Professor Shenfield's essay on anarcho-capitalism, appearing as it does shortly after the publication of Robert Nozick's excellent Anarchy, State, and Utopia, gives me hope that the debate between libertarians who believe in little government and those who believe in no government at all is reviving, and on a higher and more cordial level than before. Professor Shenfield, like Professor Nozick, has taken the trouble to read the anarchists before writing about them, and his criticisms deserve serious attention. He raises six objections to the anarchist position, and I will deal with them in order.
Professor Shenfield asserts that an anarchist society could not defend itself against aggressive Communist powers. It is true that the defense of such a society, if provided at all, must be provided by a private organization. Lacking the taxing power of the State, such an organization would presumably support itself largely by voluntary donations, such as charitable organizations do today. Defense would surely seem, to most inhabitants of an anarchist community, at least as important as medical research or Boy Scout troops.
In addition, although it would be unable to charge most of those protected (due to the difficulty of protecting from nuclear attack only those particular people who had paid their protection bill, but not their neighbors who failed to pay), a private firm providing "national defense" could still sell part of its product. Isolated areas could be protected or not according to how much local businessmen could raise to pay for the protection. Residents of the anarchic area traveling or doing business abroad could be given the same protection a State provides its citizens, provided they paid an appropriate fee. In these and other ways, a defense firm could collect, not the full value of the services it provided, but some fraction thereof.
Whether such a firm could raise enough money to defend the United States against its present enemies I do not know. Nor do I know whether our present government will continue to defend the country successfully; while it spends a great deal of money toward that end, it does not seem to spend it very efficiently. In any case, neither anarchy nor limited government seems imminent at the moment. The resources required for defense vary widely from country to country and from time to time; even if it turns out that an anarchist society cannot defend itself in North America in 1976, that says very little about its prospects at future times or in other places.
Professor Shenfield argues that the private firms which protect the inhabitants of an anarchist group against crime would become a single monopoly, through merger or conquest, and that that monopoly protection agency would thus be a State. Protection agencies, like other firms, will find that they have some optimum size, above which additional growth makes them less efficient, not more. If this optimum size is so large that there are only a few, it is likely enough that, finding theft more profitable than protection, they will combine into a government, as Professor Shenfield fears. If, as I think likely, the efficient agency is about the size of the police department of a small city, then there will be thousands, and any one that tries to fight or buy its way to a monopoly will consume its resources long before it acquires significant power. The fact that a group of organizations could, by working together, take over a society, does not mean that they will. If it did, we would have been taken over long ago by an alliance of the regular military, the National Guard, and the Federal, state, and local police.
Professor Shenfield argues that a private system cannot deal with many problems, such as air pollution, caused by situations where one person's actions indirectly harm many other people. He apparently assumes that the only market solution to such problems is redefinition of property rights so that a resource, such as a river, that was previously unowned and hence inefficiently used acquires an owner or owners. This is, in many cases, the best solution, but it is not the only one. Market courts could also, like government courts, grant injunctions against those who injure others or consider class action to force polluters to pay for the damage they cause.
Many existing property titles are, as Professor Shenfield says, unjust, or at least highly debatable and unclear. The existing State can and does provide the well-defined set of titles required for the functioning of a market by simply declaring the status quo legitimate. Without a State, who would decide, and how?
In answering this question, I must part company with many libertarians, anarchist and nonanarchist alike. What Professor Shenfield is criticising is not anarchy, but the tendency of libertarians to assume that a libertarian society will be run by moral philosophers. Precisely the same problem exists for the libertarian limited State; it too must resolve questions of ownership which are, in terms of pure libertarian principle, difficult or impossible to resolve.
I, unlike many anarchists, assume that in an anarcho-capitalist society people will pay no more attention to philosophers than they do now. Such a society will be libertarian, not because everyone believes in libertarianism (although a greater popularity of libertarian ideas might make it easier to achieve such a society), but because a "market" legal system will, for economic reasons, produce an economically efficient body of law. It so happens that freedom is efficient—the value to me of being able to order you around is rarely as great as the cost to you of being ordered around. Going from coercive to libertarian law is on net profitable, so a system which produces law for profit will rarely produce coercive laws. (This is a very brief explanation of a complicated and important argument. For a longer and more detailed explanation see my The Machinery of Freedom [New York: Harper & Row, 1973], part 3, especially chapters 29 and 31.)
It is not profitable to carry libertarianism to the extent of deciding in court how British land titles should be revised to undo the effects of the Norman conquest. It is therefore likely that anarcho-capitalist courts will accept the validity of existing titles to property except where there is an alternative claimant with a clear and unambiguous case.
Professor Shenfield complains that anarchist court systems would have no power of subpoena and hence would be incapable of convicting the guilty. But surely the prosecution or the defense or the court itself could and would pay witnesses to appear—they do, after all, provide a valuable service. If this proves too cumbersome, protective agencies might ask their clients to agree in advance to provide any evidence they possess relevant to cases that come to trial. Where the evidence is known to be in the hands of the defendant, and he has made no such agreement, the court might be unable to force him to disclose it, but it could certainly draw inferences from his failure to do so.
Professor Shenfield's sixth and last point I do not understand. He apparently believes that it requires more intelligence and vigilance to maintain peaceful and beneficent anarchy than to keep a limited state limited. Why? The limited state possesses rights and powers denied to its citizens and controls vast resources which it uses for purposes that cannot, by their nature, be easily limited—witness the activities that can be and are presently defended as necessary for national security. How can any citizen, however wise, be sure when his State has crossed the invisible line separating the just commonwealth from the embryo leviathan? And if he is sure, what can he do about it?
Under anarcho-capitalism, employees of a protection agency who coerce others will be criminals and will be treated as such, if not by their agency then by their victims. And if an agency coerces its own clients, or wastes its resources trying to conquer its competitors, or even does an inefficient job, its clients have available the instant and individual sanction of the marketplace. They can—and will—take their custom elsewhere.
David Friedman teaches at the Fels Center of Government at the University of Pennsylvania. He is the author of The Machinery of Freedom (Harper Colophon, 1973) and of numerous articles on political and economic issues.