Mises Wire

Home | Wire | Nozick and the Minimal State

Nozick and the Minimal State

  • phil


Robert Nozick’s derivation of a minimal state in the first part of Anarchy, State, and Utopia (1974) has generated a lot of criticism, and you might think there is nothing new to be said about it. I have a new point, though—at any rate, I haven’t seen it discussed—and this is what I’m going to talk about.

Nozick’s starting point resembles Murray Rothbard’s anarchism. The fundamental principle of his book is that individuals have rights. There are things you can’t do to individuals without violating these rights. Individuals are separate beings. Contrasting moral views, such as utilitarianism, don’t take seriously the differences between people. There is no collective entity that is better off if some are sacrificed to others.

These rights aren’t maximizing principles, but “side constraints.” They are like rules in a game, not moves within it. Side constraints imply that we aren’t trying to minimize the total number of rights violations by anyone. Each person isn’t even trying to minimize the total of his own rights violations. Rights just forbid you to do things. Possibly they lapse in catastrophic situations, but Nozick doesn’t commit himself about this.

The rights individuals have include rights over their own bodies. Property rights are individual. The world starts out unowned and people have to do something to resources to acquire them. What they have to do to acquire resources isn’t specified, but although Nozick is often taken to be a critic of Lockean appropriation of unowned property, I believe he supports a “first user” theory, again like Rothbard.

His project is to show how a minimal state might justifiably arise from this starting point. A minimal state, in his terms, has a de facto monopoly in a given territory over defense, protection, and justice, but it is not permissible for it to pursue other goals common in modern states, e.g., “social welfare.” People would arrive at the state through steps that are to their advantage and also morally all right.

One frequent mistake in understanding Nozick’s argument should be mentioned here. In Nozick’s invisible-hand derivation of the state, he is claiming more than that no step violates anyone’s rights. He is also claiming that each step is reasonable to take. Suppose I induce everyone on my block to surrender all their property to me, by high-pressure tactics that fall short of violating anyone’s rights. The fact that the result is, by hypothesis, one that doesn’t violate rights isn’t a very good justification for it, since the outcome involves, very likely, at least some people acting against their own interests. Nozick intends to claim more than that a state is morally permissible; this would be consistent with a state’s being always a “bad thing.”

Nozick’s procedure is to claim that, first, individuals can hire protective associations to defend their rights. When people hire protective associations, one association might turn out to be best at winning conflicts with other agencies. In that case, there will be a “cascading” process of people who join the winning agency. If this doesn’t happen, there will likely be a situation in which one agency is stronger in one area and others in another area.

If neither of these processes takes place, there will be a balance of agencies. These will come to agreement on what to do in cases of conflict. That is, they will settle on a procedure for appeals. This is less costly than fighting it out. As critics have often pointed out, there a problem here. Even if Nozick is right, why should we consider this arrangement a single dominant agency? Nozick seems to have defined a dominant agency into existence.

I’d like to concentrate on a different point, though, and that is the “cascading” argument, the claim that if an agency won most conflicts, there would be a cascading effect that would lead to its dominance. Nozick’s argument fails because it involves an arbitrary stopping point. In order for the dominant agency to transform itself into the ultraminimal and then minimal state, a substantial number of people must remain outside the state. Otherwise, we would simply have a situation in which everyone is a member of the same private agency. This would not give Nozick what he needs for his minimal state argument. What he needs is a scenario in which a substantial number of independents remain outside the dominant private agency (DPA). Then, by an elaborate argument, he tries to show that the DPA can prohibit these independents from applying risky decision procedures to its clients. In return for this prohibition, it must “compensate” the independents by offering them free or low-cost protection services. The independents don’t have to buy the protection, so they aren’t taxed, but it is greatly to their advantage to accept the DPA’s offer. Otherwise, they will be unable to protect themselves. Nozick thinks this comes close enough to taxation to qualify the dominant agency as a minimal state. Without this, he just has a dominant agency, and this doesn’t get him out of free market anarchism

If, however, Nozick’s argument for the formation of a dominant agency is correct, why would the cascading process stop short of near-universal membership in the dominant agency? Also, if the cascading argument is not correct, we don’t get to a dominant agency at all. In either case, Nozick’s effort to show that Rothbardian anarchism leads to the minimal state fails.


Contact David Gordon

David Gordon is Senior Fellow at the Mises Institute and editor of the Mises Review.

Do you want to write on this topic?
Check out our submission Guidelines
Note: The views expressed on Mises.org are not necessarily those of the Mises Institute.