Libertarian natural law theorists have traditionally focused most of their effort on solving the property rights of any given situation. Unfortunately, such focus has come at the detriment to a coherent theory of appropriate punishment for those who violate the rights of others. Libertarians generally have much more to say on the rights (or lack thereof) of children and the comatose, or what homesteading parts of the electromagnetic spectrum entails, than what a person may do to someone who has burned down their house. Murray Rothbard attempted to correct this oversight in The Ethics of Liberty [3]. I hope to show here that not only is Rothbard's analysis grossly incorrect, but that no rational punishment theory can ever be proven a priori. My analysis will end with an investigation of how a perfectly libertarian justice system could work without even trying to answer this unsolvable problem.
All ensuing analysis will have as an assumption that the libertarian ethic of non-aggression is true. Rights are herein used to define who may ethically direct the use of an object. For example, the non-aggression principle implies that all humans have the right to their own body and the property that they either make first use of in nature or acquire by voluntary means from someone else.
The ethical exercise of self-defense naturally results from the non-aggression principle. When someone else is attempting to control an object that rightfully belongs to you, self-defense merely entails directing that object's use yourself. For example, if you get home from work to see a thug walking out of your living room with your television, you are doing no more than directing the television's use by removing it from the thug's hands and placing it back. You are doing no more than directing the use of your land by subsequently depositing the thug off of your land perimeter. During an act of aggression, one may justly defend oneself and one's property. This does not necessarily resolve the question of how much force is appropriate for self-defense (the classic example of killing a boy trying to snatch a piece of candy comes to mind), but we will put this question off for some other time. I wish instead to look into what actions are appropriate after all rights violations have occurred and all actors are back within a libertarian free market environment.
Rothbard suggests that a rational punishment theory would be that the criminal loses a proportion of his rights to the victim, and furthermore that he loses them to the extent that he deprives the victim. For example, Rothbard states that if A stole $15,000 from B, that A must pay back $30,000, plus compensation for psychic costs borne by B. Such a system, which we may label the principle of 'two wrongs make a right,' is intuitively appealing. Rothbard has included three of the basic tenets most people consider are needed in any punishment ideal: restitution of the original $15,000, punishment by forcing another $15,000, and in the process deterrence of further crimes.
I wish to point out two problems with Rothbard's theory, either of which would be sufficient to disprove not only it but any a priori punishment theory. The first problem concerns the idea of appropriate compensation.
If A has stolen an object of B, then it is clear that the object's rightful owner is B, and B may reclaim it. But what if A completely destroys the object? Or what if A does not steal something of B's but instead assaults him or trespasses on his property? What then is the exact compensation, according to Rothbard's proportionality rule, that must be meted out?
Rothbard recognizes the problems of the above questions but quickly dismisses them: 'What the extra compensation should be it is impossible to say exactly, but that does not absolve any rational system of punishment . . . from the problem of working it out as best one can.' (1) Rothbard then goes on to claim that a crime of assault should be repaid by a beating of the criminal, either by the victim or by an agent thereof.
The problem that Rothbard faces is the same problem the welfare economist faces in determining whether a government intervention increases net utility: There is no means by which to measure utility. The welfare economist hopes to show that the utility gained by the beneficiary of an intervention 'outweighs' the utility lost by the target of said intervention. Likewise, Rothbard's punishment system hopes to show that utility lost by the criminal in the punishment is exactly twice that lost by the victim as a result of the crime. But as Rothbard proved in an earlier work, it is meaningless to make utility comparisons between two separate people. (2) That is, there exists no utilitometer that we can connect to one's brain that enables us to read off the happiness gained by an action. Happiness is entirely subjective: The only thing an economist can say is that an actor expected to maximize it by acting.
In fact, it is just as meaningless for an economist to claim that net social utility is increased by a tax on cigarettes as it is for Rothbard to claim that any punishment he approves of is proportional to the damage done by the crime. A simple example will clearly illustrate a few problems. What if the criminal objects only slightly to being beaten, but the victim was severely traumatized? How is the judge to determine how much utility was lost by the victim, or even when that same utility has been 'beaten out of' the criminal? The answer is that it is an impossible task. We must therefore throw out Rothbard's theory as not just impossible to implement, but meaningless in formulation. This applies also to any theory that relies on restitution or proportional punishment. Unless the restitution consists solely of returning stolen objects, it must be based on a meaningless interpersonal comparison of utility.
The second objection to Rothbard's theory of punishment is that it violates the rights of the criminal. Rothbard quietly inserts his assumption that the criminal loses some degree of his rights without ever providing a justification for this principle. Perhaps Rothbard has not been taken to task on this issue because it is almost universally accepted among libertarians that a criminal gives up some of his rights by committing a crime. But why must we automatically assume that one must be coercively punished for a crime? Is this view not more consistent with the idea of us as dogs in need of negative reinforcement by the nanny-state than as the rational adults we are?
There is no consistent principle that states that if A steals B's watch, then B may subsequently steal A's shirt. To claim otherwise, as one must do to support Rothbard's punishment theory, one must accept rights as non-absolute. But rights indicate who may justly control the use of an object. If someone has the exclusive right to some given property, those rights exist until voluntarily transferred to others or abandoned. No matter what the actions one takes, no one else has the right to usurp control over those objects. Rothbard's punishment theory therefore requires that the rights of all criminals be infringed upon. Infringing upon rights is not appropriate action for any government; nor is it appropriate action for anyone trying to 'teach a lesson' to someone else. At base, punishment theory has no grounding in proper ethics but can only be justified by claiming that the cruel delight of revenge is more important than property rights.
The above analysis may seem to indicate that we must ethically remain at the mercy of criminals. Though we can defend our property during an act of aggression, there is no ethically correct coercive punishment after the fact. What then can we do?
I submit that in a libertarian society, there would in fact be plenty of punishment verdicts for criminals. Courts would function and would decide upon liability and the amount to be repaid to the victim. Furthermore, these courts would be acting within the ethical framework of libertarian non-aggression. The reason is that they would be employing non-coercive means of punishment.
Non-coercive means of punishment could include public ostracism, blackmail, boycott, or even shunning. As such, a court could arrive at any verdict, whether regarding guilt or innocence, or appropriate punishment, and it would be ethically proper as long as it did not violently impose its decision. The punishment would require the voluntary agreement and action of those within the community. If a court made what the people thought were outrageous decisions, they would simply not go along with the non-coercive enforcement measures.
One may scoff at the abilities of a libertarian court system to actually deter crime and punish criminals. But keep in mind that in a libertarian society, there would be few places for a shunned criminal to turn to. Roads would likely be privately owned, and boycott groups would pressure businesses from dealing with such men. If the people of a community were really determined to punish a criminal, they could essentially relegate him to eking out an existence of bare sustenance on his land (or perhaps starvation). In fact, we have at our disposal historical examples of non-coercive court systems.
The common law that developed in the Middle Ages had some elements of non-coercive enforcement, though decisions were not always enforced in a libertarian manner. However, the unique law that was developed by merchants did have this character. Bruce Benson explains that merchant courts developed because the other options of the time, royal courts or common law courts, would not honor contracts regarding the payment of interest or would not allow books of account as evidence. (3) Benson also explains how the decisions of these courts were not backed by violence but rather from the voluntary boycotting of other merchants. Quoting William Wooldridge, 'Merchants made their courts work simply by agreeing to abide by the results. The merchant who broke the understanding would not be sent to jail, to be sure, but neither would he long be a merchant, for the compliance exacted by his fellows, their power over his goods, proved if anything more effective than physical coercion.' (4)
The merchant law settled contract disputes and resolved issues such as fraud. Though the scope of the law was limited, there is no reason to expect that such a system could not be applied to all law. In fact, we can cite a modern system of law with voluntary enforcement that regulates all manner of behavior. This is the Amish code of conduct.
Some libertarians might scoff at describing the Amish as a libertarian community since their conformity of behavior seems to indicate coercive enforcement of religious values on other people. However, I submit that their system is perfectly libertarian. The key, again, is not the exact values and legality they decide to enforce, but rather how they choose to enforce these decisions. The Amish follow an unwritten code of conduct, known as the Ordnung, which outlines the expected behavior of those who have accepted the Amish faith. (5) The Amish are well-known to be pacifists, and they therefore can enforce the Ordnung only by non-coercive means. Community elders decide what should be done for members who do not follow the Ordnung. A common decision is to demand a public apology. If this is not performed, the Amish may decide to shun a member, meaning that he is totally ignored by the other members of the community. Such a decision would be tough for anyone to handle, and as such, few people ever get out of line enough to require its use.
The Amish decision to shun is not made for punishment but rather as an act of love to try to win the wayward member back into the ways of the community. An Amish farmer expresses a general Amish view on justice: 'They ought to burn down all those expensive prisons because you can't force people to be good. Youngsters learn (to) be good in their families and churches but these are not respected today.' (6) Though the Amish form their 'court' decisions based on a religious moral system and value community over the individual, still they have a system of private property and do not violate anyone's rights, either through coerced taxation or unlibertarian punishment of crime. It is unfortunate that libertarians have tended to overlook Amish society as a working example of libertarianism. Even though they might not advertise themselves as anarchists, the fact that they would never violently enforce any laws reveals them to be de facto anarchists. The Amish always choose non-coercive means.
The structure of a justice system in a stateless society is an intriguing question to libertarians and anarchists. Unfortunately, many suggestions for private courts and enforcement agencies set up nothing more than little competing states. Such 'private' courts would be more like states in character because they must violently enforce their decisions. Furthermore, since it is unethical to coercively punish criminals, the stage is set for these illegitimate entities to keep enlarging their violations of rights. It is only a matter of degree between putting a thief on a chain gang and extorting payment from the people of a given region for the court's dubious service of justice. The only alternative is to recognize that no one has the right to deprive criminals of their rights, and that any punishment must be accomplished by voluntary means. Aside from being the libertarian solution for justice, such a system would promote community and prevent any court from quietly becoming a government and landing us back in the situation we have today.
1. The Ethics of Liberty , p. 89. Rothbard is here discussing the compensation of psychic damages, but the first part of his comment applies to any compensation plan, as we shall soon see below.
2. See 'Toward a Reconstruction of Utility and Welfare Economics,' 1956.
3. The Enterprise of Law, p.33.
4. Uncle Sam, the Monopoly Man, p. 95-96.
5. The Amish, together with the Mennonites, are unique among Christians in their belief that baptism is to be performed on adults, who, unlike the newborn, may decide for themselves whether to accept the church.
6. Foster, Thomas W. 'American Culture Through Amish Eyes: A Christian Anarchist Perspective.' Department of Sociology, Ohio State University , 1980.
