Contents

• Brinig, Margaret F. From Contract to Covenant: Beyond the Law and Economics of the Family. Review by Marygold S. Melli.
• Cooter, Robert D. The Strategic Constitution. Review by David D. Friedman.
• Friedman, David D. Law's Order: What Economics Has to Do with Law and Why It Matters Review by Eric A. Posner. Reply by David D. Friedman.
• Malloy, Robin Paul. Law and Market Economy: Reinterpreting the Values of Law and Economics A Memoir. Review by Lawrence E. Mitchell. Reply by Robin Paul Malloy.
• Posner, Eric A. Law and Social Norms. Review by Sidney W. DeLong.

• Eastland, Terry, editor. Freedom of Expression in the Supreme Court: The Defining Cases. Review by Ronald K.L. Collins.
• Talkback

Law and Economics for the Masses
by Eric A. Posner

Law's Order: What Economics Has to Do with Law and Why It Matters
David D. Friedman
Princeton, NJ: Princeton University Press, 2000
Cloth: $29.95
Pp. 329
ISBN: 0691010161

Law's Order is the first popular treatment of law and economics. It is intended for students and academics unfamiliar with the economic analysis of law, but also for the "intelligent layman," and it is the latter person who seems to be David Friedman's imagined reader. In form and style, the book resembles the popular but sophisticated surveys of evolutionary biology, like Richard Dawkins's The Selfish Gene (Oxford University Press, 1990) – and, indeed, Friedman cites that book as a model. (5) Friedman's book largely succeeds at the task that he sets out for himself. It is a clearly written, readable, and interesting introduction to the main ideas underlying law and economics, alongwith a large collection of specific applications.

The book has two parts. The first describes general concepts that recur through different legal-economic applications – concepts like public good, externality, and transaction cost. The second contains the applications – property, contracts, family law, torts, criminal law, antitrust law, plus a few other areas. The ideas discussed are but a sample of these topics, and these topics are but a sample of the literature.  Civil procedure, corporate law, constitutional law, and many other topics are neglected, for lack of space.  There is very little math, and disappointingly few numerical examples. Most students don't acquire subtle but fundamental intuitions, like the difference between marginal and average effects, without working through some numbers. For this, A. Mitchell Polinsky's Introduction to Law and Economics (Little Brown, 1989) remains a necessary companion to the student, though not the general reader. Finally, the book contains little bibliographic guidance. Students interested in the sources of the ideas must consult Friedman's website for the book, where there is also some math and informal discussion, but the division of materials between book and Internet, while imaginative, will be frustrating for people who do not read books in front of a computer screen.

The writing is consistently good, and although it never contains the razzle-dazzle of a Richard Dawkins, it is sometimes terrific. Chapters 4 and 5 provide as good an introduction to the Coase Theorem as one is likely to find anywhere.  The book is also gratifyingly undogmatic. It acknowledges the difficulties with the various efficiency criteria, the indeterminacies that beset evaluation of economic arguments, the corners into which economic analysis can drive one (see the discussion of "efficient murder" on 224), and the many theoretical arguments for market failure and government regulation even in areas felt by the conventional libertarian-economic mind to be sacrosanct, like private contracting. One discerns Friedman's worldview, but he does not try to force it on the reader, and instead devotes himself to the methodological project of illustrating the ways in which economics sheds light on law.

Praise is boring, so now I will turn to some disagreements.

Contract Remedies

Friedman reproduces a widely believed but incorrect or incomplete argument about the superiority of expectation damages to specific performance. When sophisticated parties enter contracts, they anticipate that one or both parties might want to breach rather than perform. To simplify, suppose that Seller promises to deliver a widget to Buyer at a later time. The parties know that the cost of production will be some amount between $100 and $200, and they settle on a price of $150. Buyer's valuation of the widget is, say, $160. Ex post efficiency requires that trade occur only if Buyer's valuation exceeds Seller's cost.

In the absence of any remedy for breach of contract, Seller will decline to perform if its cost turns out to be more than $150; it will not incur, say, $155 in costs in order to obtain $150 in cash. It can be shown that expectation damages gives Seller the right incentives to breach – that is, to breach only if its costs exceeds Buyer's valuation of $160.   By contrast, specific performance requires Seller to deliver the widget even if it costs $200 to produce, and while Seller can try to pay Buyer for a release, renegotiation is often expensive and unsuccessful. Therefore, expectation damages dominates specific performance (unless courts are likely to have trouble determining Buyer's valuation). (162-165)

The argument overlooks the parties' ability to plan in anticipation of the remedy. If the remedy is specific performance and the parties know that renegotiation might be costly, they can enter a contract holding that "Seller must deliver the widget unless its cost exceeds Buyer's valuation of the widget," or "Seller has the option to refrain from delivery upon payment to Buyer of Buyer's valuation." Both contracts allow Seller to avoid inefficient trades, and the latter requires the court to have no more information than it needs to award expectation damages. Specific performance does no worse than expectation damages, and the recent literature suggests that specific performance is, except in narrow conditions, superior.

And this should not be surprising. Specific performance, unlike expectation damages, protects the terms chosen by the parties rather than monetizing them, and so the usual defenses of freedom of contract also support specific performance. The justification for expectation damages – that it permits efficient trade when unanticipated events occur – overlooks the ability of the parties to anticipate that unanticipated events will occur, and to design the contract to duplicate the expectations remedy in those cases in which that remedy happens to be ex ante optimal, and not otherwise.

Criminal Punishment

The economic theory of criminal enforcement recommends the maximum sanction possible for bad crimes that are hard to detect. This suggests that criminals should be fined their entire wealth, and if that is not adequate to deter others from criminal behavior, they should be enslaved, tortured, or executed, with their organs taken for transplant. (237) After criticizing a number of objections to this view, Friedman argues that we reject such "efficient" punishments sometimes because they give officials an incentive to "expropriate" citizens. (240) To support this view, Friedman notes criticism of drug forfeiture laws, which give police an incentive to pursue non-dangerous drug offenders who happen to have valuable assets. But it is hard to see what officials gain by removing the organs from prisoners as long as the government takes the obvious precaution of not letting law enforcement officials keep them. Government can reduce agency costs more effectively by paying officials a fixed fee, or monitoring them closely, or declining to offer bounties, than by reducing punishments of criminals.

On a related note, Friedman wonders about the widespread taboo against cannibalism, when in fact dead people are not harmed when they are eaten. His too-clever-by-half argument is that if human cannibalism were permitted, people would kill each other for dinner. The taboo removes this incentive. (242) By the same token, a taboo against the enjoyment of any property, or property above some minimum level, would remove the incentive to steal.  Why steal if you can't enjoy what you took, or convert it to cash and spend it on things you want?

Norms

Following Robert Ellickson's Order Without Law (Harvard University Press, 1994), Friedman argues that social norms are often efficient. Like Ellickson, Friedman does not mean social norms like norms of dress, gift-giving, or politeness. He means the informal methods that have arisen for resolving disputes in small groups.  When cattle trespass on the land of farmers and ranchers of Shasta County, the landowner generally returns them regardless of whether the law gives him the right to demand payments from the owner of the trespassing cattle, and without any need of side payments in the spirit of the Coase Theorem – the outcome which is, according to Ellickson and Friedman, efficient.

Aside from empirical problems that I won't go into, this argument raises the question why we would expect people to conform to social norms. Friedman weakly says: "The reason that norms trump law in Shasta County is simple: One of the strongest norms is that neighbors don't sue neighbors. Anyone who goes to court to enforce his rights automatically loses his case in the court that matters most, the court of local public opinion." (276) It is indeed a norm that neighbors don't sue neighbors, and not just in Shasta County.  But this argument is circular: why doesn't law trump the social norm of not suing neighbors? Conversely, if public opinion can prevent the second-order violation of norms (the use of courts), then public opinion should also be able to prevent the first-order violation (letting my cattle stray onto your land, or failing to return your trespassing cattle). The general question, not answered, is why people would obey any norms, whether first-order or second-order.

The puzzle is that social norms, as Friedman and Ellickson understand them, are supposed to solve collective action problems, and yet the power of the collective action idea is its demonstration that rational individuals will not generate public goods unless an outside force, like government, enforces cooperation.  Ellickson himself appeals to some game theoretic models. Friedman presents a confusing evolutionary theory, according to which groups with efficient norms prosper, so other groups imitate them. (277) But what does it mean for a group, which by hypothesis is not controlled by a leader or government, to imitate the norms of another group? If the behavior produces a public good, then members of the second group have every incentive to free ride, even if they know that the result is that their group will not prosper.   And if this is so, how did the efficient norms of the first group arise?

Costs & Benefits

Friedman provides a nuanced (given the space) account of efficiency, describing its different definitions and uses, and acknowledging its many limitations as a normative criterion.  And these limitations are serious, including the uncertain relationship between "preferences" and "welfare," the distortions introduced by differences in endowments, and the neglect of non-consequentialist moral intuitions. Friedman's justification for using efficiency as a criterion for evaluating legal rules is twofold. First, although efficiency is not all that matters, it matters enough in enough different contexts that it is relevant for evaluating legal rules. Second, efficiency is a methodological concept, and applied unflinchingly it discloses truths about the world that we might not otherwise notice.

These are both fair points, but they are also in tension. Take Friedman's argument that in choosing optimal criminal law we should take account of the criminal's tastes as well as those of victims and the public. Should we punish the sadistic murderer less than the casual murderer, because the former derives real pleasure from the act and we don't want to overdeter pleasurable murders? I can't tell how Friedman would answer this question. He insists that by taking preferences as given, we force ourselves to learn new things. But no economic explanation, as far as I know, says that the sadistic murderer should be punished more, holding enforcement costs constant. Friedman appears to agree with, or at least entertain, the possibility that we should not give weight to morally objectionable preferences when choosing among legal rules. (See 22 and 231, but see 23 and 230).   But then what happens to the methodological concerns?

There is another confusion in the methodological discussion. A well-known defense of the efficiency criterion is that, while it ignores distributive concerns (a redistribution can never be efficient, unless you assume altruistic preferences, because the winners could not give the losers an amount greater than the winners obtained), distributive concerns are best addressed through taxes and transfers rather than "general" legal rules like tort or contract law. But there are two separate issues here. One issue is the best way to redistribute wealth. It is possible that taxes and transfers are superior redistributive instruments to, say, consumer protection laws or minimum wage laws, though the argument is more complex than Friedman's presentation suggests.

But this point is to the side of Friedman's concern, which is our uneasiness with using efficiency as a normative criterion when people have different endowments. We might decide that the tax and transfer system is so clumsy and expensive that we will tolerate a great deal of inequality, even while believing that poorer people have higher marginal utility of income than richer people do. But if that is so, then a rule or project (like a new park in a wealthy area of town) will still reflect valuations that are distorted by wealth, so that the rule or project is more likely to benefit the wealthy than the poor. The very hard problem is that if we adjust cost-benefit evaluation to take account of wealth, it will generally favor projects and laws that have redistributive effects, even if we prefer to redistribute using taxes and transfers. If we do not, then cost-benefit analysis will favor projects and laws that reflect the desires of the wealthy rather than the good of all.

Eric A. Posner is Professor of Law at the University of Chicago, and is the author of Law and Social Norms (Harvard University Press, 2000), reviewed in this issue of Books-on-Law, and editor of Chicago Lectures in Law and Economics (Foundation Press, 2000). He is also co-editor of the Journal of Legal Studies.

Professor Eric Posner correctly observes that the printed version of Law's Order contains little in the way of references or mathematics. He is also correct in believing that notes on a website are less accessible than notes in a book. But while they are less accessible, they are a great deal more useful.

From the website, which contains page images of the entire book, one click on a marginal icon brings up the corresponding note. A second click brings up the full text of a referenced case (in the Findlaw, Lexis or West database), a webbed article, even a webbed book. The reader must be online to take full advantage of the virtual footnotes, but he does not, like readers of more conventional books, have to be in a library in order to make use of their references. For readers who do not expect to be online while reading the book, the website also contains the virtual footnotes as a single long file, in pdf and Word format, to be downloaded and printed.

Professor Posner states: "Friedman reproduces a widely believed but incorrect or incomplete argument about the superiority of expectation damages to specific performance. . . . If the remedy is specific performance and the parties know that renegotiation might be costly, they can enter a contract holding that 'Seller must deliver the widget unless its cost exceeds Buyer's valuation of the widget,' or 'Seller has the option to refrain from delivery upon payment to Buyer of Buyer's valuation.'"

Sophisticated parties can indeed enter into such a contract; when they do so, however, the rule they have agreed to is not specific performance, but "specific performance unless . . . ." I did not attempt to cover all possible terms to which that parties might agree, merely some of the simpler alternatives.

"Specific performance, unlike expectation damages, protects the terms chosen by the parties rather than monetizing them, and so the usual defenses of freedom of contract also support specific performance."

Specific performance protects the terms chosen by the parties if the parties have chosen specific performance; expectation damages protects the terms chosen by the parties if the parties have chosen expectation damages. I am in favor of freedom of contract, but that does not tell us what form of contract parties will choose.

"The economic theory of criminal enforcement recommends the maximum sanction possible for bad crimes that are hard to detect. This suggests that criminals should be fined their entire wealth, and if that is not adequate to deter others from criminal behavior, they should be enslaved, tortured, or executed, with their organs taken for transplant. (237) After criticizing a number of objections to this view, Friedman argues . . . ."

The economic theory of criminal enforcement recommends, not the maximum sanction possible, but expected punishment equal to damage done minus marginal cost of deterrence (227-229). In some cases, such as crimes that are sufficiently hard to detect, that might mean no punishment at all. The starting point of the argument I present and critique is not imposing the maximum sanction possible but imposing the optimum sanction in the form that minimizes the sum of enforcement costs and punishment costs.

"But it is hard to see what officials gain by removing the organs from prisoners as long as the government takes the obvious precaution of not letting law enforcement officials keep them."

This assumes that government is a trustworthy philosopher king, even if the officials who make it up are not. But a government that gains by executing defendants and forfeiting their organs for transplant has good reasons to reward officials that help it do so, as we have observed in the analogous case of civil forfeiture.

"By the same token, a taboo against the enjoyment of any property, or property above some minimum level, would remove the incentive to steal."

A taboo against the enjoyment of property would indeed reduce the problem of theft, but it would also eliminate all uses of property. The difference between that case and the taboo against cannibalism is that the cannibalism taboo does not prevent what is almost always the highest valued use of my body – by me, alive.

"Friedman does not mean social norms like norms of dress, gift-giving, or politeness. He means the informal methods that have arisen for resolving disputes in small groups."

If that were what I meant, I would not have offered norms of honest behavior as an example of the sort of norms I was discussing.

"If the behavior produces a public good, then members of the second group have every incentive to free ride, even if they know that the result is that their group will not prosper."

I agree. That is why I argued that my proposed mechanism can produce norms such as honesty, since "it is in the interest of a pair of individuals to agree to be honest with each other," but not norms for producing a public good: "The former type of norm existed, the latter did not, with the result that nineteenth-century whalers did an efficient job of hunting one species after another to near extinction – which was probably not the efficient outcome." (277)

"But no economic explanation, as far as I know, says that the sadistic murderer should be punished more, holding enforcement costs constant."

"If deterrence is expensive, it may make sense to impose on each sort of criminal just enough punishment to deter most offenses, which requires different punishments for different people." (233)

"Friedman appears to agree with, or at least entertain, the possibility that we should not give weight to morally objectionable preferences when choosing among legal rules. (See 22 and 231, but see 23 and 230).  But then what happens to the methodological concerns?"

I entertain a wide range of possibilities. On 22, I discuss reasons why people might reject the method of analyzing law that I am proposing. The methodological concerns are one of the reasons they might, and I do, accept it.

David Friedman is a professor in the Law School and the Business School at Santa Clara University. His review of Robert Cooter's The Strategic Constitution is published in this issue.

The Strategic Constitution
Robert D. Coote
Princeton, N.J.: Princeton University Press, 2000
Cloth: $45.00
Pp. xviii, 407
ISBN: 0691058644

Suppose you are building a new nation from scratch. You face a daunting array of issues. The easiest is the choice between democracy and dictatorship. With the communist block down to North Korea and Cuba and non-communist dictatorships almost entirely restricted to the poorer nations of what used to be called the third world, it is clear that liberal democracy is the way to go.

But what about the rest of the menu? Do you want a centralized nation like Japan, a moderately decentralized nation like the U.S., or a coalition of sovereign states like the E.U. or the CIS? A unicameral or bicameral legislature?  Legislators elected by plurality vote in their district or by a system of proportional representation? A chief executive chosen by the legislature (England) or independently elected (the U.S.)?

Once such grand decisions are made, a host of details remain, some of which may make or break your project. How many seats should there be in the legislature, and how should they be allocated?  How are powers to be divided among executive, legislature, and judiciary – assuming that you follow such a now-conventional division? Should your local governments be defined by geography (county) or function (school district)? Do you want a small number of governments, each with many layers of administration to cover a wide range of territory and issues, or a large number, each with only a few layers of administration? How and by whom will the officials administering policy be monitored and controlled? What parts of the administrative hierarchy should get wide discretion, and which should be tightly constrained by rules? To what degree and in what ways should government be limited by the rights of citizens?

Robert Cooter attempts these questions, and many more, using the tool of economic analysis, primarily public choice theory and economic analysis of law. Considered as an algorithm for constructing a constitution – plug in the variables and turn the crank – it is, of course, a failure, since that project is not doable at the present state of our knowledge. But considered as a construction kit, a set of intellectual tools for thinking through these issues, it is a very useful book. And it should be useful not only to those readers who happen to be drawing up new constitutions, but also to the much larger number concerned with understanding, and perhaps improving, the existing institutions under which they live.

The economic analysis of political institutions can be approached as a set of ideas, each of which may be applied to a wide range of issues, or as a set of issues, each to be analyzed using a common toolkit of ideas. Cooter, I think wisely, does both. He starts by sketching out a small number of basic ideas, and then applies them to issues ranging from legislative structure to individual rights. The most important of the ideas is the threefold division among voting, bargaining, and ordering as ways of getting things done.

Voting, Bargaining, Ordering

Consider a single-issue initiative or an election in a society with a well defined left/right political spectrum. A candidate a little right-of-center can be defeated by a competitor one step to his left; so the logic of voting tends to produce the outcome preferred by the median voter. That is not always a good thing, since it ignores intensity of preference; a ballot initiative weakly supported by 55% of the electorate will pass, even if the other 45% passionately oppose it.

One solution to that problem is to decide issues by bargaining instead – vote trading within a legislature, for example. But with multi-issue voting, we may no longer have a determinate result, median or otherwise. Preferences are likely to be intransitive: proposal A defeats proposal B defeats proposal C, but C wins over A. The result can be endless cycling, indeterminate results, or a system where outcomes are decided not by the preferences of the legislators but by the preferences of whoever gets to determine what proposals are voted on in what order.

A further problem with bargaining is that it makes the doings of legislators less transparent. A voter cannot easily know whether his representative voted against a proposal that the voter supported because the representative was really opposed to it (in which case he should be voted out at the next opportunity) or in exchange for another representative's vote on some other issue, more important to both legislator and voter. Hence, bargaining in a legislature makes it less likely that representatives will actually represent.  This is a problem that Cooter is aware of but, in my view, gives too little weight; much of the book seems to assume that legislators do what their constituents want them to do, and that local governments do what serves the interest of their residents.

Both voting and bargaining involve substantial costs per decision made – and modern governments make a lot of decisions. Most are by administrative hierarchies of one sort or another. Such hierarchies employ the third alternative: orders.   Subordinates do things because their superiors tell them to.

This raises the problem of agency. The superior ("principal" in the language of agency theory) can only imperfectly monitor the subordinate ("agent"), leaving the agent some freedom to serve his own preferences at the expense of those of his principal. Multiply that problem by many layers of administration, and you may get officials at the bottom almost totally unconstrained by what the elected president at the top wants them to do.

One response is to substitute rules for discretion; it is easier to tell if a subordinate violated a rule (buy the model of computer that goes with your job classification) than a broader standard (buy the computer that provides the features you need at the lowest cost). But in order for superiors to know if the rule is being obeyed, it must be defined in terms of information available to them ("job classification"); an employee with discretion can take advantage of the much more detailed information available to him ("what I'm actually doing"). Hence, the use of rules constrains the agent to follow the principal's objectives at the cost of reducing his flexibility, and thus making it harder for him to follow those objectives.

The Structure of Governments Within a Nation

Having worked through the threefold division among voting, bargaining, and ordering, the author applies it to a range of issues. Consider, for one example, the structure of governments within a nation.  A centralized system has few governments, implying deep hierarchies and a heavy reliance on orders. A decentralized system eliminates many layers of hierarchy, replacing orders with voting: The mayor is elected by the population of the town, rather than appointed by a district official appointed by a regional official appointed by a national official appointed by the (elected) president. This reduces agency problems at the cost of increasing the number of elections and (this is one of the least developed parts of Cooter's analysis) depleting the reservoir of civic spirit that makes elections work.

A second problem with decentralization is that the mayor's decisions may affect people who do not have the opportunity to vote against him; he may, for example, be issuing permits to factories that pollute the river that flows through his town.   The obvious solution is to shift to a more centralized structure, putting all such decisions under a higher level of government with layers of administration leading down to the officials granting permits.

A less obvious solution is bargaining. If decisions made by one local government affect another, or several others, let them get together and voluntarily work out some mutually satisfactory solution.  How well this works and what outcomes it produces depends on the number of governments involved and the effects being bargained over.

Local governments can be constructed on geographical lines or by function. A geographical government "splices" issues: the governor is elected not for his position on education, or on highways, or on law enforcement, but for his combined position on all three. As with bargaining, intensity of preferences matters, but multidimensional issues lead to the possibility of cycling, unpredictable outcomes, and decision by agenda control. A functional government, such as a school board, instead "factors" the issues, producing something more like a median voter result.

Government Design & Judicial Power

Voting, bargaining, and ordering reappear in the design of a single government, most notably the one at the top. Consider the choice between plurality voting for legislators and proportional representation. Plurality tends to produce two strong parties, which makes bargaining over legislation simpler than in the legislature of many small parties that is a likely result of proportional representation. But representatives under a system of proportional representation come closer to actually representing the preferences of their constituents. A similar tension exists in setting the size of the legislature. The smaller the individual district, the more accurately the representative can represent the voters, but the larger the number of representatives who must bargain their way to a majority vote on legislation.

A bicameral legislature with an independently elected president (the American model) requires a three-way bargain to pass most legislation. In a unicameral legislature with a prime minister chosen by parliament (the British model, given the current impotence of the House of Lords), legislation passes if and only if the prime minister wants it to. Bargaining among branches is replaced by bargaining with the majority party over who will be Prime Minister and what legislation he will support.

This has implications for (among other things) the power of the judiciary. A court that attempts to achieve its preferred result by interpreting legislation in a way different from the intent of the people who produced it can be reversed by new and less ambiguous legislation. But new legislation requires cooperation – in the American model, cooperation among both houses and the executive. So a court can be reversed by legislation only if there is some interpretation which all three prefer to the interpretation provided by the court. Cooter concludes that the greater the number of parties required for legislation, the greater the power of the judiciary.

But ...

There is a problem with this argument. Cooter posits that bargaining within a legislature permits parties to express the strength of their views via vote-trading. That, in his view, is its chief advantage over simple one-issue majority vote. But in the context of bargaining among different parts of governments, as in America, he assumes that each issue is decided separately; hence legislation will pass only if all the parties whose consent it requires support it.

Suppose we instead analyze bargaining among two legislatures and the president in the same way in which Cooter analyzes bargaining within a legislature. There is some version of the legislation that maximizes net gains summed over all three parties. If bargaining costs are sufficiently low, that is the version that will pass – even if one of the parties, preferring no legislation at all, must be compensated for its support by getting what it wants on some other issue.   In the American system, such trades can be embodied in multi-issue bills, supported by different people for different reasons. If the court chooses to interpret the legislation in a way inconsistent with what the parties have bargained, the same process will produce a new and clearer version. Whether all three parties prefer the new version to the court's interpretation is irrelevant. Cooter's argument is unconvincing, but its conclusion, for which he also offers empirical evidence, is probably true.  The more parties that must agree in order to pass new legislation, the higher the bargaining costs are likely to be, so a court can probably get away with a wider range of interpretation in a bicameral presidential system than in a unicameral parliamentary system.

This example illustrates a general problem, not so much with Cooter's book as with the wider enterprise of which it is a part. In order to maintain a tolerable level of expositional and analytic simplicity, it is necessary to employ drastically simplified models. The analysis of such models tells us a good deal about how to think about political institutions. But the conclusions they generate cannot be trusted without much more extensive analysis, because it is hard to know to what degree the results are artifacts of the simplifying assumptions.

Competition & Monopoly

If the most important idea of the book is the division among voting, bargaining, and ordering, the next most important is the contrast between competition and monopoly.  Cooter justifies democracy as a system where politicians compete for power, arguing that it produces better results than alternatives for reasons analogous to the advantage of competitive markets over monopolies. He defends the division of power among the legislative, executive, and judicial branches (while noting that it is much less sharp in political practice than in political theory) as a way of making it harder for a majority party to create a cartel among the branches and use it to preserve and increase its power.

Competition for taxpayers forces governments, especially local governments, to give them something for their taxes – ideally, the optimal package of taxes and services.  Less obviously, mobility permits local governments to specialize, with different governments tailored to different groups of potential citizens. This leads to one of the more interesting points in the book: the observation that, in a competitive system, reducing the rights of individuals may sometimes make them better off. If local governments are free to restrict the rights of their citizens to buy, sell, and possess pornography, some communities will ban it, others permit it, and people will locate themselves accordingly. Hence Cooter argues that, the more mobile the population affected, the weaker the constitutional restrictions on local governments should be.

Choice of law provisions permit the parties to a contract to specify that it will be interpreted according to the law of a particular state within the U.S., a particular nation, or the private law provided by a specified arbitrator.   Cooter argues that such provisions force jurisdictions to compete to provide legal rules under which parties wish to contract. He discusses other forms of competition among governments as well, such as permitting parents to send their children (accompanied by a tuition voucher from their local government) to the school of a nearby government. He even discusses the competition between an actual government and a potential government implicit in the threat of secession, arguing that in at least some circumstances it provides a desirable constraint on the ability of a government to exploit a minority within its territory.

Good Theory, Dubious Applications

The book does a good job of sketching out the economic logic of political institutions and the resulting tradeoffs among the available alternatives. It does a less satisfactory job of applying the analysis.  Consider some examples:

In an early chapter, Cooter argues that lobbying is a form of investment, and that there are other investments a firm can make. Hence there are substitutes for lobbying; the demand for lobbying must be highly elastic; and increases in the cost of political favors will result in a decrease in the amount of money spent purchasing them. From this, he draws a variety of further conclusions about the effect of various possible political changes on lobbying expenditures.

But the fact that you can classify two different expenditures as "investment" tells us nothing at all about whether the things they buy are substitutes for each other.  Consider a polity where a firm requires fifty units of political influence in order to be permitted to stay in business. If the price of political influence gets high enough so that the firm buys only fifty units, further increases in price will result in increased expenditure until it reaches a point where the firm, unable to buy the necessary influence, goes out of business. It cannot solve the problem by "investing" in only twenty units of influence, and replacing the rest with an investment in a new factory.

How Not to Tame the Coastal Commission

Mr. Nollan wishes to expand his cottage. The California Coastal Commission informs him that he will be permitted to do so only if he agrees to donate a strip of his property as a coastal path. Should such bargains be permitted? The U.S. Supreme Court's answer (in Nollan v. California Coastal Commission [1987]) was "no." The Commission is entitled to condition permission on Nollan's modification of his plans to reduce the visual impact of development. But it is not entitled to make permission conditional on an unrelated transfer of rights from him to it. Cooter defends this decision as a way of limiting a government's ability to expropriate by regulation, while noting that it may sometimes produce an inefficient outcome. Since "amelioration is usually imperfect," the Commission may simply refuse permission, even though both it and the landowner would prefer development plus unrelated transfer to no development.

Cooter, therefore, proposes an improved version of the Court's rule: the Commission may not require an unrelated transfer, but may require amelioration and agree to accept an unrelated transfer instead. He writes that "to speak of mitigating more than 100 percent makes no sense, so the upper limit to the mitigation is the full extent of the harm," and concludes that his rule will limit the Commission to extracting compensation for the harm done by the development.

The mistake here lies in confusing the value of mitigation with its cost. In the book's example, $300 of redesign produces $250 worth of mitigation. Presumably some larger sum, say $400, would produce $275 worth of mitigation. If mitigation is necessarily imperfect, nothing Nollan can do short of canceling the project will produce complete amelioration. Hence there is no limit as to how much the Commission can require him to spend in coming closer and closer to that objective – a conclusion amply confirmed by the actual history of the Coastal Commission. Unless the damage from expanding the house can be eliminated at a cost that still makes the expansion worth doing, which Cooter has been assuming is not an option, the proposed rule puts no limit on what the Commission can charge for its permission. Thus, it is equivalent to the result which Cooter just argued that the Court correctly rejected.

Campaign Expenditure Limits

Cooter reasons that, by reducing expensive advertising and forcing candidates to depend on donations of labor, expenditure limits produce election outcomes that take account of intensity of preference rather than merely the desires of the median voter. But donations of money also reflect intensity of preferences; and money, like labor, produces votes, not merely median voter outcomes -- which is why candidates want it. Expenditure restrictions simply eliminate one of the ways in which intensity of preferences can be expressed.

Incumbents, as Cooter observes elsewhere in the book, are usually reelected. One reason is that they start the election with large resources of publicity and political support. To match those resources, challengers usually must heavily outspend incumbents. Expenditure limits keep them from doing so, making the political market less competitive. That, Cooter has argued repeatedly, is a bad thing, since competition produces better results than monopoly.

Hence, using a different part of the same toolkit, I conclude that such restrictions are unambiguously undesirable, rather than possibly desirable. That conclusion is reinforced when we note that the total amount spent on campaigning is well under a tenth of a percent of the amount spent by the governments the candidates are competing to control.   Thus, savings from expenditure limitations are tiny compared to the gains from any significant improvement in the functioning of the democratic system.

Conclusion

Despite some lapses at the level of application, this is, on net, a good book. It provides a broad-ranging and intelligent introduction to the use of economic theory in understanding political institutions. And the author is refreshingly open to what a more conventional writer might regard as offbeat options – secession as a way of protecting threatened minorities, restrictions of freedom by local governments as a way of increasing the options available to citizens, transferable development rights, competition among governments, and many more.

David Friedman is a professor in the Law School and the Business School at Santa Clara University. His most recent book, Law's Order: What Economics Has to Do with Law and Why It Matters (Princeton University Press, 2000) is reviewed in this issue of Books-on-Law.

Law and Market Economy: Reinterpreting the Values of Law and Economics
Robin Paul Malloy
New York, NY: Cambridge University Press, 2000
Cloth: $64.95 / Paper: $22.95
Pp. 179
ISBN: 0521782147

This is a frustrating review to write of a maddening book. Frustrating, because Professor Malloy's heart is in the right place, and I deeply want to praise it. Maddening, because despite Malloy's best intentions, Law and Market Economy is painfully obscure and largely irrelevant.

Malloy claims to have developed a new theory of law and economics, one that challenges the neo-classicism of scholars such as Richard Posner and the Chicago School (although most of Malloy's attack is aimed directly and explicitly at Posner). This is the book's irrelevance. Generals are said always to be fighting the last war. In that tradition, Malloy is fighting a war that has ended. The strict neo-classicism of which he complains can, of course, still be found in some, generally not very good, legal scholarship. But law and economics has far surpassed that strict Coasian/Paretian/Kaldor-Hicksian model peddled by Posner and his Chicago brethren like Frank Easterbrook and Daniel Fischel in the late 1970s and early 1980s. As even a casual tour through the legal literary landscape of the last decade will reveal, law and economics (a discipline that I, too, have criticized for its narrowness), while still deeply flawed, has broadened to address many of the concerns that Malloy raises. He is right to raise them. But consider norms jurisprudence (of which he makes only brief critical mention – and his brief critique is right), game theory (especially of the brand practiced by innovative economists like Ken Binmore), behavioral economics, and more "humanistic" approaches to neo-classical economics (of which Robert Frank may be the leading exemplar, using Adam Smith's moral sense theory to far better effect than does Malloy), theories of network externalities and multiple equilibria, and even descriptions of team production. All of them have sprung from the premise, central to Malloy's thesis and purpose, that law and economics must be far more contextual and nuanced in order to provide meaningful explanations of human behavior. None of these theories may get it exactly right, but they're trying, and it is a serious defect in Malloy's work that he barely mentions any of them, let alone attempts to come to grips with them and explain how his theory improves upon these attempts. Almost no serious legal scholar or economist of whom I am aware buys into the kind of law and economics that Malloy is challenging.

It is true that in attempting to create context for economic phenomena, in order to attempt to create models to explain behavior that is considered "irrational" in the traditional law and economics Malloy challenges, economists and legal economists tend to be trapped within a paradigm of rational behavior that they have difficulty escaping (although Frank does a pretty good job of it).   Malloy, however, in failing to address these approaches, gives the reader no basis for understanding why they fail, or how his theory advances his goal of developing a more human law and economics.

And that is the other failing of the book. What purports to be a theory is no theory at all; it is merely a perspective. The semiotic theory on which Malloy relies explains nothing, at least as he presents it, beyond his call for a broader law and economics. Malloy claims to employ semiotics, especially as developed by Charles Peirce, to create a theory of law and market economy that posits that market exchange shapes, and is shaped by, meaning beyond mere wealth maximization in a communal context. This creates the possibility of increasing wealth and enhancing value – indeed, of creating new value through the very process of creative exchange (echoing, of course, Joseph Schumpeter's famous description of capitalism as creative destruction).

Despite the fact that Malloy reiterates (without clarifying) the theory more times than I could count, this seems to be the essence of it. And as the essence of it, it is not theory at all. What is it that the theory predicts? What is it that the theory explains? What is it that the theory really models? Malloy devotes almost two-thirds of the book to attempting to explain what he spends the first third asserting, by use of examples and metaphor, but he often discusses his examples without even applying the theory. When he does, it appears to this reader, at least, that the theory is no more explanatory of his examples than is the simple observation that markets exist in a cultural context and this affects exchange, wealth creation, and the meaning of markets, an observation made with much more specificity, persuasiveness, and force, in the contemporaneously published collection of essays by Lawrence E. Harrison & Samuel P. Huntington, Culture Matters: How Human Values Shape Progress (Basic Books, 2000) or the organizational psychology so importantly advanced by Tom Tyler and his colleagues.

More frustration lies in the presentation of the theory itself. It may not be Professor Malloy's fault, but that of his publisher, that the book appears to be almost entirely unedited. The redundancy of the first fifty pages (which lightens somewhat, but not completely, as the book goes on) would have led me to put it down had I not agreed to review it. John Rawls appears redundant, too, in his own writings.  But in each apparent repetition in Rawls, we get a subtle advance on his theory, a further explanation of what he has meant so far. Malloy simply repeats himself.

He also uses terms in both his theory and his explanations that clearly have some meaning to him, but meaning that is not obvious to the reader, terms that go unexplained. He tells us early and often that markets are "communities" and exist in "communities." What does he mean by communities? He never tells us. Thus, his conclusion that the communal nature of markets shapes values is hollow and unconvincing, even as it is intuitively right. The same is true of his entire semiotic theory. What are the "signs" he is talking about? How do they translate into transformed meaning?   He doesn't ever tell us. He lays the theory out a number of times, but does so repetitively, without adding explanation to what he presumably believes to be self-evident. This is never a good scholarly technique, but it is particular troubling when the author is seeking to persuade the unpersuaded that his theory is important and worth pursuing. For the mark of a truly accomplished scholar is the ability to take an idea, particularly one like semiotics with its own language and relational concepts, and explain to others who are not versed in the literature what the theory means, much in the way Kant deconstructs his synthetic concepts and distinguishes them from analytic concepts so that even a lay reader can understand what the language of his theory amounts to in terms that mean something to them. Of course, I don't mean to hold Professor Malloy to the standard of Kant. But at least to this reader, he doesn't even try, leaving his truths to be self-evident.

He is equally frustrating when he gets to policy. He discusses, for example, how his theory could result in the creation of greater wealth than does "traditional" law and economics, but evades a very important question of the distribution of that greater wealth or even why we would necessarily care about greater wealth. He corrects for this a dozen pages or so later when he asserts that equality of opportunity generates more wealth than the absence of such equality, but he doesn't explain what he means by equality of opportunity or even how it creates more wealth (other than that it permits more people to be creative in market transactions, an assertion which is neither obviously true nor means anything by itself). And in making this argument, he subtly shifts from the microeconomics with which "traditional" law and economics is concerned to a more macro-economic perspective, without seeming to acknowledge the importance of the shift.

It is surprising that in a scholarly book such as this, there is no bibliography, and all the more so for the relative paucity of footnotes. It would help me to know whether Malloy has read the work of Amartya Sen, Joseph Schumpeter, Joseph Stiglitz, Ken Binmore, Robert Frank, or the law and economics literatures that have been emerging that I have mentioned above. It would help me to know whether he's read any of the rich psychological or sociological literature that in recent years has leveled powerful critiques at traditional neo-classicism. It would help me to know this, because then I could surmise, even if he doesn't tell us, that Professor Malloy finds all of these approaches inferior to his. But he limits his examination of the literature pretty much to the Posnerian view of things. So even if I understood his theory better than I do, I don't know why he thinks it's superior.

It seems terribly unfair in a review of this length to criticize Professor Malloy so harshly, since I lack the space to go through chapter and verse of the book's failings. Rather than attempt to pick out a few examples, and thus leave myself open to the charge of being overly selective, I stand simply by the broad critique I have leveled thus far.   I am sorry to do so. But the danger of a book like this is not only that it will persuade nobody who is not already persuaded, but that it provides ammunition to the small remaining rear-guard of the traditional school to charge the critics of law and economics with the inability to offer useful alternative approaches. To the extent that Professor Malloy is arguing for a broader and more contextual approach to economics, he is one of many voices in the larger academy. To the extent that he is attempting to offer a theory in which to express that approach, I regret to say that he has failed.

Lawrence E. Mitchell is John Theodore Fey Research Professor of Law at The George Washington University. He has written extensively on matters of corporate law and jurisprudence.  His books include Stacked Deck: A Story of Selfishness in America (Temple University Press, 1998) and the forthcoming Corporate Irresponsibility: America's Newest Export (Yale University Press, 2001). Professor Mitchell contributed a chapter to Robin Paul Malloy's earlier book, Law and Economics: New and Critical Perspectives (Peter Lang Publishing, 1995).

Editors' Note: In a previous issue of Books-on-Law, Professor Mitchell reviewed Liberalism and the Limits of Justice (Cambridge University Press, 2^nd ed., 1998) by Michael J. Sandel.

Mitchell's struggle over a conceptual understanding, at even an elementary level, is illustrative of one of the key points of the book. He brings a biased interpretive expectation to his reading of the text, and is frustrated by the fact that the text violates his preconceived notion of what should be included in the book. This interpretive difficulty arises in a variety of ways in both law and in market theory. My book develops a new way of understanding these interpretive problems, and demonstrates that such problems are of substantive import. In other words, linguistics, semiotics, and rhetoric are not simply interesting fields of study; they have substantive implications for law and for market theory.

I explain the way in which markets operate as places of meaning and value formation.  I explain what I mean by law and market economy, and illustrate the problem of mediating among people in alternative interpretive communities. I show how basic concepts in market theory operate as interpretive devices in framing and influencing law and legal institutions. Ultimately, I argue for the development of new courses and new perspectives on the relationship between law and market theory – courses that I would identify as law and market economy.

In my book, I clearly advise the reader that my efforts involve a desire to revisit some of the initial ideas developed in Deirdre N. McCloskey's classic book, The Rhetoric of Economics (University of Wisconsin Press, 1998). I also claim that I am interested in and sympathetic to building on ideas addressed in the book Beyond Economic Man: Feminist Theory and Economics (University of Chicago Press, 1993), edited by Marianne A. Ferber and Julie A. Nelson. All this – plus the content of the favorable endorsing quotes on the book cover, the wording of the Preface and opening chapter, the clear identification of my sources and points of reference – should help the reader to understand the nature of my undertaking. Somehow, it seems to have been missed by the reviewer.

In presenting such a work I kept Umberto Eco's conception of the intended reader in mind. My book is written for a particular audience, and it has been praised highly by people doing semiotic interpretation theory and humanistic aspects of law and society, and by legal theorists critical of the various and assorted undertakings of the American Law and Economics Association.  My examples and illustrations point to common elements in basic law and economics, so that the complex interpretive devices can be shown as relevant and as useful to the intended reader. I am directing my attention to a new audience.  I am not writing to persuade Mitchell or others like him. My only suggestion to Mitchell would be that there is space for developing some entirely new approaches to the way in which we think about the relationship between law and market theory. One of the things that should be in this space is a theory that attempts to use methods of interpretation to broaden our understanding of exchange. This is important because lawyers do interpretation theory everyday. This is at the heart of the book; yet, Mitchell says nothing about it.

For the record, I am familiar with the various sources that Mitchell cites in his musings about my book. But, does he really think that his references make that much difference to my arguments about the frame shifting function of interpretation theory? I am also aware of the various new and interesting ideas at play in the "new" law and economics. But, does Mitchell really think that the "new" work strays very far from the conceptual foundations of mainstream law and economics?  These new approaches are primarily variations and refinements on a given theme. Law and market economy takes an interpretive turn that asks us to begin to develop a better understanding of the meanings and values of exchange.

I also want to add that my editors at Cambridge did an excellent job; and I thank them for agreeing, at the proposal stage, to support a book that addresses an important area of scholarship without being tied to "the usual cast of characters."

In conclusion, Professor Mitchell, though his heart may be in the right place, simply fails to say anything of interest or relevance about the book. The most troubling thing, however, is that he doesn't even understand that the book is about something other than an economic analysis of law.

Robin Paul Malloy is Professor of Law and Economics at the College of Law and Maxwell School of Syracuse University. He is Director of the College's Program in Law and Market Economy. Malloy has published eight books and over thirty articles and essays on law and market theory, law and semiotics, and on real estate transactions and development.

From Contract to Covenant: Beyond the Law and Economics of the Family
Margaret F. Brinig
Cambridge, MA: Harvard University Press
Cloth: $47.50
Pp. 382
ISBN: 0674002164

There is widespread agreement today that the American family is in a state of crisis. Divorce rates are up, marriage rates are low, births to unmarried parents have escalated. Scholars and public policy makers ponder new ways to examine and assess such family changes in an attempt to determine possible responses to them. For more than a decade, Professor Margaret F. Brinig has explored these family problems, producing a body of scholarship – often controversial – that uses a law and economics approach to analyzing them. In From Contract to Covenant: Beyond the Law and Economics of the Family, Brinig attempts a life-span analysis of family law – from marriage and children through divorce and reorganized families to the family relations of elderly parents and their children – describing the insights that can be discerned from economics for these family forms.

I was anxious to read the book. Although my specialty is family law, I have no background in economics, and I must admit that I have been skeptical of the value of the law and economics approach to the problems of the family. I can't say that I am a convert to law and economics after reading the book. But in From Contract to Covenant, I did find a way of thinking about the family and family relations that is completely different from standard models and that recognizes the enduring nature of family relationships.

The Analysis

First, some comments on what I regard as the major contribution of the book – Brining's view of the family as a relationship bound by covenant, i.e., not by law, but by the mutual commitment of family members to each other, a commitment that results in a lifelong interconnectedness.

The book is a thoughtful and innovative analysis of the various stages of family development; it is also a wise recognition of the depth and strength of family relationships. Using metaphors of market, covenant, and franchise, Brinig divides family relations into three stages.  The first is family formation: marriage and becoming parents. At this stage, Brinig suggests that the most useful analyses are contract and market.  The second stage – the ongoing family relationships of husband and wife, parent and child – involve different commitments; here, Brinig introduces the idea of covenant to describe the unconditional nature of marriage and parenting. She proposes that the economic structure of a firm is the closest analogy to this family form. Finally, as the family breaks up – the parents divorce, the children are emancipated – Brinig turns to the concept of franchise to describe the relationships that are present after the legal bonds are dissolved by divorce or emancipation.

Family Formation. The chapters on marrying and becoming parents contain traditional law and economics analysis of markets and bargains. But even here Brinig points up the limitations of applying law and economics to family law. Decisions to marry, she notes, raise problems for rational choice theorists because people who marry are usually "in love" – a noneconomic, nonquantifiable state – and tend to ignore some obvious problems. As an example of this, she cites a survey of marriage license applicants finding that, although statistically a high percentage of their marriages would end in divorce, the applicants did not consider that their marriage would end in that fashion.

The Family Covenant. The principal contribution of this book is the use of the concept of covenant as a framework for the relationship of husband and wife, and of parent and child, in an ongoing marriage. Brinig first explored the family covenant idea in a review of Milton Regan's Family Law and the Pursuit of Intimacy (New York Univ. Press, 1993), published in 79 Cornell L. Rev. 1573 (1994). The term covenant conveys the unconditional nature of marriage and parenting. A covenant, Brinig explains, is an agreement enforced not by law so much as by individuals and their social organization. Family covenants draw much of their power from the mutual commitment of family members to each other and to the family. This perspective on family, not as a collection of rights and duties but of commitment and permanence, is important because it views the family as a positive cooperative relationship, and not an individualistic one.

Here, Brinig fulfills her objective (stated in her introduction) to use her analysis to describe characteristics of intact families, and to discuss that area of family law that is not explored because students of the family are preoccupied with divorce.

The modern family, Brinig notes, can no longer rely on the traditional model of specialization that Gary Becker, the Nobel Prize-winning economist, introduced in his book, A Treatise on the Family (Harvard University Press, 1991). In that model, husband and wife specialize at those functions – household production or market production – that he or she does best. In terms of efficiency, the model posits that the wife should be the household producer, as she commands a lower wage in the market – both because, as the child bearer, she has commitments to child care and because she is the object of discrimination. In turn, it is more efficient for the husband to be the market producer.

Brinig finds several problems with this model. For one thing, it places great psychic stress on both husband and wife; on the wife because of her inability to pursue career interests, and on the husband because the model denies him fulfilling family relationships.   Moreover, in the modern world, one income producer is not sufficient; it often requires the work of both partners to support a family.

Therefore, the modern partnership family, where both husband and wife share labor force participation and household tasks, may be the more efficient. The concept of covenant – with its goals of giving and commitment – is important to this family sharing. The work Brinig has done suggests that couples who are not keeping track of who is doing what around the house are happier.

As with husband and wife, the covenant is an analogy for parent-child relations, which cannot ever completely dissolve. Even if the parents divorce, parents are still parents. And parents are still parents when their children are grown.

The Family Franchise. This is the term Brinig uses to encompass the relationships that are still present even after legal bonds are broken by divorce, emancipation, or adoption. She sees the post-divorce family in much the same way as Constance Ahrons does in The Good Divorce (HarperCollins, 1995), not as separate entities but as a reorganized family of second husbands and wives, stepchildren and stepparents.

There is another model of the post-dissolution family, which Brinig calls the sovereign nation model, that views the marriage relation as contractual and sees divorce as a clean break. But, Brinig notes, there is no such thing as a clean break where there is a marriage of sufficient length to produce rights in property or earning capacity, or where there are children whose custody and support must be considered.  I have made this same observation in an essay entitled "Whatever Happened to Divorce?" (2000 Wisconsin L. Rev. 637), noting that, if anything, public policy has moved to more, not fewer, post-divorce ties.

Brinig's concept of franchise is even more apt in another family relationship area she explores – that of parent and adult child. Parents have invested in their children, providing them with education and skills to be successful, expecting them to uphold the family reputation and, maybe, to provide companionship and support in their old age.

Finally, in her treatment of the ties of elderly parents to their adult children, Brinig observes that those ties have become much looser in modern America. One reason for this, as we all know, is that families now live continents apart. Brinig acknowledges this, but also points to the fact that we now have social security and a tax system that encourages private pensions and Medicare.  She suggests that, perhaps if we did not have those public supports for the elderly, "there would probably be private provision for old people – arguably . . . a good thing for both the elderly and minor children." (218)

I am bothered by this theory. I was a devoted daughter to my mother, particularly in the 17 years after my father died. My mother was an integral part of our household, giving my children wonderful memories of their grandmother, but I was very grateful that she had a state pension and social security that, along with my financial help, enabled her to remain in her own home.

Moreover, I do not think that the fact that my husband and I are very well provided for in our old age has had any effect on our tightly-knit, Italian-American family.  Our children call almost daily and visit often. They would be stunned to think that someone thought they do this in the expectation of inheriting money. Brinig is right: family ties are very enduring.

Family Values

My second reaction to From Contract to Covenant relates to the role that our values – and, here, Margaret Brinig's values – play in how we analyze problems. I think we would all agree that there is no way of approaching a subject like the family and family law in a value-neutral way. But, for the uninitiated like myself, the use of a somewhat scientific approach to analyzing family relationships suggested that, maybe, it would result in some type of neutral approach. I was wrong.

Margaret Brinig does not at any point claim that her analysis is not value-driven. On the contrary, she would say that an attempt to be neutral is a mistake. She argues that the law we choose affects the way we order our family relations.  An example of this is no fault divorce.   Brinig attributes the high rate of divorce today to the fact that all American states make some provision for no fault divorce. No fault divorce, she argues, not only allows people to divorce easily, but easy divorce affects marriage. Partners do not invest in marriage because they do not regard it as permanent. In her view, making marriage more difficult to get out of empowers the parties to make a greater investment in the marriage.

In the chapter on "The Marriage Market," the section on "Cohabitation" discusses the variety of reasons for which that living arrangement might be a substitute for marriage: the desire to know more about a prospective spouse; the desire to avoid a commitment or, in the case of some elderly couples, to avoid losing the monetary benefits of being widowed; the inability to marry because one partner is already married or is of the same sex. "Looking at this phenomenon through the lens of economics," notes Brinig, "reveals several interesting patterns. The most obvious is that states want to encourage marriage, since only a very secure relationship leads to the kind of costly specific investment, particularly in children, that states consider desirable." (28)

Brinig then suggests that the state can promote marriage either by increasing the costs of nonmarital cohabitation – it can, for example, be penalized – or, in the alternative, the state can make marriage more attractive. One way in which marriage can be made more attractive is the authorization of "covenant marriage" as an alternative to standard marriage. Covenant marriage differs in the way in which it can be ended: no fault divorce applies in the case of standard marriage, while covenant marriages requires proof of fault or a long period of separation to obtain a divorce. To Brinig, the advantage of covenant marriage over standard marriage is that it encourages the partners to invest in the marriage, given their assurance that it is more permanent.   This is a theme running throughout the book: Brinig's conviction that the current high rate of divorce is attributable to the availability of no fault divorce.

Brinig's views on unwed parenting are also clear. In the chapter on "Becoming Parents," her discussion of the single mother focuses on two issues:

(1) The fact that "marriage, as opposed to some alternative family arrangement, is necessary for 'first best' child rearing." (57) Here, Brinig cites studies that point to the unfortunate effects of unwed parenting, such as poverty, adjustment problems for children, and intergenerational marital instability.

(2) "From a basic economic perspective it might seem uncontroversial to suggest that public assistance results in increased births. Subsidize something and theoretically you will always get more of it." (60)

Not surprisingly, Brinig recommends the welfare system as a definite candidate for reform.

The Organization

Finally, a few comments on the unorthodox organization of the book – something that I found, at times, to be confusing and, at other times, delightful. Each chapter in the book consists of several short essays on discrete subjects; the subjects all relate in general to the topic of the chapter. For example, the chapter on "The Marriage Market" contains, in the following order, a discussion of the changing courtship practices in the United States, a section entitled "Search and Fraud in the Marriage Market" that deals with the law of annulment, and sections on cohabitation as a substitute for marriage, contracts between same sex partners, second marriages, antenuptial contracts, and actions for breach of contract to marry.

The discrete essay approach has both its advantages and disadvantages. One disadvantage is that there is not a coherent unifying framework, resulting in a fragmented treatment of family law issues.  One advantage is that the approach allows the author the flexibility to discuss in detail some issues that are either minor in terms of the principal focus of the chapter or peripheral to it.   Here are some examples.

In the chapter on marriage, there is an essay entitled "Can Promises to Marry be Enforced?"   It treats the decline and abolition of actions for breach of contract – not usually a major topic of interest in mainstream modern family law. Brinig discusses research she has done establishing that, in response to the decline in the action for breach of promise to marry, women began to rely on the receipt of a diamond engagement ring to protect them from the economic loss associated with the cancellation of a marriage. She then suggests that the decrease in the demand for engagement rings since the 1960's can be associated with the change in social mores.  Socially acceptable sexual activity is no longer completely confined to marriage, so the need for a bonding device before consent to intercourse is greatly diminished. Furthermore, in an era of working women, the cost to a woman of a broken engagement is no longer as significant as it once was.

In the chapter on law reform, where the focus is on the family law reforms that would promote the covenant approach to marriage and family relations, there are two little essays that I found fascinating, but peripheral to the main focus. The first discusses the effect of our federal system on family law reform, using the case of migratory divorce as an example.   Brinig suggests that the ability of married people to shop for a favorable forum in which to obtain an easy divorce – a divorce that, because of the law of our federal system, was then valid in all American jurisdictions – was one of the reasons that led to the widespread adoption of no fault divorce in the United States. She also speculates on the future of same sex marriage. The fact that it is now recognized in at least one state raises the possibility that only a few states could make same sex marriage valid for the greater majority without any legislative or judicial action in those states. In this regard, she speculates on the possibility of an attack on the constitutionality of the federal Defense of Marriage Act, which allows states to decide not to recognize same-sex marriage and thus circumvent the Full Faith and Credit clause.

The second essay in the chapter on law reform begins with the observation that "family law reform . . . always seems to benefit attorneys." (213) The example that Brinig gives deals with changes in the federal tax laws for divorce in 1984 and 1986.   These changes were sought by tax attorneys and the Family Law Section of the American Bar Association, and Brinig claims that "there were no obvious gains by divorcing husbands or wives." (214) The beneficiaries were lawyers because the revised tax provisions were complex and probably beyond the competence of non-lawyer mediators who compete with lawyers for no fault divorce work.

Another example that Brinig gives of the phenomenon of lawyers profiting from family law reform is the change from legislative to judicial divorce in the 19^th Century. Early in the history of American divorce law, an aggrieved party, usually the wife, could obtain only a legal separation in court and had to obtain a divorce through the legislature. Most legal historians have reported that the legislative divorce was problematic: it was prohibitively expensive for poor, often deserted, wives and it was subject to manipulation and fraud. The change to judicial divorce was, therefore, considered a great reform.  In fact, the Constitution of the State of Wisconsin, adopted in 1848, prohibited the legislature from granting divorces. Brinig does not discuss the forces behind the movement, but she notes that the change to judicial divorce benefited lawyers in two ways: first, by lowering transaction costs, more people were able to seek a divorce; and second, by changing from the simple divorce bill to the more complicated need for a summons, complaint, and judicial proof, a new form of lucrative lawyer practice emerged.

Incidentally, this theory is contradicted by Brinig's observation at an unrelated point that, prior to no-fault divorce, the practice of family law was regarded as "a rather disreputable one (only slightly above criminal defense work, and perhaps below 'ambulance chasing' and loan collection in terms of public perception." (214) And, in the vast majority of cases, not very lucrative.

Conclusion

From Contract to Covenant is a thoughtful analysis of family relationships using an innovative economics-based framework that stresses the enduring nature of these relationships. The book is also an exposition of a conservative view of the law and family relations. But it is not necessary to agree with Brinig's conservative agenda – I, for example, do not believe that the current high divorce rate is the result of the adoption of no fault divorce – to appreciate the value of her analysis.

Marygold S. Melli is Voss-Bascom Professor of Law Emerita at the Law School of the University of Wisconsin-Madison, where her specialty was family law.

Editors' Note: For other Books-on-Law reviews related to family law, see Michael Grossberg's review of The Family, Civil Society, and the State (Rowman & Littlefield, 1999), edited by Christopher Wolfe (and Wolfe's reply), and Judge Richard Ross's review of Beneath The Fault Line: The Popular and Legal Culture of Divorce in Twentieth-Century America (University Press of Virginia, 1997) by J. Herbie DiFonzo. See also Katharine K. Baker's review of Unbending Gender: Why Family and Work Conflict and What to Do about It (Oxford University Press, 1999) by Joan Williams.

Law and Social Norms
Eric A. Posner
Cambridge MA: Harvard University Press, 2000
Cloth: $39.95
Pp. 260
ISBN: 0674001567

In Law and Social Norms, Professor Eric Posner tells us why we behave.   As you might guess, it's not because we are good. It's because we are greedy. But Posner's explanation of exactly how we serve self-interest in complying with non-legal social norms is unusual in the literature of Law & Economics, because it rests on an account of meaning as well as behavior. Drawing on game theory and psychological research, Law and Social Norms elaborates the traditional model of economic analysis in order to enable legal theory to take account of symbolic behavior. Posner makes a powerful argument that legal policy must take into account both the existing environment of social norms and the hermeneutic effects of legal norms on that environment.

Posner sets out to answer one of the most persistent criticisms of the economic analysis of human behavior, one often associated with the Law & Society movement. Contrary to economic assumptions, people often seem to act against their self-interest when they conform to socially determined standards of conduct. Social norms (by definition) dictate most of our behavior, including modes of dress, speech, and manners; workplace rituals; religious observance; political engagement; and family relationships. Norms supply the more abstract, traditional virtues such as trustworthiness, loyalty, courage, and honesty, as well as producing less honorable behaviors such as racism, sexism, and religious intolerance.

To conform to a social norm that regulates behavior is often to incur the cost of frustrated desire. Yet, people often choose to obey even when they gain no obvious benefit from performance and would incur no obvious cost by disobedience. In contrast to rational, self-interested market behavior, voluntary obedience to behavioral norms casts doubt on the descriptive accuracy and comprehensiveness of economic analysis.

The literature of Law & Economics usually responds that costly obedience to social norms is consistent with economic assumptions. In some cases, disobedience is more costly than obedience because financial or social sanctions will follow, such as loss of business opportunities or personal relationships. (But it is unclear why society would incur the cost of punishing the deviation.) In the more problematic cases in which no punishment would be forthcoming, obedience must reveal the actor's preference or taste for whatever behavior the social standard happens to demand. Gift-giving satisfies a preference for altruism, patriotism a preference for flag-waving, marital fidelity a preference for the conjugal bed, and so on.  But this invocation of convenient preferences seems too pat an explanation, threatening to turn economic analysis into a tautology.

Posner takes the opposite tack. It is precisely because social norms conflict with preferences that they can serve the actor's self-interest in obtaining future wealth or utility. His thesis is that social norms are behavioral signals of trustworthiness, the function of which is to enhance the reputation of the actor and facilitate the formation of cooperative relationships. Obedience to social norms is thus a rational, self-interested effort to obtain the utility such relationships can provide.   Gift-giving, patriotism, racism, and other socially approved behaviors are all intended to identify desirable partners for mutually beneficial exchanges. (As the example of racism illustrates, Posner does not assume that social norms are per se desirable, and notes that they often have effects on third parties that outweigh any benefits they may provide.) If successful, Posner's theory would improve the accuracy of economic analysis and permit a more informed relationship between legal policy and social norms.

Not Your Parents' Norms

Law and Social Norms is not an inquiry into the substance or validity of the underlying values that give rise to social norms. Posner defines social norms as patterns of behavior, not as values. Social norms are "nonlegal mechanisms of cooperation."   More precisely, a social norm is any pattern of observable behavior that can serve as a signal and that is enforced by social sanctions, such as criticism, ostracism, or stigma. In Kantian terms, Posner views all social norms as instrumental rather than categorical, as ways of getting something done rather than as expressions of ultimate value. What the norm accomplishes is largely unrelated to its content. From this perspective, there is no difference in kind between the most deeply held moral principles and the most faddish styles of dress or recreation. The approach is norm-neutral.

Nor does Posner evaluate the content of specific social norms, although he occasionally speculates about their efficiency. Disavowing faith in the evolutionary value of conventional morality expressed by Posner père (Richard A. Posner, The Problematics of Moral and Legal Theory (Harvard Univ. Press, 1999)), Posner fils is agnostic about the net value of social norms: "Functionalism – the view that social practices and norms are efficient or adaptive in some way – is empirically false and methodologically sterile." (172) The focus of the book is on the genesis and function of norms as behavioral patterns with informational effects. His model of the mechanisms by which social norms do their work is as valid for a coven of witches as for a cartel of corporations.

Law and Social Norms generally rejects the preference hypothesis as well as the more commonplace claim that people conform to social norms out of a genuine felt desire or sense of obligation. Posner acknowledges that such motivations may exist, but he finds the preference hypothesis "methodologically sterile" because it does not lead to interesting or provable conclusions about exactly what norms exist, how they come into being, and why they manifest themselves exactly as they do. For example, the hypothesis that gift-giving reflects a preference for altruism is weak in answering such questions as why we do not give family members gifts of cash, or why we usually give gifts only on holidays, or why charitable gifts are rarely anonymous. Posner believes the signaling hypothesis does answer such questions.

The book contains three sections. Part I describes Posner's model of non-legal collective action and his theory of social norms as signaling devices, and answers some possible objections to it. Part II applies the signaling model to diverse legal fields: gifts and gratuitous promises; family law; criminal law; voting and symbolic behavior; racial discrimination and nationalism; and contract law and commercial behavior.  Part III explores the normative implications of the theory for efficiency and distributive justice; incommensurability and commodification; and autonomy, privacy, and community.

The Model & Its Parts

Posner posits that social norms represent solutions to the fundamental problem of human coordination, the problem of trust (which Posner conceives as an information problem). Aside from rare simultaneous exchanges, most human utility is derived from sequential exchanges that take place over time. Long-term commercial contracts, friendships, and more intimate relationships all require that parties must frequently trust each other not to take advantage. If each of the parties values the stream of future benefits of cooperation highly, then each will have an incentive to maintain the relationship and refrain from cheating.  But if a party does not value the future benefits highly, or if he is offered alternative opportunities that he values more highly than the future income from the relationship, then he will cheat, as shown by such diverse phenomena as opportunistic breach of contract, cowardly retreat on the battlefield, the one-night stand, or the confidence game.

The two attributes of the trustworthy trading partner are that she values future relational benefits highly (what Posner calls a low discount rate) and that she lacks attractive alternatives to continued cooperation.   Posner calls such partners "good types" and their opposites "bad types." Presumably, we all want to enter relationships with good types and avoid bad types. Assuming that each type knows her own characteristics, the information problem is that we cannot identify the good types easily. Past performances can create reputations for trustworthiness, but reputations are sometimes inaccurate and new traders have none.   The solution to the problem is to "signal" one's type to potential partners. But assuming that both good and bad types will claim to be good types, what kind of signals can we trust?

Enter the social norm, a pattern of behavior designed to segregate the good from the bad types, and achieve a "separating equilibrium." A clear example is the custom of giving gifts by potential business trading partners. If the value of the gift clearly exceeds the potential gain from an early defection by a cheater, then cheaters cannot afford to give the gift.  The giving of such a gift is an unmistakable sign of a good type with a low discount rate. The good type hopes to recoup the expense of the gift in the future benefits of the relationship, and so his trustworthiness is assured. Repeated gifts throughout the relationship help to maintain it against the threat of changes in discount rates or alternatives.

Social norms are signaling behavior that has occurred regularly enough to become a pattern of human interaction. As a signal, an effective social norm must have two characteristics.  First, it must be observable behavior.   (Posner's theory has nothing to say about an actor who obeys rules even when she is certain she is not being observed.)  Second, conformity to the norm must be costly to the actor. If he actually prefers to engage in the behavior, it has less value as a signal of his discount rate. (Thus, a relative who gives you a gift because she really loves you is failing to send the signal that she is a good type.) Any behavior meeting these two criteria can serve as a Posnerian social norm if it is recognized as such by enough people.

To be a successful signal, a social norm must actually separate good types from bad types. If bad types can mimic the signal by conforming to the norm, then the norm will result in a "pooling equilibrium" in which everyone conforms. The norm would then represent a pointless social cost (unless it produced valuable external effects). Norms will usually change at this point, as good types endeavor to find other ways of distinguishing themselves. "Norm entrepreneurs" (otherwise known as moral reformers) are agents of change who supply new social norms and persuade us of their meaning.

The product of all social norms is thus reputation, specifically reputation for trustworthiness in long term relationship. Although this behavior can have positive or negative welfare effects, all conformity to social norms regardless of their form is aimed solely at generating reputations.

Social Norms & Law

Posner offers a "methodology that enables a systematic analysis of the relationship between the law and non-legal mechanisms of cooperation." (5) The most interesting section of Law and Social Norms applies the model to a variety of normative behavior that is subject to legal regulation.  Posner's accounts here draw on research of various kinds. The analysis ranges from the scientific to the speculative and intuitive. The following examples are illustrative:

Criminal law and social stigma: Throughout history, criminal law enforcement has used methods such as branding and public punishment to identify persons convicted of crime, subjecting them to social stigma and shaming. Although social penalties increase deterrence at no cost to the state, stigmatizing is an imprecise method because its effects are highly variable. Stigmatizing criminals creates a criminal caste forced to rely on each other and whose only option is continued criminal activity. Indeed, shaming penalties can become a badge of honor within a deviant sub-community. The signaling model thus suggests that such effects be taken into account by lawmakers.

Welfare policy: Posner's analysis of stigma in welfare law departs from traditional Chicago School positions in several respects. On the assumption that people generally believe that the poor are less trustworthy than the non-poor, Posner shows how several features of contemporary welfare policy stigmatize the welfare recipient by making her more identifiable to the public and less able to obtain employment. Food stamps, public housing, and official indignities can have the perverse effect of causing the deserving (low discount rate) poor to forgo aid and leaving it to the undeserving, who incur fewer costs from stigmatization.   He also notes that two frequently attacked methods of wealth redistribution, the minimum wage and rent control, have the benefit of avoiding stigmatizing the recipient. These benefits may offset the costs of such laws.

Contractual marriage: Marriage is a value-generating relationship that the law seeks to support. Performance of the marriage obligation is enforced by social sanctions, such as shunning and gossip that penalize nonsupport, abuse, and infidelity. By limiting the particular forms that marriage can take to one set of "default" obligations, the law makes it easier for society to recognize and punish deviation. If the law permitted a wide range of marital contracts, society would not know what behavior was a breach of trust and could not effectively enforce marriage contracts. Conversely, if community enforcement of marriage is reduced (e.g., by permitting more contractual forms of marriage), legal enforcement through stricter divorce laws might be necessary. But this may have other undesirable consequences.

Same-sex marriages: Legal recognition of same-sex marriage would change the social meaning of marriage, which is seen as evidence of trustworthiness and conventional attitudes.  Society penalizes unmarried people in various ways, including unattractive stereotypes. Recognition of same-sex marriages or other liaisons would dilute the benefits of marriage enjoyed by existing married couples and may weaken the inference that married people are trustworthy.

Flag desecration laws: At times of national crisis, laws often compel symbolic behavior attesting to loyalty to the nation or to the Constitution. This behavior is undertaken largely for reputational purposes, signaling that one is trustworthy. Predicting the hermeneutic effect of flag desecration laws is difficult.  If law demands the behavior, it loses its meaning as a signal and may result in a pooling equilibrium in which both good and bad types issue the same signal for different reasons. The law "ambiguates" the meaning of the signal.

Commercial promises: Parties who enter long-term commercial relationships must trust each other to perform their commitments. Nonlegal or legal methods of enforcing promises are available to such parties.  They select the methods partly based on the informational problems inherent in litigation. Courts may provide superior sanctions, but have inferior information about what would constitute breach of trust. Courts use the consideration doctrine to distinguish purposefully legal from purposefully nonlegal promises in relationships that include a mix of obligations. Courts are more competent to determine whether a contract was created than they are to find its precise elements. This suggests a criticism of laws, such as Article 2 of the Uniform Commercial Code, that require courts to ascertain and apply immanent commercial norms to contract disputes.

In all of this, Posner's analysis is stymied at critical points by a problem familiar to Law & Economic analysis: the lack of critical empirical information about the social system. This includes ignorance of actual costs and benefits of cooperation, uncertainty about the existing "equilibrium" reflected in any particular social norm, and uncertainty about the behavioral effects of rule changes.   This "information problem" leads many of the analyses to speculation and guesswork, precluding specific policy recommendations. But Posner candidly acknowledges these limits, and does not try to push the model beyond its capabilities.

Nevertheless, this limitation means that his central injunction may prove impossible to heed. He urges lawmakers to take the existence of pre-existing social norms into account in designing law: "The desirability of a proposed legal rule . . . does not depend only on the existence of a collective action problem on the one hand, and competently operated legal institutions on the other hand.   It also depends on the way nonlegal systems always already address that collective action problem and the extent to which legal intervention would interfere with those nonlegal systems." (4)

Some Questions

Law and Social Norms falls prey to a paradox inherent in all accounts of social meaning that claim to be novel insights. Posner is telling us something we presumably do not already know about the informational content of our social behavior. Yet we are the senders and receivers of these messages. It seems unlikely that behavior could function as a signal when neither the sender nor the receiver is aware what it is or what it means. Given general ignorance of Posner's thesis, he seems to be forced to admit either that the meaning of signals is lost on most participants (which rebuts his thesis), or else that signals act subliminally or subconsciously. His model lacks any mechanism for subconsious transmission of information.

Social norms are ubiquitous and highly variable. In attempting to propound a Unified Theory of Everything, Posner must force some behavior into his model that doesn't seem to fit.  Thus, he argues that racists demonstrate their trustworthiness to each other by incurring the cost of shunning minorities; but lots of racists feel no cost at all in shunning, and it is a premise of their racism that there is no cost. Posner argues that rich people give each other expensive gifts because they have more to gain by defection from cooperative relationships; but isn't it equally plausible that they give expensive gifts simply because they can afford luxury and it is expected of them? Posner argues that patriotic displays in times of war are not really intended to support the national effort but to curry favor with others for future commercial relationships. This would be unsurprising to cynical war profiteers, but surely misdescribes sincere flag-wavers.

Posner concedes the possibility that much social norm behavior could result from preferences for such behavior or from beliefs that underlie ethical values, although he rejects such explanations on methodological grounds.   But the admission that people do, indeed, act out of emotions such as love and compassion, fear of looking different, or a felt sense of moral obligation means that strategic signaling accounts for only some, an indeterminate amount, of social norm behavior.   Although Posner is unquestionably correct that his model accounts for some social norms, it is difficult to draw useful conclusions without knowing how much normative behavior is produced for these reasons.

Sidney W. DeLong is an Associate Professor at Seattle University School of Law, where he teaches and publishes in the areas of Contracts and Commercial Law.

Freedom of Expression in the Supreme Court: The Defining Cases
Terry Eastland, editor
New York: Rowman & Littlefield / Ethics & Public Policy Center, 2000
Paper: $22.95
Pp. 396
ISBN: 0847697118

He is the author of "numerous publications on a wide variety of political and legal issues."   His several books include Religious Liberty in the Supreme Court: The Cases that Define the Debate over Church and State (Ethics & Public Policy Center, 1993) and Benchmarks: Great Constitutional Controversies in the Supreme Court (Ethics & Public Policy Center, 1995). He is a "renowned constitutional scholar."

By such measures, offered up by his publisher (Rowman & Littlefield in conjunction with the Ethics & Public Policy Center), Terry Eastland is ably suited to offer Freedom of Expression in the Supreme Court: The Defining Cases. Of course, the resume of this "renowned constitutional scholar" is far more noteworthy than credited – including in matters touching on the First Amendment.

Backdrop: Meese's Man

Ages ago, Terry Eastland was a speechwriter for Attorney General William French Smith; before that he was the editor of The Virginian-Pilot; and before that he studied philosophy and theology at Vanderbilt University and Oxford.  Along the way, he wrote for publications such as Commentary, National Review, Policy Review, and The Wall Street Journal. In 1979, he co-authored (with William J. Bennett) Counting By Race: Equality from the Founding Fathers to Bakke and Weber (Basic Books), a work highly critical of affirmative action. But Mr. Eastland is, perhaps, best remembered for his work with Attorney General Edwin Meese III.

Edwin Meese "will go down as one of the
great attorneys' general in our history."
        -- Terry Eastland
        The [Bergen] Record, Sept. 28, 1986

A March 5, 1988 Los Angeles Times editorial opened with the following observation: "From the beginning, President Reagan's Justice Department has approached law enforcement as if it were a moral crusade. Instead of simply prosecuting wrongdoers or investigating civil-rights violations, the department has unabashedly sought to remake law and to overturn Supreme Court precedents that it dislikes. Stare decisis is not in its vocabulary."

The editorial continued: "[L]ast week the Baltimore Sun unearthed a memo setting out the Justice Department's agenda for the last year of Reagan's term. Entitled 'A Strategy for the Remaining Months,' the memo is a remarkable document that urges department officials to 'polarize the debate' on key issues. 'We must not seek "consensus,"' the memo argues. 'We must confront.' What this means in practice, the memo says, is that the department should champion compulsory AIDS testing. Obscenity cases shouldn't bog down in First Amendment considerations. Capital punishment must be enforced for the sake of 'deterrence, retribution and incapacitation (i.e., decapitation),' the memo says."

Terry Eastland, then spokesman for the Regan/Meese Department of Justice, was "the memo's author."

Mr. Eastland (an "ideological provocateur") was one of the central figures in the campaign waged by Attorney General Edwin Meese to change the direction of American constitutionalism.  (James Lyons, "Justice's Spokesman-Scholar," Legal Times, November 10, 1986, p. 1) To that end, the campaign espoused reactionary brands of constitutional interpretation. (Such interpretation – once tagged as "radical positivism and relativism" – was attacked by some notable conservatives. (Harry Jaffa, Original Intent and the Framers of the Constitution (Regnery, 1994).)

In a July 1985 speech to the American Bar Association, Mr. Meese took constitutional exception to the tried-and-true doctrine that the Bill of Rights (including the First Amendment) applies to the states. That constitutional extremism was repudiated by, among others, Justice Stevens in an October 1985 speech to the Federal Bar Association. (See also Floyd Abrams, "Mr. Meese Caricatures the Constitution," NYT, July 25, 1986, p. 31.)

And, then, there was the infamous 1986 Meese Commission Report on Pornography. The report was denounced for "its glaring and persistent biases in gathering and evaluating evidence," and its recommendations were said to "pose a serious threat to free expression." (Carole S. Vance, "The Meese Commission on the Road," The Nation, Aug. 2/9, p. 76, at p. 77)   It was also soberly critiqued by, among others, Gordon Hawkins and Franklin E. Zimring in Pornography in a Free Society (Cambridge University Press, 1988).

There is more from the Meese years, including the First Amendment case of Meese v. Keene (1987). (See Stuart Taylor, "Court Backs 'Propaganda' Label for 3 Canadian Films," New York Times, April 29, 1987, sec. A, p. 28 (Eastland applauding decision).)

Fast-forward: Today, at 50 or thereabouts, Terry Eastland is the publisher of The American Spectator, among other things.

All of this should suffice to better inform the reader of Mr. Eastland's constitutional credentials. Even so, and in fairness, his book deserves to be judged on its merits, to which I now turn.

How "Splendid"?

In their blurbs on the back cover, University of Virginia political scientist Henry J. Abraham and UCLA Law Professor Eugene Volokh respectively describe Freedom of Expression in the Supreme Court as "a splendid compendium of the subject" and a "splendid textbook."

As a reference book, Freedom of Expression in the Supreme Court is the kind of work much to be welcomed by First Amendment enthusiasts. It is, after all, a handy compendium of sixty 20^th Century free speech cases (circa 1919-1998), accompanied by some interesting editorial commentaries and prefaced by an informative introduction, often tracking the thought of historian Leonard Levy, law professor David Rabban, and political scientist Walter Berns.

Notice that Mr. Eastland's First Amendment timeline begins in 1919. His case history starts with Schenck v. United States (1919).  Justice Holmes' opinion in Patterson v. Colorado (1907), in which he declined to apply the First Amendment to the states and likewise announced a restricted notion of free expression, is thus absent from Freedom of Expression in the Supreme Court. Granted, Mr. Eastland does discuss the case somewhat, though he thinks it too marginal to include in his text. Hence, among other things, Justice Harlan's notable dissent in Patterson goes unnoticed.

As a textbook, Freedom of Expression in the Supreme Court's chronological presentation – as contrasted with a coherent doctrinal presentation – makes for dizzying reading. Generally speaking, the offering moves in perplexing ways from clear and present danger cases to a prior restraint case/ then back to a clear and present danger case/ onto a picketing case/ then to a licensing case/ next to a "two-tier theory" case/ onto a fighting words case/ then back to more clear and present danger cases/ next to obscenity cases/ then to a right-to-reply case followed by a campaign financing case/ back to another obscenity case/ followed by a freedom of association and an intentional infliction of emotional distress case/ then to a flag burning case/ back to another libel case and to another commercial speech case/ onto an Internet case/ with various other kinds of cases sprinkled here and there, depending on their dates of decision.

The result: Chronology trumps clarity, at least the kind that comes with some comprehensive notion of how a doctrine evolves.

"Defining Cases"

How seriously are the Abraham-Volokh endorsements to be taken, in light of the fact that all of the following noted First Amendment cases are absent from Eastland's "splendid" selection of cases?

• Stanley v. Georgia (1969) [obscenity & privacy]
• Gertz v. Welch (1974) [libel law: public figures]
• Young v. American Mini Theatres (1976) [indecent speech: adult entertainment]
• Wooley v. Maynard (1977) [compelled expression]
• FCC v. Pacifica (1978) [indecent speech: broadcasting]
• Central Gas & Electric Corp. v. Public Service Comm. (1980) [commercial speech]

While such noted First Amendment cases are excluded, other relatively little known cases, such as Kunz v. New York (1951) (offered as a prior restraint case), are included. In differing ways, the six First Amendment cases listed above, among other omitted cases, are "important as a statement of legal doctrine." Some of those same cases "extend[ed] First Amendment protection to some new activity . . . ." Others are "notable as a reflection of, or a statement about, [the] times." By such criteria – Eastland's own criteria – the above-listed cases would seem to merit inclusion in his anthology of "defining cases." While they may receive passing commentary here or there, it is odd that those cases are not classified as "defining," given that other cases are so designated despite the fact that they sometimes do little more than parrot existing doctrines (e.g., the clear and present danger line of cases Eastland offers).

Mr. Eastland's work also omits, in significant measure, at least two categories of free speech cases. First, there is the public forum line of cases – the ones so thoughtfully explained and examined by the late Harry Kalven in his famous Supreme Court Review article. Apart from the lone inclusion of Edwards v. South Carolina (1963), the relevant "public forum" cases, as Kalven tagged them, are absent from Eastland's list of "defining cases." Hence, a wide spectrum of decisions – ranging from Hague v. C.I.O. (1939) to Brown v. Louisiana (1966) to Grayned v. City of Rockford (1972) to Police Dept. of City of Chicago v. Mosley (1972) to Perry Educ. Ass'n. v. Perry Local Educators' Ass'n. (1983) to International Society for Krishna Consciousness v. Lee (1992) and beyond – find no "defining" home in Freedom of Expression in the Supreme Court. (Eastland's inclusion of Airport Commissioners v. Jews for Jesus (1987) (a "defining case"?) is offered as an "overbreadth" case.) As the Court's spirited split in Hill v. Colorado (2000) (an abortion picketing case) makes all-too-clear, the public forum cases remain controversial and perplexing – in ways that the Court's nearly forty-year-old ruling in Edwards v. South Carolina cannot begin to resolve or explain.

Standing alone, many of the public forum cases I have mentioned (and yet others) may not amount to landmark cases. Still, their collective significance is highly important. Moreover, the public forum line of cases highlights the dichotomy between "pure speech" and "speech plus;" thus, they helped give rise to the "two-level" theory of the First Amendment. (See Kalven, "The Concept of the Public Forum: Cox v. Louisiana," 1965 Supreme Court Review 1, 13-21.)

It is ironic that students of the First Amendment learn relatively little about the public forum doctrine from Eastland's work, since it is that very doctrine that gives life to the kind of expression pictured on the cover of Freedom of Expression in the Supreme Court – i.e., a man addressing a crowd from atop a car in the street.

Second, Mr. Eastland gives scant attention to the electoral process line of First Amendment cases. Thus, of the some twenty such cases handed down by the Supreme Court between 1970 and 1998 (not to mention a 1999 case rendered after Eastland's cutoff date), Freedom of Expression in the Supreme Court offers but one electoral process First Amendment case: Buckley v. Valeo (1976). Consequently, Eastland's readers are pretty much kept in the dark about the important role the First Amendment has played in the area of law and politics.

But there is more here than a predictable quarrel over which cases should be in or out; there is a basic jurisprudential question. Is the law of the First Amendment synonymous simply with the law of "the defining cases," assuming such cases can be identified? For example, just how much First Amendment law does New York Times v. Sullivan (1964) teach if virtually divorced (as in Eastland's book) from Gertz v. Welch (1974), Time, Inc. v. Firestone (1976), and Dunn & Bradstreet Inc. v. Greenmoss Builders, Inc. (1985), among other libel cases? Or, can the immense doctrinal gap (nearly two-dozen Supreme Court cases) between Virginia Pharmacy Bd. v. Virginia Consumer Council (1976) and 44 Liquormart v. Rhode Island (1996) be ignored in the commercial speech context (as Eastland largely does) without forfeiting some important understanding of First Amendment law?

The point: Law is in the details, in what lawyers and judges do with the landmark cases. Learning something about that process is, I think, far more fruitful than stacking up sixty "defining cases" and saying "go figure." No doubt, Mr. Eastland senses this, as evidenced by two facts: First, he offers some nine "clear and present danger" kinds of cases, spanning the spectrum from Schenck v. United States (1919) to Brandenburg v. Ohio (1969) (likely to incite or provoke test). All right. But what about similar treatment for other lines of doctrinal development? Second, Eastland attempts to resolve the underinclusiveness issue by his introduction to the book, and by his prefatory comments to the cases he selected. Such comments are typically one-page long and consist largely of a statement of the facts and holding.   The problem is that huge doctrinal gaps and inconsistencies cannot be explained away in a few short prefatory asides. There is far too little commentary here to redeem Mr. Eastland's "top-60" enterprise.

Selected Commentaries

Twenty-three of the sixty cases in Mr. Eastland's work are followed by editorial "commentaries from leading newspapers and magazines of the day." (The Schenck (1919) case, and it alone, is followed by scholarly commentary – i.e., an excerpt from a 1919 Harvard Law Review article by Zechariah Chaffee, Jr.) The newspaper commentaries are offered to "convey some sense of how the country responded to the Court's decisions at the time they were handed down . . . ."

Mr. Eastland's approach is an innovative addition to the all-too-often unimaginative presentation of First Amendment case law. A number of the commentaries Eastland selected are quite interesting and revealing of the times. Of course, many are entirely predictable, as noted in a 1931 Minneapolis Tribune editorial Eastland quotes: "Newspapers for the most part are so sensitively jealous of their freedom of speech that they oppose to the utmost the slightest curtailment." The lion's share of the newspaper editorials tends, therefore, to side with the First Amendment – meaning vindication of press rights.

The pro-press/pro-First Amendment trend notwithstanding, Mr. Eastland also has included a number of editorials from the Evening Star – editorially speaking, sort of a Washington Times of its day. These editorials invariably protested a variety of vindications of First Amendment rights. But, as Eastland reveals in the editorials following Dennis v. United States (1951) (a Smith Act case involving prosecution of Communists), even the New York Times (distinguishing freedom from "a furtive conspiracy aligned with foreign Governments") and the Washington Post ("We do not think the decision belittles the great principle of freedom of speech") at times opposed robust applications of the First Amendment.

Mr. Eastland has done his readers a service by way of such newspaper editorial comments.  Their inclusion following 23 of the "defining cases" raises a question, however: What about the other 37 "defining cases"? Why were they left unaccompanied by such commentaries?

Hustler Magazine v. Falwell (1988), the case about Larry Flint's ribald parody of minister Jerry Falwell, is one of Eastland's "defining cases."  The majority opinion was a gallant reaffirmation of First Amendment principles, one tendered by Chief Justice Rehnquist, no less. But the case – unlike, say, 44 Liquormart v. Rhode Island (1996) – is followed by no editorial commentary. Why? Surely it was not because such commentary was unavailable.

Two days after the Hustler case was handed down, the New York Times editorialized: "But who is to decide what is too gross, too repugnant?  [Chief Justice Rehnquist] could find no principled way – certainly not in the subjective outrage of victims or of juries – to punish Hustler yet protect more defensible satire.   Better let some outrages go unchecked, he concluded, than imperil some enriching expression. The Larry Flynts are left to paddle in their own mire.  Out of it, the Rehnquist Court has fashioned a sparkling vindication of free speech." (NYT, Editorial, February 26, 1988, sec. A, p. 34) (See generally, Rodney Smolla, Jerry Falwell v. Larry Flynt: The First Amendment on Trial (St. Martin's Press,1988).)

Likewise, no commentary accompanies the opinion in Roberts v. United States Jaycees (1984) (held: First Amendment offers Jaycees no right to discriminate against women).  The opinion, not surprisingly, drew ample editorial comment at the time. For example, there was a George Will column in the Washington Post (July 8, 1984, D-7). In it, Mr. Will argued that "the court severely circumscribe[d] a principle in the process of enunciating it. A constitutional claim, grounded in the 'preferred freedoms' of the First Amendment, has been found inadequate in a challenge to a Minnesota statute. But the constitutional argument has life in it yet." That is, there was the matter of   "whether, say, the Boy Scouts can remain sexually exclusive." Similarly, a New York Times editorial (July 6, 1984, A-22) observed: The Court "has a long way to go to resolve the churning conflict between associational rights rooted in the First Amendment and laws prohibiting discrimination. . . . The Court's criteria do not seem to apply to organizations like the Boy Scouts, Girl Scouts or private sex-segregated clubs. [M]uch is [thus] left unresolved." (The editorial properly referred to Justice O'Connor's thoughtful concurring opinion, which is omitted from the Eastland book.) Such commentaries and others thus pointed to Boy Scouts of America v. Dale (2000), where a divided Court recently backed away from Roberts and ruled that the First Amendment allows the Boy Scouts to discriminate against gays.

Finally, Cohen v. California (1971) – the "Fuck the Draft" case – is depicted by Eastland as a ruling that "expanded the definition of protected speech by . . . embracing a theory of moral relativism that effectively prevented the states from maintaining standards of public discourse." (Eastland's "moral relativism" comment, by the way, echoes that found in Walter Berns' The First Amendment & the Future of American Democracy (Regnery, 1985), pp. 190-96.) But that is the only commentary Mr. Eastland's readers receive. As for the merits of his "moral relativism" charge (a charge also leveled against Justice Holmes in Eastland's introduction), there is a First Amendment answer. While there are many old and new variations, perhaps one of the best can be found in Justice Kennedy's thoughtful statement in United States v. Playboy Entertainment Group (2000):

"When a student first encounters our free speech jurisprudence, he or she might think it is influenced by the philosophy that one idea is as good as any other, and that in art and literature objective standards of style, taste, decorum, beauty, and esthetics are deemed by the Constitution to be inappropriate, indeed unattainable. Quite the opposite is true. The Constitution no more enforces a relativistic philosophy or moral nihilism than it does any other point of view. The Constitution exists precisely so that opinions and judgments, including esthetic and moral judgments about art and literature, can be formed, tested, and expressed. What the Constitution says is that these judgments are for the individual to make, not for the Government to decree, even with the mandate or approval of a majority." (emphasis added)

"Expertly Edited"?

The notion of "truth in advertising," I grant, takes on a special meaning when it comes to the hype found on the back covers of many books.  Still, one would hope that consumers could rely on certain representations (dare I say warranties?) having to do with matters such as product-reliability. Consider in this regard Professor Abraham's assurance that Freedom of Expression in the Supreme Court is "an expertly edited work."

Mr. Eastland devotes thirteen pages to 44 Liquormart v. Rhode Island (1996), a commercial speech case. Yet, he allows only seven pages for the landmark New York Times v. Sullivan (1964) case. What is especially troubling about his editing of the latter is his omission of a key component of the constitutional holding in Sullivan. I refer to the several paragraphs in the majority opinion dealing with the "of and concerning" requirement.

As Justice Brennan put it in language missing from Freedom of Expression in the Supreme Court: "We also think the evidence was constitutionally defective in another respect: it was incapable of supporting the jury's finding that the allegedly libelous statements were made 'of and concerning' respondent."

The Alabama Supreme Court assumed, as did the counsel for L. B. Sullivan, that for libel actions it was enough that the defendants, including the New York Times, referred to Sullivan in merely his official (though nameless) governmental capacity. To that argument, Justice Brennan forcefully responded:

"Raising as it does the possibility that a good-faith critic of government will be penalized for his criticism, the proposition relied on by the Alabama courts strikes at the very center of the constitutionally protected area of free expression. [fn. omitted] We hold that such a proposition may not constitutionally be utilized to establish that an otherwise impersonal attack on governmental operations was a libel of an official responsible for those operations. [Such evidence standing alone is] constitutionally insufficient to support a finding that the statements referred to respondent." (emphasis added)

And, then, there is Mr. Eastland's editing of 44 Liquormart v. Rhode Island (1996). In his introduction to the case, Eastland purports to restate the 4-prong test in Central Gas & Electric Corp. v. Public Service Commission of N.Y. (1980). According to Eastland, the Central Hudson test requires that a regulation "must directly advance a 'substantial' state interest, and there must not be a more limited regulation that would serve equally well."

Of course, that is an incomplete and somewhat awkward statement of the test. The problem is compounded by Eastland's omission of the Central Hudson case (which remains law). Not only is there the deletion of the full statement of the test as presented in Justice Stevens' 44 Liquormart decision for the Court, but there is also Eastland's curious editing of Justice O'Connor's concurrence. In relevant part, O'Connor's unedited concurrence states:

"Under that test, we first determine whether the speech at issue concerns lawful activity and is not misleading, and whether the asserted governmental interest is substantial. If both these conditions are met, we must decide whether the regulation 'directly advances the governmental interest asserted, and whether it is not more extensive than is necessary to serve that interest.' Central Hudson Gas & Elec. Corp. v. Public Serv. Comm'n. of N. Y., 447 U.S. 557, 566 (1980)."

The bold portion of the above quote out does not appear in the Eastland book, leaving readers to fathom ("go figure") for themselves the precise tenets of the Central Hudson test.

Whether such omissions are typical, I know not. Nonetheless, it seems to be more than harmless puffing to represent a work as "expertly edited" when it omits a key portion of a landmark opinion and likewise misstates a widely-used test from a noted First Amendment case.

Bizarre Bibliography

The bibliography to Freedom of Expression in the Supreme Court is not bad; it's bizarre. Strikingly absent from the list of "useful" commentaries on the First Amendment are the oft-cited books of Alexander Meiklejohn (Free Speech & Its Relation to Self-Government (1948)) and Thomas Emerson (The System of Freedom of Expression (1970)) and the seminal article of Harry Kalven ("The New York Times Case," 1964 Sup. Ct. Rev. 191). In their place are such largely unknown tracts as The First Amendment and the Future of American Democracy (Regnery Gateway, 1985) by Walter Berns and No Liberty for License (Spence Publishing,1997) by David Lowenthal. Curiously, Eastland's reference to such conservative interpretations of the First Amendment by Straussian political philosophers overlooks George Anastaplo's powerful tome, The Constitutionalist (Southern Methodist University Press, 1971).

There are still other oddities.For example, Eastland refers to (and draws heavily upon in his introduction) a law review article by David Rabban ("The First Amendment in Its Forgotten Years," 90 Yale Law Journal 514 (1981)). He does not, however, refer to Rabban's more extended and refined work, Free Speech in Its Forgotten Years (Cambridge University Press, 1997).   And, in the famous cases category, Eastland correctly refers to Make No Law (Knopf, 1991) by Anthony Lewis, but not to Minnesota Rag (Random House, 1981) by Fred Friendly or to Fighting Faiths (Viking, 1987) by Richard Pollenberg.

What Happened to the Lawyers?

Appended to Freedom of Expression in the Supreme Court is the customary listing of Supreme Court Justices (circa 1919-1999), replete with the names of the presidents who appointed them and the justices who replaced them, among other things. What is not appended, or really mentioned anywhere in the text, is anything about the lawyers who argued Eastland's sixty "defining cases." Mr. Eastland's omission is, unfortunately, an all-too-commonplace one.

As if there were some invisible force by which cases are brought to and crafted for the Supreme Court, the general practice is to dwell endlessly on those who decided the cases but not on those who argued them. Hence, who but the dead knows that Morris Ernst successfully argued Hague v. C.I.O. (1939), or that Ephraim London prevailed in Joseph Burstyn, Inc. v. Wilson (1952), or that Jack Greenberg was triumphant in Edwards v. South Carolina (1963), or that Herbert Wechlser successfully represented the New York Times in the Sullivan (1964) case, or that Alexander Bickel successfully argued the Pentagon Papers Case (1971), or that Mel Nimmer convincingly argued Cohen v. California (1971), or that Bruce Ennis won the day in Reno v. ACLU (1997), or even that Rust v. Sullivan (1991) pitted Solicitor General Kenneth Starr against Harvard Law Professor Laurence Tribe?

The Verdict

On balance, then, what should one make of Freedom of Expression in the Supreme Court?   Well, it is the kind of book that could easily find a place in a First Amendment library. A versatile reference book, indeed – a book not without its moments. Nonetheless, it is a conceptually and sometimes editorially flawed work. If only more time and thought had gone into Freedom of Expression in the Supreme Court, it might have been a far more useful addition to the literature.

Metaphorically speaking, the First Amendment is the grand highway of American justice.  It leads to places known and unknown, perilous and perfect, odious and harmonious, even "clear and present." A long stretch with all sorts of twists.   For the faithful, it is a shining highway to heaven; for the faithless, it is a dark highway to nowhere. And for some, it beckons uninhibited journey . . .

Ronald Collins is the co-editor of Books-on-Law and the co-author (with David Skover) of The Death of Discourse (Westview, 1996) and of Comedy on Trial: The Free Speech Struggles of Lenny Bruce (forthcoming). He directs a First-Amendment project at the Center for Science in the Public Interest.  Last year he co-authored a First Amendment amicus brief (co-signed by the Society of Professional Journalists, the Association of American Publishers, and the Booksellers Foundation for Free Expression) in Texas Beef Group, et al v. Oprah Winfrey (5^th Cir., 2000). Currently, he is working (with Robert Corn-Revere) on a First Amendment appellate brief in the case of Huminski v. City of Rutland.

Editors' Note: For other Books-on-Law reviews concerning free speech, see Bruce Johnson's review of First Amendment, First Principles (2000) by John Wirenius; C. Thomas Dienes' review of Don't Shoot the Messenger: How Our Growing Hatred of the Media Threatens Free Speech for All of Us (1999) by Bruce Sanford; and Chris Finan's review of Bookleggers and Smuthounds: The Trade in Erotica, 1920-1940 (1999) by Jay Gertzman. See generally "Dissent Injustice and the Meanings of America: An Online Exchange with Steven Shiffrin" and our General Index of Reviews (under Free Speech).

JURIST would like to hear your reaction to our reviews:

© Ronald K.L. Collins and David Skover, 2000.