How, without government, could we settle the disputes that are now settled in courts of law? How could we protect ourselves from criminals?

Consider first the easiest case, the resolution of disputes involving contracts between well-established firms. A large fraction of such disputes are now settled not by government courts but by private arbitration of the sort described in Chapter 18. The firms, when they draw up a contract, specify a procedure for arbitrating any dispute that may arise. Thus they avoid the expense and delay of the courts.

The arbitrator has no police force. His function is to render decisions, not to enforce them. Currently, arbitrated decisions are usually enforceable in the government courts, but that is a recent development; historically, enforcement came from a firm’s desire to maintain its reputation. After refusing to accept an arbitrator’s judgment, it is hard to persuade anyone else to sign a contract that specifies arbitration; no one wants to play a game of ‘heads you win, tails I lose’. For repeat players, the agreement is enforced by what is sometimes described as the discipline of constant dealings.

Arbitration arrangements are already widespread. As the courts continue to deteriorate, arbitration will continue to grow. But it only provides for the resolution of disputes over pre-existing contracts. Arbitration, by itself, provides no solution for the man whose car is dented by a careless driver, still less for the victim of theft; in both cases the plaintiff and defendant, having different interests and no prior agreement, are unlikely to find a mutually satisfactory arbitrator. Indeed, the defendant has no reason to accept any arbitration at all; he can only lose—which brings us to the problem of preventing coercion.

Protection from coercion is an economic good. It is presently sold in a variety of forms—Brinks guards, locks, burglar alarms. As the effectiveness of government police declines, these market substitutes for the police, like market substitutes for the courts, become more popular.

Suppose, then, that at some future time there are no government police but instead private rights enforcement agencies. These agencies sell the service of protecting their clients against crime. Perhaps they also guarantee performance by insuring their clients against losses resulting from criminal acts.

How might such agencies protect? That would be an economic decision, depending on the costs and effectiveness of different alternatives. They might limit themselves to passive defenses, installing elaborate locks and alarms, or they might take no preventive action at all but make great efforts to hunt down criminals guilty of crimes against their clients. They might maintain foot patrols or squad cars, like our present government police, or they might rely on electronic substitutes. In any case, they would be selling a service to their customers and would have a strong incentive to provide as high a quality of service as possible at the lowest possible cost. It is reasonable to suppose that the quality of service would be higher and the cost lower than with the present governmental protective system.

Inevitably, conflicts would arise between one agency and another. How might they be resolved?

I come home one night and find my television set missing. I immediately call my agency, Tannahelp Inc., to report the theft. They send an agent. He checks the automatic camera which Tannahelp, as part of their service, installed in my living room and discovers a picture of one Joe Bock lugging the television set out the door. The Tannahelp agent contacts Joe, informs him that Tannahelp has reason to believe he is in possession of my television set, and suggests he return it, along with an extra ten dollars to pay for Tannahelp’s time and trouble in locating Joe. Joe replies that he has never seen my television set in his life and tells the Tannahelp agent to go to hell.

The agent points out that until Tannahelp is convinced there has been a mistake, he must proceed on the assumption that the television set is my property. Six Tannahelp employees, all large and energetic, will be at Joe’s door next morning to collect the set. Joe, in response, informs the agent that he also has a rights enforcement agency, Dawn Defense, and that his contract with them undoubtedly requires them to protect him if six goons try to break into his house and steal his television set.

The stage seems set for a nice little war between Tannahelp and Dawn Defense. It is precisely such a possibility that has led some libertarians who are not anarchists, most notably Ayn Rand, to reject the possibility of competing free-market rights enforcement agencies.

But wars are very expensive and Tannahelp and Dawn Defense are both profit-making corporations, more interested in saving money than face. I think the rest of the story would be less violent than Miss Rand supposed.

The Tannahelp agent calls up his opposite number at Dawn Defense. ‘We’ve got a problem…’ After explaining the situation, he points out that if Tannahelp sends six men and Dawn eight, there will be a fight. Someone might even get hurt. Whoever wins, by the time the conflict is over it will be expensive for both sides. They might have to start paying their employees higher wages to make up for the risk. Then both firms will be forced to raise their rates. If they do, Murbard Ltd., an aggressive new firm which has been trying to get established in the area, will undercut their prices and steal their customers. There must be a better solution.

The man from Tannahelp suggests that the better solution is arbitration. They will take the dispute over my television set to a reputable local arbitration firm. If the arbitrator decides that Joe is innocent, Tannahelp agrees to pay Joe and Dawn Defense an indemnity to make up for their time and trouble. If he is found guilty, Dawn Defense will accept the verdict; since the television set is not Joe’s, they have no obligation to protect him when the men from Tannahelp come to seize it.

What I have described is a very makeshift arrangement. In practice, once anarcho-capitalist institutions were well established, agencies would anticipate such difficulties and arrange contracts in advance, before specific conflicts occurred, specifying the arbitrator who would settle them.

In such an anarchist society, who would make the laws? On what basis would the private arbitrator decide what acts were criminal and what their punishments should be? The answer is that systems of law would be produced for profit on the open market, just as books and bras are produced today. There could be competition among different brands of law just as there is competition among different brands of cars.

In such a society there might be many courts and even many legal systems. Each pair of rights enforcement agencies agree in advance on which court they will use in case of conflict. Thus the laws under which a particular case is decided are determined implicitly by advance agreement between the agencies whose customers are involved. In principle, there could be a different court and a different set of laws for every pair of agencies. In practice, many agencies would probably find it convenient to patronize the same courts, and many courts might find it convenient to adopt identical, or nearly identical, systems of law in order to simplify matters for their customers.

Before labeling a society in which different people are under different laws chaotic and unjust, remember that in our society the law under which you are judged depends on the country, state, and even city in which you happen to be. Under the arrangements I am describing, it depends instead on your agency and the agency of the person you accuse of a crime or who accuses you of a crime.

In such a society law is produced on the market. A court supports itself by charging for the service of arbitrating disputes. Its success depends on its reputation for honesty, reliability, and promptness and on the desirability to potential customers of the particular set of laws it judges by. The immediate customers are rights enforcement agencies. But each agency is itself selling a product to its customers. Part of that product is the legal system or systems of the courts it patronizes and under which its customers will consequently be judged. Each agency will try to patronize those courts under whose legal system its customers would like to live.

Consider, as a particular example, the issue of capital punishment. Some people might feel that the risk to themselves of being convicted, correctly or incorrectly, and executed for a capital crime outweighed any possible advantages of capital punishment. They would prefer, where possible, to patronize agencies that patronized courts that did not give capital punishment. Other citizens might feel that they would be safer from potential murderers if it was known that anyone who murdered them would end up in the electric chair. They might consider that safety more important than the risk of ending up in the electric chair themselves or of being responsible for the death of someone falsely accused of murder. They would, if possible, patronize agencies that patronized courts that did give capital punishment.

If one position or the other is almost universal, it may pay all agencies to use courts of the one sort or the other. If some people feel one way and some the other and their feelings are strong enough to affect their choice of agencies, it pays some to adopt a policy of guaranteeing, whenever possible, to use courts that do not recognize capital punishment. They can then attract anti-capital-punishment customers. Other agencies do the opposite.

Disputes between two anti-capital-punishment agencies will, of course, go to an anti-capital-punishment court; disputes between two pro-capital-punishment agencies will go to a pro-capital-punishment court. What would happen in a dispute between an anti-capital-punishment agency and a pro-capital-punishment agency? Obviously there is no way that if I kill you the case goes to one court, but if you are killed by me it goes to another. We cannot each get exactly the law we want.

We can, however, each have our preferences reflected in the bargaining positions of our respective agencies. If the opponents of capital punishment feel more strongly than the proponents, the agencies will agree to no capital punishment; in exchange, the agencies that want capital punishment will get something else. Perhaps it will be agreed that they will not pay court costs or that some other disputed question will go their way.

One can imagine an idealized bargaining process, for this or any other dispute, as follows: Two agencies are negotiating whether to recognize a pro- or anti-capital-punishment court. The pro agency calculates that getting a pro-capital-punishment court will be worth $200,000 a year to its customers; that is the additional amount it can get for its services if they include a guarantee of capital punishment in case of disputes with the other agency. The anti-capital-punishment agency calculates a corresponding figure of $400,000. It offers the pro agency $300,000 a year in exchange for accepting an anti-capital-punishment court. The pro agency accepts. Now the anti-capital-punishment agency can raise its rates enough to bring in an extra $350,000. Its customers are happy, since the guarantee of no capital punishment is worth more than that. The agency is happy; it is getting an extra $50,000 a year profit. The pro agency cuts its rates by an amount that costs it $250,000 a year. This lets it keep its customers and even get more, since the savings is more than enough to make up to them for not getting the court of their choice. It, too, is making a $50,000 a year profit on the transaction. As in any good trade, everyone gains.

If you find this confusing, it may be worth the trouble of going over it again; the basic principle of such negotiation will become important later when I discuss what sort of law an anarcho-capitalist society is likely to have.

If, by some chance, the customers of the two agencies feel equally strongly, perhaps two courts will be chosen, one of each kind, and cases allocated randomly between them. In any case, the customer’s legal preference, his opinion as to what sort of law he wishes to live under, will have been a major factor in determining the kind of law he does live under. It cannot completely determine it, since accused and accuser must have the same law.

In the case of capital punishment, the two positions are directly opposed. Another possibility is that certain customers may want specialized law suited to their special circumstances. People living in desert areas might want a system of law that very clearly defines property rights in water. People in other areas would find such detailed treatment of this problem superfluous at best. At worst, it might be the source of annoying nuisance suits. Thus the desert people might all patronize an agency with a policy of always going to a court with well-developed water law. Other agencies would agree to use that court in disputes with that agency but other courts among themselves.

Most differences among courts would probably be more subtle. People would find that the decisions of one court were prompter or easier to predict than those of another or that the customers of one agency were better protected than those of another. The agencies, trying to build their own reputations, would search for the best courts.

Several objections may be raised to such free-market courts. The first is that they would sell justice by deciding in favor of the highest bidder. That would be suicidal; unless they maintained a reputation for honesty, they would have no customers—unlike our present judges. Another objection is that it is the business of courts and legislatures to discover laws, not create them; there cannot be two competing laws of gravity, so why should there be two competing laws of property? But there can be two competing theories about the law of gravity or the proper definition of property rights. Discovery is as much a productive activity as creation. If it is obvious what the correct law is, what rules of human interaction follow from the nature of man, then all courts will agree, just as all architects agree about the laws of physics. If it is not obvious, the market will generate research intended to discover correct laws.

Another objection is that a society of many different legal systems would be confusing. If this is found to be a serious problem, courts will have an economic incentive to adopt uniform law just as paper companies have an incentive to produce standardized sizes of paper. New law will be introduced only when the innovator believes that its advantages outweigh the advantages of uniformity.

The most serious objection to free-market law is that plaintiff and defendant may not be able to agree on a common court. Obviously, a murderer would prefer a lenient judge. If the court were actually chosen by the disputants after the crime occurred, this might be an insuperable difficulty. Under the arrangements I have described, the court is chosen in advance by the agencies. There would hardly be enough murderers at any one time to support their own rights enforcement agency, one with a policy of patronizing courts that did not regard murder as a crime, and if there were, no other agency would accept such courts. The murderers’ agency would either accept a reasonable court or fight a hopeless war against the rest of society.

Until he is actually accused of a crime, everyone wants laws that protect him from crime and let him interact peacefully and productively with others. Even criminals. Not many murderers would wish to live under laws that permitted them to kill—and be killed.