do), biotech and stem cell research, and, most of all, the "war on terror," which affects everything from civil liberties to federal spending to the ongoing war in Iraq.

By comparison, the issues Epstein wants to grapple with fall mostly within the realm of economics, including the minimum wage, antidiscrimination rules, collective bargaining statutes, mandatory pensions, insurance regulations, price and rent controls, and tariffs. Opposition to all these statist measures is firmly in the libertarian tradition; as Epstein says, it's all part of his plan to "reduce if not eliminate much of the welfare state."

Yet while it's fine to pound away on any and all of these issues one more time, I wonder what the ROII -- Return on Intellectual Investment -- will be. Most people, certainly most economists, accept the general proposition that markets work, and so the fight of the future is over applying Smithian wisdom in specific cases. The Institute for Justice, for example, recently filed suit against the Louisiana Horticulture Commission, which cartelizes florists. Such market-freeing cases are important, and Epsteinian thinking can help. But for the most part, the national agenda has shifted away from economics to other issues that seem more pressing. Indeed, the U.S. seems to have settled into a complacent Clinton-Bush consensus that accepts the idea that if the economy is booming, federal revenues ought to be spent -- and then some. Today prospects for reducing, let alone eliminating, the welfare state seem poor.

At the same time, prospects for expanding the warfare state -- which will, in turn, further ex-pand the welfare state -- seem excellent. In today's America, the spending of blood and treasure for foreign wars -- even those, such as Iraq, that violate international law and are based on government deception -- is practically unquestioned. A government arrogant enough to lie, big time, will never be a modest government. And then, of course, having made more enemies around the world through offensive wars, Washington must spend more on "defense," including homeland defense. Finally, after pledging a welfare state for Iraq (Washington is now a gold rush for lobbyists and contractors brandishing newfound expertise in anything "Middle Eastern"), it will be impossible not to keep and expand the welfare state here at home. Just days after the 1918 armistice that ended World War I, British Prime Minister David Lloyd-George promised "a fit country for heroes to live in." Britain's subsequent socialist bender proved that it's possible for a country to win the war and then lose the peace, and thus betray its heroes.

In today's America, war veterans will soon be granted larger benefits, but such expenditures will only be an overture to the lawsuits some of them are already filing against anyone with a deep pocket. When's the last time President Bush made a real push against the trial lawyers? He can't deal with tort reform or any domestic issue because he's too busy vindicating his foreign policy.

In addition, a government that's "strong" enough to rearrange the domestic affairs of other nations likely will feel equally confident about continuing to meddle in matters that should be private here at home, be they sexual, medical, or pharmacological. Randolph Bourne was so right: "War is the health of the state."

Some libertarians, of course, endorse the Bush Doctrine, explicitly with their words or implicitly with their silence. They argue, in effect, that the maintenance of freedom here requires us to force others to be free. I disagree with this neo-Rousseauean argument; I predict that if the Bush Doctrineers get their way, our future politics will go the cynical and perhaps dirigiste way of France, as every big-governmentalizing action is justified in the name of la gloire -- oops, I mean "democracy."

By all means, let's have a debate about American imperialism. My fear is that if we don't raise our voices, then libertarianism, a la Epstein, could become just a synonym for economics. In which case, we might have prosperity, but we won't have peace, and we won't have freedom.

James P. Pinkerton is a columnist for Newsday and TechCentralStation.com. He is also a contributor to the Fox News Channel and a fellow at the New America Foundation in Washington, D.C. He worked in the White House under Presidents Ronald Reagan and George H.W. Bush and in the 1980, 1984, 1988, and 1992 presidential campaigns.

A Moderate Responds Richard A. Epstein

Of the two basic points I made in my initial remarks, one has escaped serious criticism: that the traditional natural law justifications for freedom are not sufficient to sustain the case for individual liberty or for limited government. We are, as it were, all consequentialists now. The particular debates, therefore, are more focused. Randy Barnett and David Friedman claim that any system of forced exchanges is likely to produce more mischief than it eliminates. James Pinkerton argues that any libertarian theory that obsesses on economics runs the risk of losing the large struggles over peace and freedom. Both criticisms deserve responses.

Barnett thinks state coercion is not justified as a means to overcome major coordination problems because the risk of abuse is greater than the potential gains from the method. But his examples do not support that conclusion. Of course, the citizens of Poletown didn't want to not have their community ravaged for a pittance, nor did Donald Trump's neighbors want their homes converted to his parking lot. But it is wrong to look at this one feature of the eminent domain equation in isolation. Barnett's objections to eminent domain depend solely on the unknowable subjective value each person attaches to his or her possessions. But if that point is decisive, then eminent domain is also illegitimate when used to condemn land for national fortifications or public parks, when the public-use criterion cannot be contested at all.

Thus there are several real difficulties with Barnett's examples. First, the private use in both cases expands eminent domain power where it is least needed. Second, the puny compensation offered systematically ignored the subjective value of the private landowners and the dislocations the taking brought into their lives. Narrow the permissible set of purposes for takings and boost the compensation, and this takings problem will shrink, without using the meat cleaver of incommensurate subjective values to savage eminent domain.

Next Barnett claims we don't need eminent domain because private ingenuity can overcome holdout problems. But only sometimes. Private developers can use quiet tactics to assemble large parcels, but only because the law turns a blind eye to the low-level frauds that are used to mislead sellers about the buyer's intentions. But governments acting through public debates and appropriations can't use the same tactics to assemble land for highways. The history of oil and gas production, with the need to assemble and organize complex network industries, shows how holdout problems can stop development in its tracks. Indeed, the entire structure of property law rests on an implicit set of forced exchanges, which give first possessors, inventors, and writers priority over the rest of the world when astronomical transaction costs block voluntary negotiations. Too often private ingenuity blocks social coordination. It is much more sensible to stick with the more modest proposition that as transaction costs go down, the need for state intervention is reduced.

David Friedman's remarks are vulnerable to both these and other objections. The societies to which he refers but does not name are small communities, such as medieval Iceland, where tiny numbers and kinship relationships eased the path to social organization. Modern societies have bowed to the inevitability of some taxation, even if they have not done enough to constrain its use.

Again, it is not enough for Friedman to list the instances where state power fails. My own view gives no scope for tariffs because local protection is their raison d'être, notwithstanding all the palaver about infant industries. By contrast, the state sponsorship of scientific research has created public goods; before I ruled the National Science Foundation out of bounds, I would examine the evidence that advances in health have more than repaid their cost. Improve the system, yes; abolish it, no.

Friedman's pipe dream is that the alternative to limited government is no government at all. Not so. A large society with no central authority offers an open invitation to some sleazy individual to consolidate power in his own name. Constitutional government uses deliberation to expand the base of public support. Friedman's void at the center promotes totalitarian rule, not individual liberty.

Pinkerton charges me with an excessive preoccupation with economics. In part, I plead guilty to the charge that a short essay did not cover the waterfront and offer this belated response. First, on matters of value, nothing says that only markets matter. Indeed, as Barnett's comment and my reply to him indicate, one powerful reason for strong property rights is to allow people to pursue their subjective understanding of the good while respecting the like liberties of others. Nothing in my approach privileges certain kinds of preferences over others.

Let me turn next to Pinkerton's lists of specific issues. Pollution, how it should be attacked and when (at low levels) it should be allowed, lies at the heart of any law and economics program. The drug question also is amenable to that approach. Does drug use impose a peril on nonusers that requires some intervention before the fact? If so, what? Here's one three part program that might work: 1) abolish drug offenses, 2) remove all state subsidies for rehabilitation, and 3) refuse to reduce criminal responsibility by reason of voluntary drug use. As to terror, we all face the sticky tradeoff between liberty and security, where the only sensible program (even in Barnett's universe) asks whether additional precautions are more intrusive than beneficial. This is the heart of law and economics. Uncertainty makes utilitarians of us all.

Bush and Iraq are topics on which lawyers don't have much to say. I can predict what set of rules will help generate responsible political leaders, but I cannot make hard policy choices as a matter of general political theory. Just when should one nation intervene in the affairs of another, be it for humanitarian reasons or by way of anticipatory self-defense? Again, it is a matter of hard tradeoffs. But this is hardly an indictment of my approach. It only shows the great need for statecraft.

Nor does a focus on legal institutions trivialize issues such as character formation, family, or personal life. Here Pinkerton's criticism of me echoes that which John Stuart Mill made of Jeremy Bentham. Bentham was weak on personal virtues. There's a reason. Lawyers specialize in the question of how and when to use force. They deal with individuals as strangers, trading partners, family members, and litigants. Professionally, they don't have any inside track on matters of character, duty, and virtue, however vital.

Any libertarian who thinks you can promote virtue solely by abstaining from force and fraud is smoking some banned substance. Character helps people choose the proper course of conduct among those that are legal. Legal and political theory do themselves a disservice when they poach on questions of personal behavior. They are at their best when they deal with matters of constitutional structure and political power. It's nice, just this once, to occupy the odd position of being a moderate.