Anarchy, State, and Utopia
by Robert Nozick

First he argues that the minimal state would evolve from an anarchist society without violating anyone's rights. I completely missed his point. I thought he was still building his argument, when all of a sudden, he was claiming to have proven his thesis. After this razzle-dazzle, the rest of the book is great. He refutes John Rawls brilliantly. He also refutes several arguments for redistribution of wealth and for more than a minimum state. Finally, he shows that only a libertarian society allows utopian experimentation and is inspiring to all utopians. During his libertarian phase, Robert Nozick offered a theory to explain how a minimal state (limited to protection, compensation, and retribution services) could arise in a morally justifiable way. His argument does not lend support to any states that have actually existed. No states that we know of have arisen by means that are justified by Nozick's theory, and all historical states exercise more powers than can be justified by his theory of justice.

Nozick's theory was an attempt to provide a moral argument for the establishment of an organization that monopolizes the retribution business within a geographic area by using the political means of coercion to prevent competitors from assessing criminal guilt in ways it doesn't approve. This hypothetical organization fits the definition of a state except that it does not use the political means of taxation to finance its operations.

Nozick starts with the state of nature in which everyone has the right to defend himself against crime and to form voluntary associations to enforce justice. In this society, all the protection agencies adhere to Nozick's principles of justice. These principles are similar to mine except that (1) forcible appropriation of compensation from criminals and punishment of criminals are not considered to be crimes, and (2) it is OK to prohibit people from doing risky activities if you compensate them for this violation of their rights.

Nozick predicted that if the free market were allowed to operate in the protection industry, one of the competing protection agencies would inevitably become dominant through an invisible-hand process (a series of rational actions by people who do not have this goal in mind—the result of human action but not of human design). This dominant protection agency would evolve into an "ultraminimal" state without violating anyone's rights.

Next, Nozick used his principle of compensation to justify the transformation of the ultraminimal state into the minimal state. This handy principle allows you to morally violate someone's rights. In other words, it allows you to violate someone's rights without violating their rights. The legitimacy of the minimal state rests on this principle.

Nozick offered no convincing moral arguments for his principle of compensation. It is less likely that people would agree on this principle than that they would agree on procedures to protect their rights. Yet Nozick made the opposite claim and used the compensation principle to justify the suppression by the dominant protection agency of the risky procedures of competing protection agencies.

It is purely a conjecture that in the free market there would have to be a dominant protection agency. Let's assume, for the sake of argument, Nozick's prediction is correct that one agency would become dominant in the rights-protection market. Let's also assume, for the moment, that everyone agrees it is OK to prevent people such as the Intrepid Protection Agency from doing something that risks violating the rights of a suspect or a witness or anyone else, as long as you compensate the Intrepid Protection Agency for doing so. Does it follow that the dominant protection agency could and would use the compensation principle to suppress the procedures used by all of its weaker competitors and thereby become a minimal state?

Given the avaricious nature of successful businessmen, it is likely that the dominant agency would want to establish a monopoly. If the compensation principle could be exploited for this purpose, it is reasonable to suppose that it would be. Would the dominant agency be able to apply the compensation principle objectively to establish a monopoly?

Unfortunately for Nozick's theory, there is no objective way to measure compensation or to determine what compensation is commensurate with any particular violation of rights. Different people value their rights differently and value the same compensation differently. Some people would fight and risk death to defend their rights. Other people are less idealistic or more cowardly and would surrender their rights easily. No compensation could buy off people in the first group. They would always be a destabilizing force and a threat to the minimal state. The second group could be brought into compliance by mere threats, without any need for compensation. Between these two extremes are an infinite number of possible positions. Any particular compensation would be exactly right for only a small percentage of the people. For some it would be too little compensation and for others it would be more than enough. We have no objective way to determine the correct compensation.

If some people continue to fight for their rights despite the fact that they have been given something in compensation for violating their rights, it is safe to assume that the compensation was not enough. If some people surrender their rights under duress after receiving something in compensation, we cannot be sure that they really prefer this situation. It may not be satisfaction with the compensation that stills their resistance. They may have stopped defending their rights because they are out-gunned. The prospects for successful resistance may discourage them from openly rebelling against what they regard as injustice. This is quite different from being satisfied by co-optive compensations.

The right that Nozick believes the dominant protection agency should usurp is the right to participate with another protection agency that uses procedures for determining guilt that are unreliable or risky in the opinion of the dominant protection agency. If such procedures actually violate the rights of innocent people, it is legitimate for the dominant agency (or anybody else) to stop the violations. However, sometimes risky procedures do not violate the rights of innocent people. In these cases, it is not legitimate to stop these procedures. Suppose the alleged criminal and the alleged victim both agree to obey the decision of a particular court that, in the opinion of the dominant agency, uses risky procedures. From whom does the dominant agency get the authority to overrule that court?

It is surprising that Nozick singled out the issue of procedural rights to be the centerpiece in his argument. The determination of the legitimacy of fact-finding procedures is not as difficult or controversial as the determination of appropriate compensation for crimes. More important, the potential for rights violation by risky fact-finding procedures is not as great as the potential for rights violation by unproven principles of compensation and punishment.

Nozick's argument that one protection agency should monopolize retributive justice would have been stronger if he had concentrated on the widespread disagreements about what punishments and compensations are appropriate instead of concentrating on the less controversial subject of procedural rights. Suppose the Hanging Tree retribution company uses the same fact-finding procedures and recognizes the same procedural rights as the Fastidious Justice company, which is the dominant protection agency in the area. Suppose the Hanging Tree company administers the death penalty for all felons convicted by its non-risky guilt-establishing procedures. Suppose the Fastidious Justice company believes in administering less severe punishments. Wouldn't Fastidious Justice have the right to prevent Hanging Tree from administering the death penalty in cases where Fastidious Justice believes the crime calls for a lesser punishment? Doesn't Fastidious Justice have an obligation to do this in cases where the condemned person is one of their clients? Isn't this a stronger argument for Fastidious Justice to monopolize the justice industry than the argument that Nozick offered?

After exercising a lot of ingenuity and effort to present a moral justification of a minimal state, Nozick concluded his book by appealing to utilitarian values. He advocated a framework for society that allows people to voluntarily form, join, not join, dissolve, or resign from communities that make up their own living arrangements and club rules. He argued that this framework is the ideal utopian process, because it allows a multitude of communities to experiment with different life-styles, and it offers a free market in communities competing for the allegiance of individuals. The competition among the various communities within this free-market framework will lead to ever-improving social experiments as the individuals migrate from the less satisfying communities to the more satisfying ones.

The free-market framework does not impose a static order on society. Nor does it force people to continually experiment with different life-styles. If a social framework did the former, it would halt progress. If a social framework did the latter, it would lack the market test for determining the relative success or failure of social experiments.

The free-market framework allows individuals to keep living a life-style they are content with or to try an alternative life-style, if they think that would make them happier. It does not impose an ideal life-style on anyone, nor does it impose experimentation for its own sake. The free market treats individuals as moral agents who are responsible for their own lives.

Nozick called this free-market framework the utopian process, because it is the process necessary for making progress toward the ideal society in which everybody is as happy as they can be. He may have been correct, and I think he was, but some people don't like freedom and diversity, and they would not be happy in such a society. No social framework can be proven to be the best from a utilitarian point of view, because happiness cannot be measured. However, the free-market framework can be justified by natural law, because it is the only possible framework for society that respects the rights and autonomy of moral agents. The general framework that Nozick advocated in the utopian part of Anarchy, State, and Utopia is the same as the framework that I advocate.

Although Nozick was definitely on the right track, he had some misconceptions about rights that raise theoretical problems for which he had no answers. It may be possible to offer such generally attractive compensations that the majority of people would be content to lose their right to choose their means of protection. But the compensation principle cannot justify premeditated crime. The fact that most people can be bought off or intimidated is hardly a principle upon which to build a moral argument.

Year Read: 1976

Back to Libertarian Essays by Roy Halliday
Back to Nonfiction Book Notes
Back to Fiction Book Notes
Back to Book Notes by Author

This page was last updated on November 2, 2011.
This site is maintained by Roy Halliday. If you have any comments or suggestions, please send them to