The Violation of Rights

In a previous post, I have said that the progress of rights is constrained by two principles, that of homesteading and consensual exchange. The first says that property may not be acquired from nature without demonstrating ownership in some way. Originally this must require altering it some obvious way, but as a society develops a more sophisticated system of law, it may come to recognize more subtle methods of homesteading. The second requires that all exchange of rights be consensual. In libertarian theory, then, to violate rights is to violate these axioms: crimes must involve either defending something which is unowned, preventing a consensual exchange, or forcing a nonconsensual exchange to take place.

Crimes, of course, should never happen, but naturally they will happen anyway. The problem is to understand what is to be done about them. There are two broad kinds of retribution that one might make. One is self-defense, that is, using force or its threat to prevent the criminal from succeeding, and another is punishment, which is to use or threaten force against the criminal after his criminal activities have ceased so as to make him to atone for his crime. Libertarians have attempted to derive conclusions about the just retribution from the axioms of law, but this attempt is a dead end. We will see that the principles of retribution are not a part of the axiomatic theory of law, and instead are best justified instead as having been established by contract according to the principles of free market law I have described earlier.

The axiom of law is often called the nonaggression axiom, and stated as, “it is wrong to initiate aggression.” I prefer to call it the axiom of consent and state it as, “everything should be consensual.” Neither of these statements says anything about the proper response to aggression. They neither permit, nor prohibit, any form of self-defense or punishment. From the way I previously established the axiom, we can also see that the proper response to aggression is an open question: the axiom is the absolute minimum required of any system of law, and a theory of punishment not being strictly necessary, the axiom implies no theory of punishment. Settlers who first make contact with one another must assent to the legitimacy of consensual exchange between any of them, but as long as they continue to interact peacefully with one another, they have no need to agree upon the proper responses to the violations of the law. Principles of self-defense and punishment, therefore, are not an a priori part of law, but must instead develop as needed within a system of free market law.

In our theory, rights are defined in terms of the axioms of law: they are whatever results from the entire history of consensual interactions. When one person transgresses upon another without a resolution having been predefined by any prior agreement, the process by which rights are established has been disrupted and our theory of rights no longer applies. We are left with no process by which rights can be established. If Alex should steal from Bjorn, for example, we certainly want to say that Bjorn may retrieve his stolen property. However, to do so will necessarily involve a transgression on Alex; it may be necessary to trespass on his land, to assault his person, to destroy his property, and even to kill him. Depending on what was stolen, any of these may seem reasonable or unreasonable. Is it reasonable to trespass to retrieve a stolen thumbtack? Is it reasonable to kill a man to retrieve an automobile? One can propose ad hoc principles to decide the matter purely on common sense, but there is no reason to believe that everyone is going to assent to them.

From an a priori standpoint we cannot even say that the just retribution to a crime should not involve infringing upon a third party. Though this might at first seem to go against all intuition and common sense, it is easy to come up with examples that make it sound plausible. For example, suppose a man is threatening to blow up a bomb that would kill many people, but is protecting himself with an innocent hostage. In this case, the best response may be to fire at the criminal even if this puts the hostage at risk.

Having gone outside the bounds of the libertarian axiom once a crime takes place, we are left with no a priori method of saying precisely how to return within those bounds. The main problem that any theory of retribution must overcome is that the severity of a transgression is subjective; therefore people will disagree on whether a transgression deserves a given punishment. When there is a conflict, rights are ambiguous until the conflicting parties should agree that the conflict is resolved and establish ownership over the disputed property once again. Unfortunately, there is no guarantee that this will ever happen. Other people can get involved and act as agents to either side. When a criminal is sentenced to a punishment, anybody may decide that the punishment is worse than his original crime and attempt to rescue the criminal from it. There is a danger that a transgression will have a protracted resolution as more and more responses pile up on it without a general agreement an which ones were just and which were not. We know this to be a very real danger given the history of vendettas all over the world.

Although I have shown that the axiom of consent implies no theory of retribution, this does not show that a rationalist theory of retribution is impossible. Therefore, I have would now like to discuss a theory proposed by Rothbard and elaborated by Kinsella which purports to be such a rationalist theory of retribution. This is the theory of proportionality, which says that the aggression imposed upon a wrongdoer should not exceed his aggression against his victim. Thus, a man who stole $100 would have to give up $100 to his victim, after having returned the $100 he originally stole. Rothbard does not give an argument for this principle and perhaps believes that it follows from the axiom of consent. Kinsella points out that a criminal cannot argue against his own punishment, provided it is proportionate to his crime, without violating the axiom of moral equality. This is a very good argument, but it is not good enough to provide a substantive theory of punishment. To say that the punishment must not be disproportionate to the crime is a completely unhelpful criterion; whether a punishment is disproportionate or not is entirely subjective. Although the principle of proportionality may seem to give reasonably good answers in most cases, taken in general it is nothing more than a useless tautology.

Suppose Alex punches Bjorn, leaving him with a black eye. By the Rothbardian theory, Bjorn is then entitled to give Alex two black eyes. In the first place, let me say that this conclusion seems absurd to me even granting Rothbard’s premises. In the case of the thief who stole $100, he must relinquish $200 to the victim because the first $100 is simply a return of what had already been stolen. However, in the case of a black eye, nothing is stolen and therefore there is nothing to return. Rothbard ought to say that Alex deserves just one black eye, and perhaps may be billed for Bjorn’s medical treatment on his black eye. Now, let us look at whether this conclusion actually makes sense. What if Bjorn is a hemophiliac? Then to him, a black eye would be a life-threatening injury, far worse than any pair of black eyes he could gives to Alex. On the other hand, suppose that Bjorn suffers from analgia. His black eye would then cause him no physical pain at all, so the proportionate response to Alex might then be much less than a black eye. It is therefore not necessarily true that injuries given in kind are proportional to one another.

Suppose that Alex has only one eye, and suppose that he pokes out Bjorn’s eye. Is it really a proportional punishment to poke out Alex’s remaining eye, leaving him blind? A one-eyed man’s eye is likely to be of greater marginal utility to him than one eye of a biclops. Just or not, to poke out Alex’s eye is certainly not proportional.

Similarly morbid examples can be endlessly hypothesized. Suppose a legless man chops off another man’s toe. Does he just get away scott-free, or do we start to chop at his fingers? Is a finger proportional to a toe?

Suppose that someone were to kill my cat. Am I then entitled to kill two of his cats in retaliation? The whole idea is absurd; I would never kill an innocent cat! Must the attacker then get away scott free? Am I to be defenseless against the felinacidal maniac? One might propose that he is to reimburse me twice the market value of the cat, but this, too, is absurd: the market value of a cat is practically nothing, and in any case is far less than the value of my own cat to me. Such a punishment would not even begin to cover the injury done to me. The only way of objectively establishing some value for the crime, and therefore to determine what I am owed by the criminal, would be for the crime itself to have a market value. Its value may then be appraised according to the conditions of the market. However, I am getting ahead of myself already. 

Even in matters involving stolen money the proportionality principle as interpreted by Rothbard can lead to absurd answers. Suppose a man steals money during a period of hyperinflation. If he is apprehended by the end of the day, the money he stole might be practically worthless. What was a huge loss one morning could be a minor inconvenience later that evening. If the punishment cannot be more than what was stolen, then there are circumstances where terrible crimes have insignificant legal sanctions.

Kinsella does not discuss self-defense in his article, but as I recall Rothbard seems to apply the same principle of proportionality to self-defense as well. It is of course easy to come up with similar absurdities here. Suppose a pickpocket steals something and then hides it in such a way that to retrieve it would cause far more property damage than the cost of the original item. He might lock it in a state-of-the-art safe whose value is far greater than that of the stolen good. Assuming he then refuses to open the safe, is it then impossible to retrieve the stolen item without committing a crime?

Roderick Long proposes in his series of lectures that a punishment need not always be in kind but rather that the criminal may always choose to pay with money instead. Long’s proposal in many ways seems to make more sense than Rothbard’s because it does not require that a criminal be punished according to the way he violated the rights of others. There are many things a criminal might do that would simply be too horrible to do back to him. Long’s proposal moreover emphasizes restitution to the victim rather than simply that the criminal be symmetrically brutalized. It makes much more sense to me that a criminal who had assaulted another man should be required to pay his victim’s medical bills and some extra rather than that he should endure the same injury. However, Long’s proposal does not get around the subjectivity of value and hence does not objectively establish what sorts of punishments are just and unjust.

It is clearly not enough for the criminal to pay the market value of what he destroyed or stole. His victim not having already sold his property at the market price indicates that he values those items more than their market value. On the other hand, the criminal and his victim cannot be expected to negotiate a fair price for the crime. We might threaten him in some way so as to force him to negotiate reasonably, but it is still a subjective matter as to whether it is just to follow through on those threats.

Long’s theory also fails to be a priori in a much more obvious way than Rothbard’s. Because money does not exist in the state of nature but rather requires a history of market interactions before it exists, a theory of retribution that involves paymens in money cannot be a part of a priori law. Putting a money value on transgressions is only possible in societies that actually have money.

Any theory of retribution runs against the problem of the subjectivity of the just retribution. This problem is insurmountable for any attempt at an a priori theory because there is no reason that people must agree on the just retribution in order to form a peaceful society. The search for the just retribution is really just as foolish as the just price.  It is possible to propose common-sensical theories that will make sense in most cases, but for the most general theory of retribution, one must describe a process by which principles, whatever they may be, can become generally established in a society. Though one cannot predict those principles in detail beforehand we can know that there certainly will be principles sufficient to resolve the ambiguities in dealing with a criminal.

The system of free market law I have described in earlier posts describes such a process. People will naturally tend to create a system of law among themselves because they have a need to foresee conflicts and establish a means of resolving them beforehand. The division of labor being so much more efficient than self-sufficiency, a particular industry will devote itself to the problem of foresight in contractual matters. It will produce bodies of law that predefine the rights for any sort of conflict, which people will then adopt to fill the gaps of foresight of all their contracts. People may then sign on to these bodies of law with various organizations to demonstrate to one another that they are law-abiding and have agreed to certain legal procedures should they be implicated in a crime. To the extent that crime is an important issue to them, then they will have a strong incentive to bind themselves to a system of law; those who do not will find it difficult to form business relationships with other people. Through the organizations of law, everyone will have a certain contractual relation with everyone else, even those he has not met. From this marketplace of laws, standards will develop that state what retributions are appropriate to any transgression. The free market of law is indeed a market that can establish the value for crimes; people will not sign up for systems of law or to courts that mandate punishments which are too harsh or too lax. The just retributions to crimes are therefore established according to the prefferences people demonstrate in the market.

Although it is possible that standards of law could evolve which seem oppressive, it is because people have found them better than the alternatives that people have agreed to them enough that they have become standardized. However, since the law must be flexible enough to please everybody, I think it can be expected that they will be fairly permissive. Groups which prefer to live under stricter rules must resort to banishment as the punishment for those who don’t follow them.

A system like this, one that naturally emerges in any peaceful anarchistic society, completely resolves the problem of subjectivity. The criminal has already stated beforehand what rights he should lose for what kinds of crime. He has agreed to the system of justice that will be applied to him, and he has agreed that his sentence is the just consequence of his actions. Everyone else has also agreed to respect the law and to the circumstances under which a victim shall have exceeded his right to self-defense or to punish his transgressor. Therefore, it is now an objective matter whether any form of retribution is just and it is no longer to be expected that crimes may escalate into endless vendettas.

There is of course more that needs to be discussed on this issue, and that will be the subject of another post. Something still must be said about the tough cases in which a criminal or victim has not agreed to some sort of law. In addition, although we have proposed a theory that allows us to state just sentences for criminals, this is a logically separate matter from the actual infliction of those sentences. We have a theory of the judge, but not yet one of the executioner.