[One version of this book review was published in the University of Michigan Law Review; since this is from my hard disk it is probably not identical. I am grateful to that journal for its permission to republish the article here]
Robert C. Ellickson, Order Without Law, Harvard University Press, Cambridge 1991.
This book starts with a famous academic parable--Ronald Coase's parable of the farmer and the rancher, intended to explicate the relation between liability rules and land use. Coase asked how the behavior of a farmer and a neighboring rancher would be affected by whether the rancher was responsible for damage done by straying cattle. In one of the first statements of what came to be called the Coase Theorem, he concluded that, as long as the parties could readily bargain with each other over their actions, the legal rule would affect the pattern of side payments--who had to pay what to whom--but not the use of the land.
Shasta county California is a patchwork of open range and closed range, designed by the accidents of history to test Coase's conclusion. In open range the legal rule is that the farmer is responsible for fencing out straying cattle, while in closed range the rule is that the rancher is responsible for fencing his cattle in. Allowing for the fact that real legal rules are somewhat more complicated than academic parables, the situation corresponds almost exactly to Coase's hypothetical.
Robert Ellickson looked at what happened in Shasta county, using techniques of investigation modeled on the approach of Law and Society scholars. He used interviews, aerial photographs, and searches of the relevant records to try to discover what actually happened with regard to straying cattle, land use, fence building, and a variety of related issues. He studied an area that included both closed and open range over a period that included one successful and one unsuccessful attempt to persuade the board of supervisors to convert territory from open to closed. His conclusion was surprising and interesting. Coase was correct in predicting that land use would be unaffected by the legal rule. He was incorrect in predicting that side payments would be affected.
So far as Ellickson could tell, the legal rule on who was liable for straying cattle had no effect at all on the behavior of ranchers and farmers, insofar as it involved cattle straying onto the farmers' land. His observations and interpretations appeared in a series of articles, of which the best known is probably "Of Coase and Cattle: Dispute Resolution among Neighbors in Shasta County."[1] He has now combined and expanded the articles into a book that should be of quite general interest, to legal academics and many others. His argument draws on both the Law and Economics and the Law and Society traditions, combining economic analysis, anthropology, sociology, game theory, and a variety of other approaches to understanding human society.[2] One of his conclusions is that law is very much less important than most of us believe.
What Ellickson discovered about Shasta County was that interactions between neighbors, with regard to straying cattle and many other things, were controlled not by law but by a system of norms, a private law code having no connection to courts, legislatures, or any other agency of state power. When a rancher was informed that one of his animals was trespassing, he was expected to apologize, retrieve the animal, and take reasonable precautions to keep it from happening again. If significant damage had been done, the rancher was expected to make up for the damage--in one case by helping replant a damaged plot of tomatoes.
The system was self enforcing. If a rancher consistently let his animals stray, or failed to offer to make up for significant damages, the victim would respond by initiating true negative gossip--spreading the word that that particular rancher was not behaving in a proper neighborly way. If that failed to work, straying animals could be transported far from the victim's (and owner's) property--imposing significant costs on the owner who had to retrieve them. In extreme cases trespassing animals might even be deliberately injured. The one thing good neighbors did not do, even under severe provocation, was go to court.
Straying cattle were not the only thing to which legal rules were irrelevant. California has quite detailed laws specifying under what circumstance one of two adjoining landowners can build a fence and charge part of the cost to his neighbor. Landowner's in Shasta county build fences, and their neighbors sometimes end up paying for them. But what fences get built and who pays for how much of the cost are unaffected by what ought to be the relevant law.
Legal rules do seem to have some effect on how willing ranchers are to let their cattle stray onto highways, where they might be hit by passing motorists. Ranchers are much more careful to avoid such accidents in closed range than in open range--that, according to Ellickson, is the main reason that people care enough about the legal categories to lobby the board of Supervisors for and against proposals to close areas of previously open range. The reason ranchers are more careful is that they believe that in open range the driver is strictly liable for damages to cattle (the motorist "buys the cow") while in closed range the rancher is strictly liable ("buys the car").
The odd thing in the situation is that, by Ellickson's account, the ranchers are wrong: Whether an accident occurs in closed or open range has little or no effect on who is liable for what after a collision. The one form of behavior affected by the legal distinction between closed and open range is one to which it is almost irrelevant.
Ellickson's account of how conflicts involving traffic accidents with animals are settled raises a number of different issues. The first, and simplest, is why the same people who are almost never willing to go to court on one set of issues are frequently willing to on another. The answer is fairly clear. Both trespass and the construction and maintenance of fences involve frequent minor conflicts between neighbors--people locked into a long term relationship with each other. Highway accidents are rare, sometimes large, conflicts between people who usually do not know each other and have no reason to expect to be in contact in the future. On almost any plausible theory of norms, it is easier to use them to deal with the former than the latter sort of problem. And indeed it appears, from Ellickson's account, that minor highway accidents involving neighbors--the kind of accidents for which norms are adequate--do get handled by norms rather than law.
The second and harder problem is to explain why ranchers are ignorant of the relevant law--why they believe that they "have the rights" in open range, when in fact highway accidents in both open and closed range are covered by essentially the same rules of negligence. The corresponding ignorance with regard to trespass and fence law is easy enough to explain. Since these conflicts are resolved by norms rather than by laws, knowledge of the law is of no value to the people concerned, and there is no good reason for them to pay the cost of acquiring it. They are exhibiting what economists' call "rational ignorance."
But accident claims are resolved by law rather than by norms. Furthermore, according to Ellickson's account, the ranchers' false beliefs about the law actually influence their behavior in at least two important ways. Farmers who have been grazing their cattle during the summer in unfenced areas[3] (typically upland forests leased from timber companies, the U.S. Forest Service, or the Bureau of Land Management) move elsewhere when the range is closed, not because they are worried about being sued for trespass but because they are afraid that one of their cattle will wander onto a road and be involved in an accident, leaving them liable for the damages. And ranchers, whether or not they let their cattle roam free, campaign fiercely against proposals to close additional areas of range.
Assuming that Ellickson has correctly described the law, the ranchers' beliefs, and the ranchers' behavior, his observations seem inconsistent with the economist's usual assumption of rationality. Information about law is not free, but it is not all that expensive. By Ellickson's account, the ranchers are aware that the courts have in several cases found the owners of cattle liable for accidents that happened in open range--striking evidence that the ranchers' view of the law is wrong. Instead of changing their views, they interpret the cases as experimental error--the judge got the law wrong.[4] That does not look like a rational approach to processing information when something valuable is at stake.
Ellickson's explanation of the paradox is that political clashes over proposals to close parts of the range are symbolic rather than actual battles. The ranchers are really fighting not over liability law but over status. By winning, they establish the fact that Shasta County is still Marlboro Country--a place where the ranchers are in the saddle, politically and socially as well as literally. Since they are not willing to admit, even to themselves, what they are fighting over, they instead maintain their false beliefs about the law in order to provide a plausible excuse for their own actions.
As an explanation of political behavior, this may make a good deal of sense. If the only decision affected by ranchers' beliefs about the law was whether to oppose petitions to close the range, then the symbolic stakes involved, although they lie outside the sort of values usually included in economic analysis, might provide a rational justification for the ranchers' actions. Their beliefs would then be a costless way of concealing, from themselves and others, the nature of what they were doing.
But this is a much less convincing explanation in the case of ranchers who change where they herd their cattle in response to such false beliefs.[5] If acting on more accurate beliefs about the law would make them better off, it is hard to believe that there is no excuse they could find for politically opposing closed range ordinances while continuing to herd in closed range.[6]
The answer, I think, is that acting on more accurate beliefs about the law would not make the ranchers better off. Although their beliefs are incorrect, their actions are probably correct.
The explanation for this paradox is that the cost to a traditionalist rancher--one who lets his cattle run free during the summer--of not leasing land in closed range areas is essentially zero, since land is available to lease at the same price in open range areas. Although the difference in expected liability costs between open and closed range is not nearly as great as the ranchers believe, it is quite likely, for reasons Ellickson discusses, that there is some difference--that under some circumstances an animal's owner might be found liable for an accident in closed range when he would not in open range.[7] If so, then it makes sense for traditionalist ranchers to avoid closed range areas.
This explanation suggests two new puzzles. Why is the price for leasing land the same in both areas, when the consequence is that the owners of unfenced grazing land in closed range have no tenants and no rents? Why do the owners not cut the price they charge enough to compensate the ranchers for the small increased cost of liability? And, given that this is happening, why do we not find that the owners of such land, rather than the ranchers who rent it, are the ones leading the opposition to petitions to convert areas from open to closed range? It is the landowners, after all, who are actually bearing the cost of the conversion.
The answer to both questions is very simple. Most of the land used for unfenced grazing is owned by either the Bureau of Land Management or the U.S. Forest Service. Neither organization is willing to delegate to its local agents the authority to set price--with the result that both of them charge a uniform price for grazing leases across the entire western United States. The remaining land is owned by large timber companies, whose policies are only slightly more flexible.[8] The amounts at stake in grazing leases on upland forests are not large enough to make it worth the cost, to either the public or the private owners, of overcoming the principal-agent problems that prevent them from either fine tuning their leases or intervening in local politics to prevent areas they own from being closed.
If the explanation I have offered is correct, then Ellickson's account of how ranchers behave is consistent with conventional economic ideas of rationality. Rationality, after all, is an assumption about what people do, not why they say they do it. Ellickson's explanation of why ranchers persist with their false beliefs does suggest interesting possibilities in the direction of explaining why holding false beliefs may be rational--not because the beliefs are true, but because they are useful.
Ellickson does not limit his discussion to Shasta County. He also discusses, in somewhat less detail, norms that apply to orchard men in the Pacific Northwest, whalers in the 19th century, and modern American academics. The last example is the cruelest, at least so far as his academic readers are concerned, since he offers evidence that professors, in photocopying each other's articles, ignore the relevant copyright laws while adhering to professional norms well designed to serve the interests of our academic community--in some cases at the expense of our publishers.
One of the charms of the first--descriptive--half of the book is the way in which it uses simple observations of the real world to throw cold water into the faces of legal theorists of almost all sorts. It is hard to see how legal rules can serve the roles assigned to them by doctrinal scholars, economists, feminists, or practically anyone else, if nobody pays any attention to them.
Part of what Ellickson accomplishes in this book is an attack on legal centralism, the doctrine that assigns law a central role in determining what people do and what society is like. His purpose, however, is not destructive but constructive. It is to explain the nature of the institutions that actually do a large part of what law is supposed to do. His central thesis is a simple one--that close-knit groups tend to develop efficient norms. He concludes that while formal law is important and useful in human affairs, it is less important and less useful than generally believed. In a wide variety of situations, people not only succeed in resolving their conflicts without recourse to law, they do it by mechanisms that work considerably better than the legal system.
One important issue Ellickson does not explore is where such norms come from and why they tend to be efficient. If rules are well designed, the obvious explanation is that someone designed them. While there may be some systems of religious norms that were actually the result of deliberate design, that does not seem a very plausible explanation for the norms that apply to straying cattle and fence building in Shasta country, or to who owned a whale when two different ships had been involved in killing it.
The obvious alternative to deliberate planning is evolution. Perhaps, over time, societies with better norms conquer, absorb, or are imitated by societies with worse norms, producing a world of well designed societies. The problem with that explanation is that such a process should take centuries, if not millenia--which does not fit the facts as Ellickson reports them. Whaling norms, for example, seem to have adjusted rapidly to changes in the species being hunted.
Perhaps what is happening is evolution, but evolution involving much smaller and more fluid groups than entire societies. Consider a norm, such as honesty, that can profitably be followed by small groups within a society, applicable only within the group. Groups with efficient norms will prosper and grow by recruitment. Others will imitate them. Groups with similar norms will tend to fuse, in order to obtain the same benefits on a larger scale. If one system of norms works better than its competitors, it will eventually spread through the entire society. When circumstances change and new problems arise the process can repeat itself on a smaller scale, generating modified norms to deal with the new problems.
This conjecture about how norms arise and change suggests a prediction: Even if a norm is efficient, it will not arise if its benefits depend on everyone following it. Consider the whaling norms that Ellickson discusses. It is in the interest of any pair of captains to agree in advance to an efficient rule for dealing with whales that one ship harpoons and another one brings in, just as it is in the interest of a pair of individuals to agree to be honest with each other. But a rule for holding down the total number of whales killed so as to preserve the population of whales is useful only if everyone follows it. The former type of norm existed, the latter did not--with the result that 19th century whalers did an efficient job of hunting one species after another to near extinction.[9]
This conjecture explains one feature of the norms that Ellickson observes--their apparent failure to deal with certain sorts of problems. There are at least two other superficially puzzling features that Ellickson discusses. One is the tendency to use inefficient punishments to enforce norms. A second is the apparent preference for non-monetary over monetary payments.
For an example of an inefficient punishment, consider the case of straying cattle. Suppose that, for the tenth time, a herd of your cattle breaks down my fence and destroys my garden. I know, from the previous nine incidents, that calling you up to complain will do no good--you will retrieve the cattle but do nothing to reimburse me for my damages. Calling up a lawyer and suing you would violate the strong norm against taking neighbors to court. What do I do?
The obvious efficient solution, at least in a rural context, is to convert one of the ten animals into packages of meat in my deep freeze, then call you up to complain that nine of your cattle have gotten into my garden. I thus punish you and reimburse myself simultaneously. After a few repetitions you are likely to get the point--and if not, at least I will eat well.
This solution is obvious and efficient, but it is not what happens. Instead of slaughtering one animal, I drive the whole herd to some inconvenient location many miles from both my farm and your ranch. Instead of imposing costs on you in a way that benefits me, thus holding down the net cost of the punishment, I use a punishment that is costly for both of us. Why?
Ellickson recognizes the problem and suggests, although he does not fully explicate, the answer. In his words, "The major disadvantage of a seizure, as opposed to a destruction, is that someone who loses property to an intentional taker is more likely to interpret the event as an act of initial aggression than as an exercise in self-help."
The issue is not merely one of misinterpretation. As Locke pointed out many years ago, every man is a biased judge in his own case. If the victim, sitting as his own judge and jury, knows that the higher the punishment he imposes the larger his own benefit, he will have a strong incentive to find the defendant guilty.[10] If the punishment is costly for both parties, the victim will only be willing to impose it when he really believes that the trespasser has been behaving unreasonably and must be stopped. The victim, knowing that, will be much more willing to accept the judgement and reform his behavior. Thus a norm limiting the injured party to inefficient punishments is actually an efficient rule. It raises the cost of punishment, but it makes it much more likely that a punishment will have its desired effect and less likely that it will set off an escalating feud.[11]
The second puzzle is why the equivalent of damage payments in a system of norms takes a non-monetary form--reimbursing a farmer by helping him replant his garden instead of simply paying for the damage done. Ellickson's explanation is that cash transactions are "cold and impersonal, ... . A consitutive norm favoring in-kind transfers ...puts members through the ritual of signalling that they are in solidarity, rather than at arm's length." But this is not really an answer--it leaves us with the question of why money transactions are seen as cold, and whether this perception, and this characteristic of Shasta County norms, apply more generally. There are plenty of societies, after all, where money is explicitly involved in marriage arrangements, even though an important element of such arrangements is the creation of solidarity between the newly related kindreds. In one of the most famous feuds in literature, the killing match in Njal's Saga,12 Njal and Gunnar maintain the peace by passing a purse of silver back and forth to compensate each other for the killings alternately instigated by their feuding wives.
Robert Ellickson does not, and cannot be expected to, fully answer all of the question that he raises. Understanding the nature of non-legal norms is a hard problem, and, if it is ever fully solved, the process will almost certainly require a variety of different sorts of expertise, probably including economics, game theory, and anthropology as well as law. What he does do is to make it overwhelming clear that norms are an interesting and important issue, and to make a reasonable start at sketching out an approach to understanding them.
David Friedman
Olin Fellow in Law and Economics
University of Chicago Law School
Back to my home page.