For some years I taught a law school seminar on legal systems very different from ours. The central idea was a simple one. All human societies face about the same problems. They solve them in an interesting variety of different ways. There is no good reason to think that we are wiser than other people in other times and places, so all of those solutions ought to be taken seriously—not as historical curiosities or primitive practices but as ways in which a group of adults tried to solve the same problems that we attempt to solve with our legal system. I am not trying to figure out what is the best legal system—indeed, I doubt there is one. I am simply trying to understand how different systems worked, what problems they raised, how they dealt with them.

I invented the course as a commitment strategy, a way of forcing myself to learn things, and it worked. For most of what I have learned you will have to read the book based on that course that I am currently writing. This chapter is about one particularly striking conclusion: The legal system I invented forty some years ago and described in part III of this book is less imaginary than I thought and much older. What I reinvented in the context of a developed modern society was a more elaborate version of the system on top of which many, perhaps most, of the legal systems of the world were built. That system was stateless and decentralized, and its basic logic was quite simple: If you wrong me, I threaten to harm you unless I am compensated for the wrong.

There is one requirement such a system must satisfy to function as law enforcement rather than extortion—that my threat to harm you is more believable if you actually have wronged me than if you have not. There must be some reason why right makes might.

That requirement has been met in a variety of different ways in different societies. In saga period Iceland, the most organized such system that I have looked at, what converted right to might was the existence of a legal code and a court system. Once a court had outlawed you for failing to pay me damages, everyone else in the society knew that you were the one at fault, with the result that my friends would support me in any subsequent conflict and your friends would not support you—for details, see Chapter 44.

The legal system of the Rominchal gypsies, the largest gypsy group in modern-day England, is a more primitive version of the same system. If you wrong me, I threaten to beat you up if not compensated. Both of us know that, if you have wronged me as judged by the customary rules of our society, my friends will back me and yours will not back you. Hence if you are in the wrong you either compensate me or leave town.

There are many other examples of societies, past and present, whose legal system works on the same principle. Most enforce traditional rather than legislated law. Most have some formal mechanism for settling disputes, some equivalent of my arbitration associations. In some, such as the Icelandic, enforcement is by the victim and an ad hoc group of his friends, relatives and allies. In others, such as the traditional legal system of Somaliland, the northern part of Somalia, the coalition that supports an individual in such a conflict is created by an explicit, sometimes written, contract specifying in advance the obligations that the members of the coalition have agreed on, a rights enforcement agency in a society with less specialization and division of labor than ours. In still others, such as the Bedouin, the coalition is defined almost entirely by kinship, membership in a tribe or its subgroups.

Just as in our system, while law is enforced by the threat of violence, actual violence occurs only when the threat fails to work. Despite the popular image of a blood feud with a long chain of revenge killings, actual feuds in such systems usually terminate quickly. A feud system (“feud” not “feudal”; the words are unrelated in both meaning and etymology) may or may not be better than the alternatives, but it is an approach to law enforcement that has functioned successfully over long periods of time in many societies.

One of the things I concluded about feud systems was that they are the foundation on which a variety of other legal systems were built, as demonstrated by the existence within the later systems of the fossilized remnants of the earlier. Traditional Jewish law provides one striking example. If you killed someone, were tried, found guilty of the equivalent of first degree murder and sentenced to be executed, the sentence was supposed to be carried out by the avenger of blood, the heir of the person you murdered. If you were found guilty but of a non-capital killing, you were to be escorted to one of the cities of refuge to remain there until the high priest died—an exile of uncertain length. If, however, the avenger of blood caught up with you on your way to the city of refuge, he was free to kill you. Pretty clearly, that is what was left of a system in which, when a man was killed, it was the right and duty of his heir to avenge him.

Further evidence from Jewish law is the rule forbidding the avenger of blood from accepting a money payment to let the killer off. The fact that there is a law against something is pretty good evidence that it is something people do, or would do if the law did not exist. In this case, the law was blocking the payment of wergeld, the damage payment for killing that, in a feud society such as that of Iceland or Somaliland, was used to prevent the revenge killing that would otherwise occur.

Another example of fossilized feud is provided by Islamic law, which treats killing or injuring someone not as a crime against the state but as a tort against the victim. If the offense is a capital killing, the kin of the victim are entitled to retaliate. They have the option, however, of instead accepting diya, the Arabic wergeld, a fixed payment (specified in camels) to compensate the kin for the killing, or of accepting a smaller payment, or no payment at all. If the offense is only an injury, the victim has the option of accepting a fixed damage payment whose amount is specified in law, agreeing on a smaller payment, or simply forgiving the offense. Pretty clearly, this is a remnant of Bedouin feud law adopted by Mohammed into his legal system.

The two most important modern examples of legal systems built on the ruins of feud law are European civil law, based on Roman law, itself based on a feud law whose remnants can be seen in the Law of the Twelve Tables, and Anglo-American common law. The latter evolved out of an Anglo-Saxon legal system similar to the Icelandic through a process in which the King claimed that certain offenses were violations of the King’s peace as well as of the rights of the victim, hence that the offender owed damage payments to both.

Feud law, in substance although not in form, is alive and well in modern day America. Consider high tech patent litigation, where the potentially profitable wrong of suing someone who has probably not violated your patents in the hope that you can persuade a judge or jury that he has is deterred by the threat that he will retaliate in kind. The mechanism by which right makes might in that context is provided by a court system which makes it harder to win a case when you are in the wrong than when you are in the right—although perhaps not hard enough. The most notable failure of that system is the problem of non practicing entities, referred to by their critics as patent trolls, firms that buy patents in order to sue or threaten to sue other firms that it claims infringe them. Patent trolls do not themselves produce anything, so cannot be counter sued, so are immune to retaliation.

For a solution to that problem borrowed from Periclean Athens, whose legal system I like to describe as the invention of a mad economist, you will have to read my next book.