From Harvard Law professor Adrian Vermeule’s paper “Many-Minds Arguments in Legal Theory”:
B. Evolution and Hayek
In legal theory, evolutionary many-minds arguments are likely to focus either on Hayek or Burke, whose ideas on this subject overlap to some degree. I will focus on Hayek here, bringing Burke into the picture in the next section. In legal theory Hayek is often associated with two ideas: the division of knowledge23 (the epistemic analogue to Adam Smith’s division of labor) and an evolutionary argument for the virtues of the common law.24 The first emphasizes the dispersed and tacit character of knowledge in markets, an argument that Hayek famously used to win the debate over the feasibility of socialist calculation – whether a socialist planner could acquire and process sufficient information to make efficient centralized decisions. The second celebrates the common law by an extended analogy to markets, an analogy that is shaky at best, or so I will claim.
Hayek’s central claim about explicit markets was true, important and largely original, which led him to draw ever-more-extended analogies between markets and other social institutions, such as custom, morality and the common law. The analogies are evocative, which accounts for his enduring appeal. But Hayek was an unsystematic thinker who had a strong tendency to think that all things he counted as good – spontaneous order, the division of knowledge, markets, morality, custom and the common law – somehow all go together. They do not; Hayek’s analogies are flawed. Moreover, the version of his thought about the common law on which he seems to have finally settled — that the common law is socially beneficial as a system in competition with other systems, rather than efficient taken rule-by-rule – strips his defense of the common law of any implications for action within an ongoing common law system.
We must start with some distinctions to come to grips with Hayek. One distinction is between intentional mechanisms and invisible-hand mechanisms. In the Jury Theorem model of information aggregation, participants are trying to get a particular answer (whether the preference-independent right answer or the answer that would, when collectively adopted, maximize the satisfaction of their own preferences). The epistemic value of their collective effort – the many-minds argument – is the intended result of their actions, although the group competence may of course differ from individual competence by virtue of the miracle of aggregation.
In evolutionary many-minds arguments, by contrast, the epistemic value of many minds arises as a byproduct of actions taken for other reasons. Evolutionary many-minds arguments are thus a species of invisible-hand arguments,25 in which the epistemic competence of the group is “the result of human action, but not of human design.”26 Hayek’s core commitment is to spontaneous order, which means to invisible-hand mechanisms (although, as we will see, Hayek seems to have assumed that all invisible-hand mechanisms must be evolutionary, which is not the case). The Condorcetian mechanism is a model of aggregated intentions, not an invisible-hand mechanism, whereas Hayek thinks that the aggregation of information must occur through the action of the invisible hand. In this sense, there can be no Condorcetian interpretation of Hayek.
Within the category of invisible-hand mechanisms, there is a further distinction between aggregative and evolutionary mechanisms.27 Aggregative mechanisms are synchronic, explaining the emergence of an equilibrium at a given time; markets are an example. Evolutionary mechanisms are diachronic, explaining not the emergence but rather the maintenance and development of order over time. Thus, in biology, Darwin’s mechanism of natural selection requires variation, heritability and differential reproductive fitness of the genotype (expressed in the phenotype), but natural selection is entirely agnostic about the origin of the genotype, which may arise through random mutation. The point is to explain the survival, or not, of the genotype over time.
Hayek seems to have systematically conflated these two types of invisible-hand mechanisms. Starting from an epistemic version of Adam Smith, inspired by aggregative markets, he slid over into an evolutionary account of morality, custom and the common law, without realizing that the subject had changed. Intrinsically, the price-system strand in Hayek’s thinking has nothing to do with evolution. When Hayekians emphasize the dispersed character of information, and the ability of market-like mechanisms using explicit prices (such as so-called “prediction markets”) to aggregate that information, their arguments are synchronic rather than diachronic, and thus not evolutionary at all.
Separately, Hayek and the Hayekians have a commitment to evolutionary accounts of morality, custom and law but this commitment need not follow from the commitment to aggregative spontaneous orders. Spontaneous orders arising at a given time and extended over space, such as a market, are analytically different than spontaneous orders evolving over time and there is no simple analogy between the two. To be sure, Hayek also means to offer an evolutionary account of markets as adjusting efficiently, over time, to exogenous shocks arising from technical, economic and social change, and such an account is fully in the spirit of Austrian economics. But an account of that sort needs separate justification and cannot be derived from the basic Hayekian insight that the price system aggregates dispersed information, which is a strictly synchronic insight taken by itself. In Part II, I will examine an institutional analogue of these problems: the distinction, crucial for many-minds arguments in legal theory, between simultaneous and sequential decisionmaking.
Finally, of course, spontaneous orders of either variety may either be desirable or not.28 Hayekians acknowledge this in their cautious moments but seem to hold at least a presumption that spontaneous order is functionally beneficial. But there is no such presumption. In the case of aggregative invisible-hand mechanisms, the only requisite is that a patterned social structure arises as a result of human action but not through human design. Collectively suboptimal patterns fit this requirement just as well as collectively optimal ones. The Tragedy of the Commons is a spontaneous order, just a bad spontaneous order that produces a structured social pattern of depletion and waste.
What about evolutionary invisible-hand mechanisms? Here there is a strong temptation to invoke “the test of time” and to think that there is at least a presumption that evolved patterns must be functionally beneficial in some sense or another. However, within the domain of evolutionary invisible-hand mechanisms, biological evolution is very different than social or economic or cultural evolution. Here too Hayek seems to have overlooked a crucial distinction. The rate of change in the biological (“natural”) environment is, plausibly, much slower than in the social or economic or cultural environment. The slow pace of environmental change in biology makes it plausible to think that natural selection produces organisms well-adapted to their ecological niches. If the rate of change in social, economic and technological environments is high, however, then social evolution faces a shifting target: even if social structures constantly evolve towards efficiency, they may at any particular point remain very far from it.29 Once again the diachronic/synchronic distinction is crucial: the trend towards efficiency over time just does not support the very different assumption or presumption that observed institutions are beneficial at a given time.
Hayek and the common law. For legal theorists, the most obviously relevant of Hayek’s many analogies – morality is a spontaneous order like the market, custom is a spontaneous order like the market, and so on – is the Hayekian claim that the common law is a spontaneous order that incorporates dispersed information, and thus proves superior in some sense to the collectivist and intentional legislation that a Benthamite might favor.30 Here again, Hayek or at least the Hayekian tradition conflates several different things that must be disentangled. Moreover, the version of Hayek’s argument on which he finally settled is far narrower than usually assumed, and has few concrete implications.
The first step is to distinguish Hayek’s ideas from several near and not-so-near relatives. To begin with, we must distinguish (1) an intentionalist argument for the beneficial properties of the common law from (2) an invisible-hand argument for the beneficial properties of the common law. As the common law is a diachronic institution, extended over time, invisible-hand arguments in this domain are inevitably of the evolutionary variety. Moreover, we must distinguish (2a) evolutionary arguments at the level of particular common law rules from (2b) evolutionary arguments at the level of the whole society, including the legal system. It turns out that Hayek’s view is (2b), not (1) or (2a), and this blocks an easy application of Hayek’s views to the particular rules of the legal system.
Position (1) can be associated with Richard Posner,31 while (2a) later became a standard view in law-and-economics.32 Position (1) is simply the argument that judges aim(or, in a normative variant, should aim) to maximize social wealth. That position is distinctly un-Hayekian, because it sees judges as a cadre of centralized, even Benthamite decisionmakers who rely for the most part upon their first-order reason to develop efficient rules, subject only to the weak constraints that higher-level legal rules must be respected and that precedents might convey useful information. Thus true-blue Hayekians reject the Posnerian vision of efficiency-seeking judges (and are especially suspicious of the Supreme Court, which they see as a centralized rulemaker). These Hayekians see custom as a genuine and highly beneficial spontaneous order, but deny that the common law system, at least as it currently operates, actually incorporates custom.33 I return to these issues in Part II.
Position (2a) is the argument that the common law will, at the level of individual rules, evolve towards efficiency. The basic mechanism is that inefficient rules will impose deadweight losses and will thus be more likely to be challenged by litigants. The losers from the rule will be willing to pay more to overturn the rules than the gainers will be willing to spend to defend them (this follows from the assumption that the rule creates a deadweight loss) and so, over time, inefficient rules will tend to be eliminated.
Among the many implicit conditions that make this hypothesis fragile and somewhat lacking in generality, I will confine myself to one observation.34 Even if all the conditions required by this mechanism actually obtain, it only shows that the common law will evolve towards efficiency. By itself, it says nothing about the rate at which it will do so. The common law is path-dependent and hence sticky;35 the very existence of a mistaken precedent reduces the ability or incentive of later actors to challenge it, in part by making it costly to switch to a different legal rule. If statutes can innovate more rapidly than the common law when circumstances change, “the fundamental trade-off is between evolution towards efficiency when the social optimum does not change, and rapid legal innovation when it does. . . . This price [i.e. the cost of inefficient legislation] is only worth paying when social change in sufficiently intense.”36
The crucial question then involves the relative rates of change in the common law and in the surrounding economic and political environment. If the environment changes slowly, then it is plausible that the common law will evolve to become quite efficient; perhaps the common law of the 17th or 18th centuries was like this. If, however, the nonlegal environment changes rapidly, then the common law will constantly be facing a moving target, and may be much less efficient at any given time than, say, statutes intentionally designed by Benthamite legislators, or even statutes designed by ignorant or self-interested legislators. In a rapidly changing environment the common law may be constantly be evolving from 0% efficiency to 5% efficiency, over and over again, and then statutes need not be very efficient at all to be comparatively superior.
Hayek, however, does not offer evolutionary arguments at the level of individual rules in any event. Rather his arguments operate at the level of the whole society, including the legal system.37 The suggestion is that the common law and related liberal institutions, such as the market, enable the social coordination of individual expectations, and that a society with a legal and economic system that coordinates expectations in this way will outperform competitors.38
This argument is ill-defined. “Hayek is noticeably unspecific about the precise mechanism by which selection works. On different occasions, there are hints of different models of selection, based variously on imitation, inter-group migration, competition for resources, differential rates of reproduction, and wars of extermination; but none is developed in any detail.”39 On the other hand, there is a burgeoning literature on “legal origins” suggesting that nations whose legal system descend from the common law outperform nations with other types of legal system, on various measures of economic and political well-being, perhaps due to the security of economic expectations that a common law system provides.40 The broad thrust of Hayek’s view might have been correct even if he did not understand or failed to spell out the mechanisms that make it so.
I will focus on a different problem with the systemic version of Hayek’s argument for the superiority of the common law. Even if well-specified, convincing, and empirically validated, the argument has few implications for action within an ongoing common law system, for the following reasons. The systemic version of the Hayekian approach is extremely narrow: it underwrites no conclusions whatsoever about the efficiency or desirability of any particular common law rule. That the common law as a system is superior to (say) a civil law system, as measured by some criterion of fitness, just means that one package of institutions and rules is superior to another package of institutions and rules. Any individual rule in the superior system might be inferior to the alternative rule found in the other system.
Moreover, the general theory of second best41 implies that the superior system might actually contain more rules that are inefficient or unfit than the inferior system. If a given system departs from the optimum in one respect, as it inevitably will, then the theory of second best implies that other departures from the optimum are necessary. The system whose rules most closely approach the optimum, evaluating the efficiency of rules one by one, might well be the inferior system. It is even possible that all rules in a given system are suboptimal and indeed inferior to those of the competing system, but that because of the particular structure of their interaction, the given system is superior overall. Because systems compete as packages, an evolutionary account of the superiority of the common law pitched at the system level tells us little about particular questions within an ongoing common law order.
22 Edelman, supra note 18, at 339-48, provides several examples of arguments that equivocate between different interpretations of the Jury Theorem.
23 See Friedrich A. Hayek, The Use of Knowledge in Society, in INDIVIDUALISM AND ECONOMIC ORDER 77-91 (1945).
24 FRIEDRICH A. HAYEK, LAW, LEGISLATION AND LIBERTY, VOL. I: RULES AND ORDER (1973).
25 Edna Ullmann-Margalit, The Invisible Hand and the Cunning of Reason, 64 SOC. RES. 181, 186 (1997).
26 Paraphrasing ADAM FERGUSON, AN ESSAY ON THE HISTORY OF CIVIL SOCIETY 119 (Fania Oz-Salzberger ed., Cambridge Univ. Press 1996) (1767).
27 The next few paragraphs draw heavily upon Ullmann-Margalit, supra note 25.
28 Robert Sugden, Spontaneous Order, in THE NEW PALGRAVE DICTIONARY OF ECONOMICS AND THE LAW 485, 494 (Peter Newman ed., 1998).
29 See JON ELSTER, ULYSSES AND THE SIRENS 6-7 (rev. ed. 1984).
30 See generally Hayek, LAW, LEGISLATION AND LIBERTY VOL. I: RULES AND ORDER, supra note 24.
31 RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 321-27 (1973).
32 See Paul H. Rubin, Why is the Common Law Efficient?, 6 J. LEG. STUDIES 51 (1977); George L. Priest, The Common Law Process and the Selection of Efficient Rules, 6 J. LEG. STUDIES 65 (1977).
33 John Hasnas, Hayek, The Common Law, and Fluid Drive, 1 N.Y.U. J. L. & LIBERTY 79, 92-98 (2005); Todd J. Zywicki, The Rise and Fall of Efficiency in the Common Law: A Supply-Side Analysis, 97 NW. U.L. REV. 1551, 1628-29 (2003). However, Hayekians of this stripe think that the common law was efficient in an earlier period, when it competed with other law-supplying institutions, such as equity and church law. See Zywicki, supra, at 1620.
34 This paragraph and the next apply to law a point made by Elster, supra note 29. For a more extended treatment, see Vermeule, supra note 21.
35 See Oona Hathaway, Path Dependence in the Law: The Course and Pattern of Legal Change in a Common Law System, 86 IOWA L. REV. 601 (2001).
36 Giacomo A. M. Ponzetto & Patricio A. Fernandez, Case Law vs. Statute Law: An Evolutionary Comparison, J. LEGAL STUD. (forthcoming), at 26 (available at http://www.people.fas.harvard.edu/~ponzetto/PF.pdf).
37 See Gerald F. Gaus, Hayek on the Evolution of Society and Mind, in THE CAMBRIDGE COMPANION TO HAYEK 232, 238 (Edward Feser ed. 2006).
38 See Todd J. Zywicki & Anthony B. Sanders, Posner, Hayek, and the Economic Analysis of Law, 93 IOWA L. REV. 559 (2008).
39 Sugden, supra note 28, at 494.
40 Rafael La Porta, Florencio Lopez-de-Silanes & Andrei Shleifer, The Economic Consequences of Legal Origins, J. ECON. LIT. (forthcoming 2008), available at http://www.economics.harvard.edu/faculty/shleifer/files/jel.nov_combine.pdf.pdf; Paul G. Mahoney, The Common Law and Economic Growth: Hayek Might be Right, 30 J. LEGAL STUD. 503 (2001).
41 R.G. Lipsey & Kelvin Lancaster, The General Theory of Second Best, 24 REV. ECON. STUD. 11 (1956).