View source for Some Problems with Spontaneous Order

<!-- you can have any number of categories here -->
[[Category:Timothy Sandefur]]
[[Category:Spontaneous Order]]
[[Category:Friedrich von Hayek]]
<!-- 1 URL must be followed by >= 0 Other URL and Old URL and 1 End URL.-->
{{URL | url = http://www.independent.org/pdf/tir/tir_14_01_1_sandefur.pdf}}
<!-- {{Other URL | url = }} -->
<!-- {{Old URL | url = }} -->
{{End URL}}
{{DES | des = "It is not possible to distinguish in principle, either descriptively or normatively, between spontaneous and constructed orders. Instead, the difference depends entirely on the observer’s frame of reference."  A convincing explanation of how Hayek pulls a hand-waving fast one. | show=}}
<!-- insert wiki page text here -->
<!-- DPL has problems with categories that have a single quote in them.  Use these explicit workarounds. -->
<!-- otherwise, we would use {{Links}} and {{Quotes}}  -->
{{List|title=Some Problems with Spontaneous Order|links=true}}
{{Quotations|title=Some Problems with Spontaneous Order|quotes=true}}

{{Text |
F. A. Hayek’s famous observations about the nature of social order are a cornerstone of much conservative and libertarian thought. Hayek’s argu- ments are frequently cited not only as a useful description of how social orders evolve, but also to support normative positions about the relationship between the state and society, specifically the argument that the social order ought to change as the result of decentralized, or “bottom-up,” actions on the part of individuals—that is, “spontaneous order”—as opposed to the centralized, “top-down” planning implemented by bureaucracies, socialist central planners, or the post–New Deal administrative state, which yields “constructed order” or “constructivist rationalism.”
The problem is that the distinction between spontaneous and constructed orders, though useful descriptively, cannot bear much normative weight because the difference between spontaneous and constructed orders collapses on close examina- tion. Although Hayek’s observation that social orders emerge from an aggregate of individual choices is indisputable, it is not possible to distinguish in principle, either descriptively or normatively, between spontaneous and constructed orders. Instead, the difference depends entirely on the observer’s frame of reference. This difficulty explains why Hayek fails to provide a convincing account of the process of conscious social change or the elimination of unjust institutions. The distinction between spontaneous order and planned order may be useful in an explanatory model at a certain level of generality, but it cannot serve as a normative guide in a particular policy dispute (Ogus 1989, 393; Wax 2005).
Timothy Sandefur is a senior staff attorney with the Pacific Legal Foundation in Sacramento, California. The Independent Review, v. 14, n. 1, Summer 2009, ISSN 1086–1653, Copyright © 2009, pp. 5–25.
   5
6 F TIMOTHY SANDEFUR
The Distinction Between Spontaneous Order and
Constructivist Rationalism
Hayek’s most enduring contribution to political economy is probably his explanation of how social orders emerge through a decentralized process of choices by an indefi- nitely large number of actors: such orders are the product “of human action, but not of human design” (Hayek 1967); they evolve much as animals do. Genes evolve by a nonrandom selection process applied to random mutations, and the selection pres- sure is provided by the environment; human social institutions evolve through non- random selection among different courses of action, with selection pressure being provided by competition between individuals and groups, and the proposed actions in question being undertaken by individuals. Social institutions such as language are therefore not invented “by” anyone, but emerge out of each person’s choices and pursuits. They are grown rather than made. Because social institutions serve people’s subjectively perceived needs, the result is that they evolve so that those that obstruct human needs are discarded and those that serve these needs are retained (CL, 63).1
We should appreciate the audacity of this seemingly simple observation. At the time Hayek was formulating this theory, virtually every respectable intellectual had concluded that free markets ought to be eliminated in favor of an economy organized by allegedly rational central planners, experts who could direct resources to their most efficient uses. Opposing the ambitions of New Dealers, socialists, and other central planners was a small coterie of dissenters with very little intellectual ammuni- tion. Liberals had largely concluded that limiting government was not sufficient for providing the grounds of human flourishing and that planned economies could do better (Dewey 1935). Against their accusation that free markets waste resources by rewarding the trivial and ignoring the impoverished, conservatives of the period had not yet formulated an effective response, and as a consequence many liberals came to believe that the only decision left was between romantic nationalist traditionalism (represented in its most extreme form by the Nazis) and modernist centralized planning (whose most extreme form was communism) that would eliminate individ- ual choice and achieve an allegedly objective historical “progress.”
Hayek, however, appealed to the traditionally liberal concerns for individual choice and flourishing, but used them to advocate limited government and individual freedom. He formulated an explanation of economic choice that responded to the argument that markets were inherently wasteful and irrational, showing that, on the contrary, social and economic institutions can originate, change, and disappear with- out the intervention of an omniscient planner. He was therefore among the pioneers
1. For ease of reference, I have used the following abbreviations for four of Hayek’s works most frequently cited in the article: CL = The Constitution of Liberty (1960); R&O = Law, Legislation, and Liberty, vol. 1: Rules and Order (1973); MSJ = Law, Legislation, and Liberty, vol. 2: The Mirage of Social Justice (1978); PO = Law, Legislation, and Liberty, vol. 3: The Political Order of a Free People (1979).
  THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 7
 in modernizing classical liberalism, and his contribution was a major accomplishment (Doherty 2007, 98–111).
On top of his observations about the nature of social order, Hayek built a normative argument: not only did social institutions evolve without planning, but they often evolved so that their utility could not be assessed in isolation from a particular transaction in a particular case. Certain traditional practices might have nonobvious utility, which a planner would overlook and might damage beyond repair when, regarding it as irrational, he took steps to eliminate it. The unanticipated consequences might then be much worse than the supposed irrationality of the now destroyed institution. Because a central planner cannot possibly have the knowledge necessary to evaluate an evolved social institution fully or to anticipate the conse- quences of altering it, decentralized spontaneous order is preferable to centralized bureaucratic planning.
In the three-volume work Law, Legislation, and Liberty (1973–79), Hayek elaborates on the difference between planned and grown orders. He attributes the central planners’ ambitions, which he calls “rationalist constructivism,” to classical liberalism’s rationalist tradition, tracing it ultimately to John Locke, and he links the idea of “spontaneous order” to Edmund Burke’s political philosophy. Rationalism sought to redesign social institutions (or whole societies) on the basis of exogenous principles of justice, principles rooted in concepts of natural law and similar ideas; it was basically radical, willing to substitute the judgment of allegedly objective thinkers for the evolved network of social institutions, a network too sophisticated to be understood in the abstract, and willing to achieve a rational optimum though the heavens fall in the process. The results were generally disruptive, frequently violent, and usually a failure. The archetype of that failure was the French Revolution, when proponents of the new Goddess of Reason slaughtered people and obliterated tradi- tional social institutions, including even the old Roman-derived names of the months of the year. Hayek echoes Burke’s challenge to the French Revolution: reasoned inquiry cannot understand the true nature of ancient social structures and therefore cannot appreciate the evils of changing them.
Problems with the Spontaneous Order/Rationalist Constructivism Dichotomy
Useful as Hayek’s observations are for understanding how social orders evolve, two significant and interrelated problems obstruct its use as the basis of a normative argument in favor of gradual, decentralized social change and against radical, planned, or conscious social change. First, we have no principled way to categorize the choice that a person might make as either an element of “rational constructivism” or an element of “spontaneous order.” This categorization rests entirely on the observer’s subjective frame of reference. The distinction is therefore trivial and
VOLUME 14, NUMBER 1, SUMMER 2009

8 F TIMOTHY SANDEFUR
 useless when we decide on a course of political or social action. Second, any attempt to change the social order or to reform an institution on the basis of conceptions of justice—the need for which Hayek himself recognizes—requires an appeal to exogenous principles, or conceptions of political right and wrong that are produced by conscious rational reflection and not simply by adherence to spontaneously generated mores.
When Would Hayek Pour The Cement?
We can appreciate the weakness of the distinction between rational constructivism and spontaneous order through a simple thought experiment.2 Suppose a new col- lege is being planned, and the architects must decide where to put the sidewalks for students to use when they cross the quad from one class to another. On the one hand, the architect might simply lay out the sidewalks and pour the concrete in the places he takes to be the best walking paths. He might even design the walkways to deviate slightly from the most efficient routes between the buildings in an attempt to manipulate students into keeping off the grass. In such a case, the architect would be a rational constructivist, dictating the terms on which students conduct their lives. On the other hand, an architect reared on Hayekian principles would wait a year before pouring the concrete, to observe how the students actually do cross the quad. They would trample down the grass—producing a spontaneous order—and he could then pour the concrete where they actually walk. The Hayekian architect would better suit the needs of actual individuals.
The problem is that life is dynamic: classrooms are reassigned, and students change their walking patterns. We may make an equally valid argument for the architect’s waiting another year before pouring the concrete. If he pours after one year, he is likely to find that the students have now changed their habits, rendering his plans obsolete. Yet at some point he must pour the concrete. Whether he waits a second year or not, he will be accused of engaging in constructivist rationalism the instant he begins to make the sidewalk, and plausibly so because by pouring the concrete, he will put a stop to (or radically alter) the dynamic evolutionary process in which students choose their own paths. Just as the observer cannot avoid interact- ing with the observed, so the concrete can never be poured without interfering in some fashion with the spontaneous order. It is not possible to label the architect either a constructivist or a participant in the decentralized spontaneous order. No matter how long he observes the students’ walking patterns, the instant he does pour, he can be accused of constructivism.
Consider another example: health care for workers. A political leader who decides to replace a free market for health care with a single, government-run health care
2. I owe this example to Hillsdale College professor Gary Wolfram, whose excellence as a teacher I am pleased to recognize.
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 9
 network for all citizens would seem to be the most extreme example of a rational constructivist. But imagine a large corporation working in a free market, whose ambitious new CEO decides that the firm should provide a single, universal, health care plan for employees. Is he a rational constructivist or merely a participant in the spontaneous order?
Seen from within the firm, he is clearly the former: he has consciously devised a new plan for organizing the provision of employees’ health care, rationally deciding how it will be provided to serve the needs he considers most important. But step back and observe the corporation as only one entity amid the bustling multitude in the free marketplace, and the CEO appears as simply an actor within the spontaneous order, proposing one experiment in one firm, an experiment that can be changed if the employees or directors regard it as a failure or if the corporation goes out of business. The director, like any kind of constructivist, is bound to incorporate into his plans some elements of the existing structures he seeks to reform; in this case, the existing health care resources or his knowledge of the illnesses his employees are most prone to contract. By incorporating these elements into his plan, he might claim to be merely taking one step within the spontaneous social evolution, no matter how radical his plan might be. The same is true if we imagine a government official making such decisions—say, a state governor who implements a government-run health care system. Seen from within the boundaries of one state, he is clearly a constructivist. Seen as the governor of one of fifty states in the nation, he appears as a single experimenter in the process of spontaneous order, particularly if he incorpo- rates elements of the existing freer market for health care into the plan.
This paradox came to the forefront recently when Judge Richard Posner wrote that Hayek’s emphasis on spontaneous order was “in considerable tension with his great admiration for the Constitution of the United States,” given that the Constitu- tion is a written plan of government formulated by a committee of experts and thus apparently a constructed order (2005, 151). Donald Boudreaux demurred on the grounds that the Constitution’s framers “did not seek to create all or even most law de novo” or “to replace wholesale one set of laws with another,” but instead incorporated “[t]he evolved common law rooted in English experience and modified by the more recent experience in the colonies” (2006, 162–63). But any committee of experts preparing a plan of legal or social reform, no matter how radical, will incorporate into the plan some element—usually a very large element—of the existing legal and social background. Hayek himself recognizes that “[e]ven when as a result of revolution or conquest the whole structure of government changes, most of the rules of just conduct, the civil and criminal law, will remain in force” (R&O, 135). It is difficult to think of any reform that did not do so to some degree, and thus it is difficult to imagine a circumstance in which an allegation of constructivism cannot be “refuted” by pointing out that some existing social institutions have been incorporated into the constructed reform proposal in question.
To reiterate the point simply, defining an action as rationalist constructivism or as an element in the spontaneous order depends entirely on the observer’s frame of
VOLUME 14, NUMBER 1, SUMMER 2009

10 F TIMOTHY SANDEFUR
 reference. Social change is constructivist in the short run and spontaneous in the long run. If the observer draws the circle narrowly around a single transaction, a single reform proposal, a single firm, a single state, or a single nation, then the action may appear as rationalist constructivism. But if he steps back and views that transaction, proposal, firm, state, or nation in the context of the enormous multitude of interac- tions between individuals and firms, any action—including even extremely radical, state-mandated reforms with long-lasting effects—can appear to be an element in the spontaneous order. In this way, the distinction appears to parallel (appropriately enough) the distinction between microevolution and macroevolution in biology. Although that distinction may be useful in some contexts, it is not a distinction in principle: all evolution is microevolution that becomes macroevolution when viewed over the course of aeons (Dawkins 2004, 601–5). There is much wisdom in the saying that “all politics is local.”
Making this distinction is not a hypothetical exercise. Consider the Shaker com- munities of the nineteenth century. The Shakers believed in the complete abolition of privacy, and they accordingly had no private property, no private conversations, and no families.3 Their lives were governed with a rigor that in some ways exceeded even that of twentieth-century totalitarian dictatorships: Shakers were told how to fold their hands (right hand on the outside) and how to walk up stairs (right foot first). Some communities employed elders to spy on communicants to ensure that they followed the rules. A more extreme example of constructivist rationalism probably cannot be found in American history. Yet the Shakers were only one among a great many religious communal movements of the era that were experimenting with new social institutions. In The Constitution of Liberty, Hayek observes that the open society will allow and even encourage such experimentation as a means of discovering the best social institutions: “[t]he existence of individuals and groups simultaneously observ- ing partially different rules provides the opportunity for the selection of more effective ones” (CL, 63). Seen from afar, in the whole context of nineteenth-century religious history, alongside the Oneida Community, the Mormons, New Harmony, and other groups, the Shaker movement was simply one instance in the wide-ranging process of spontaneous social evolution. Thus, the distinction between spontaneous order and rationalist constructivism depends on how one defines the frame of reference.
Randy E. Barnett (2005) employs Hayek’s observations regarding spontaneous order to criticize the Restatements, a series of legal encyclopedias that endeavor to lay out the principles of common law as derived from cases. Barnett criticizes the Resta- tements in Hayekian terms for two reasons. First, “once promulgated, a Restatement tends to freeze that common law evolution in amber at the moment of its creation.” Second, “[the] Restatements as we all know are more than mere restatements. But
3. At least in theory. Excavations of Shaker communities have revealed evidence that many Shakers managed to keep private property and probably had black markets for alcohol, cosmetics, and other proscribed items (Starbuck 1999).
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 11
 whatever ‘reforms’ [they] may include are themselves also frozen in time.” These critiques, however, fail for the same reason that criticism of the architect in my first example fails: at some point, he must pour the concrete for the sidewalks, and at that time, whenever it comes, he may be accused of obstructing the process of dynamic social evolution. Likewise, the Restatement authors may be accused of doing the same by writing down, restating, the content of common-law principles, no matter how faithful they are to the law’s actual content. The same is true of any judge who sifts through the cases to distill a rule of law as part of an opinion. Both judges and the Restatement authors may be criticized for obstructing the evolution of common- law doctrines whenever they say “the cases say that the rule is x,” and yet it is impossible for them not to take this step. Moreover, later judges are free to alter, distinguish, or recharacterize the content of a common-law principle, and once they do so, legal historians will regard the prior law, no matter how consciously planned it was at the time, as merely one element in the spontaneous order.
In short, any act by a person or firm in society may be characterized as “ratio- nalist constructivism,” in particular actions aimed at reforming an institution by making it more compatible with a preconceived criterion, such as justice. Yet at the same time that action may be characterized as one of the manifold experiments going on in the spontaneous order, from which people learn in time for the next round of experiments. The critique of an alteration of the social order is best aimed at the merits of that specific action, not at planning per se, because we expect and need to plan in most areas of life. If freedom is a discovery process, the discoveries must be implemented at some point; a time must come when spontaneity solidifies (however temporarily) into order.
The Problem of Spontaneously Generated Injustice
In at least two ways, Hayek’s inability to distinguish in principle between spontane- ous order and constructivist rationalism seriously hampers his ability to deal with the ways in which a social order may be reformed. Not only does he reach no satisfying conclusion with regard to the proper mechanism for reforming unjust institutions, but he provides no account of how people are to recognize institutions as unjust in the first place.
For Hayek, social orders persist when they succeed, but their success does not rest wholly on their truth, goodness, or beauty. One critic describes Hayek as believing that “[w]hat is ‘good’ or ‘bad’ is . . . a question of what proves to be effective in terms of survival” (Ogus 1989, 404). This claim is not entirely fair because although Hayek does make this claim (R&O, 99)—as well as the claim that people “have . . . no choice but to submit to rules whose rationale we often do not know, and doing so whether or not we can see that anything important depends upon their being observed in the particular instance” (CL, 66–67)—he recognizes elsewhere that social institutions may evolve and perpetuate themselves for reasons unrelated to their utility, equality, or justice.
VOLUME 14, NUMBER 1, SUMMER 2009

12 F TIMOTHY SANDEFUR
 Moreover, he recognizes that sometimes a “necessity” arises for “radical changes of particular rules” and even that “occasions [may arise] when it is recognized that some hereto accepted rules are unjust in the light of more general principles of justice.” These occasions, he writes, “may well require the revision not only of single rules but of whole sections of the established system of case law” (R&O, 89). Hayek therefore does not explicitly embrace the Panglossian positivism that defines the “good” as whatever rules perpetuate a social order and that therefore concludes that all is just in this most just of all possible worlds (see, for example, CL, 67). Yet his explanation of how reforms are to be undertaken remains unsatisfying.
One example of a longstanding, emergent, unjust social institution is slavery. This institution was never created by a constructed plan, but rather evolved—indeed, it is one of the oldest and most persistent of all human institutions—and its elimina- tion, in the United States at least, was extremely disruptive, even aside from the Civil War. But how does a reformer recognize an institution as being unjust or recognize a “necessity” for “radical change”? And how should the reformer eliminate such institutions?
Recognizing Injustice
How can the citizen in a spontaneous order recognize an undesirable social institu- tion? Hayek acknowledges that the common law may develop rules that “could not meet the more general requirements of justice” (R&O, 89), but his writing on the question of justice is riddled with inconsistency. Perhaps nothing illustrates this confusion better than the contrast between his claim in volume 2 of Law, Legisla- tion, and Liberty that “[s]ince only situations which have been created by human will can be called just or unjust, the particulars of a spontaneous order cannot be just or unjust” (MSJ, 33) and his recognition in volume 1 of the “necessity” for “radical changes” in the law when “some past development . . . produce[s] conse- quences later recognized as unjust” (R&O, 89, emphasis added).4 This inconsisten- cy is traceable to his antirationalism (Petsoulas 2001, 191).5 Hayek opposes the renovation of social institutions on the basis of exogenous principles—that is, the abstractions of contemplative reason—and even defends obedience to rules that are “based merely on tradition” and “[can]not be fully justified on rational grounds” (R&O, 10). Indeed, he argues at length in favor of “submission to undesigned
4. See also CL, 68: “[a moral principle should] be accepted as a value in itself, as a principle that must be respected without our asking whether the consequences in the particular instance will be beneficial . . . as a creed or presumption so strong that no considerations of expediency can be allowed to limit it” (emphasis added).
5. Hayek’s antirationalism is not as extreme as that of, say, John C. Calhoun or his modern epigone Russell Kirk, and in his famous essay “Why I Am Not a Conservative” it is balanced by an acute assessment of some of the problems with antirationalism (CL, 404–5). Yet the final verdict must be that Hayek’s antirationalism was strong enough, in the end, to direct his thought toward historicism. I do not have space here to defend this counterintuitive thesis at length, but later in this essay I explain it briefly.
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 13
 rules and conventions whose significance and importance we largely do not under- stand [and] reverence for the traditional” (CL, 63). Yet at the same time he argues that individuals, including state actors, ought to “endeavour to make society good in the sense that we shall like to live in it” (R&O, 33) and that legislation is necessary to “correct” a spontaneous order that has developed in “very undesirable directions” (R&O, 88–89).
Recognizing an institution as unjust requires one to compare it to some base- line, some principle of justice by which deviations can be evaluated. Hayek refers to “justice” as defining this baseline, yet this statement seems an invitation to construc- tivism. As Chandran Kukathas observes, Hayek believed that reason “can identify inconsistencies among rules within a situation (or tradition of behavior) but cannot stand outside the evolutionary process to evaluate different states of affairs that rational action might lead to” (1989, 197).
Hayek never explains where the citizen or the judge will obtain the understand- ing of the “more general requirements of justice” or the “general principles of justice” to which he refers. Philosophers in the “social contract” tradition have usually argued that social institutions may be evaluated in the light of exogenous, objective standards of natural justice accessible by reason, but Hayek’s antirationalism bars him from giving this answer (Kukathas 1989, 199). Yet if the baseline is an endogenous, socially relative one, fluctuating with society’s evolving mores, then it will be impossible to recognize an institution as unjust; no objective standards will allow the citizen to measure deviations. Hayek is therefore frequently left with the apparent self-contradiction of, on the one hand, arguing that “our morals are not a product but a presupposition of reason” (CL, 63) and inveighing against those who would “construct [social] rules by deduction from explicit premises” (R&O, 21), but, on the other hand, dealing with the problem of reform in terms that would please the most radical constructivist: we recognize evolved social institutions as unjust, he writes, by comparing them with “an ideal picture of a society which may not be wholly achievable, or a guiding conception of the overall order to be aimed at,” a picture that is “not only the indispensable precondition of any rational policy, but also the chief contribution that science can make to the solution of the problems of practical policy” (R&O, 65; see also Hasnas 2005, 97). This ideal picture will be drawn from abstract conceptions of justice—conceptions that often “must . . . be dogmatic and make no concessions to expediency” (R&O, 61). What French revolu- tionary would have disagreed?
In his effort to articulate endogenous principles by which a judge might recognize a social institution as needing reform, Hayek offers two other candi- dates. First, the judge might recognize that the institution is the product of an inequality in the procedures by which the institution has evolved. This inequality might have led to the formation of rules that “could not meet the general principles of justice” (R&O, 89). Hayek presumably has in mind the legal insti- tutions of segregation, which resulted from racial inequalities in the United
VOLUME 14, NUMBER 1, SUMMER 2009

14 F TIMOTHY SANDEFUR
 States.6 But, again, without a standard of proper equality, it will not be possible to determine whether the perceived inequality is worth fixing or not. Inequalities cannot be perceived as unjust or even as unequal without reference to some exogenous standard. In the case of segregation, the standard by which the insti- tution’s justice or injustice was measured was the (presumably “rationalistic”) conception of equality enunciated in the Declaration of Independence. But other kinds of differential treatment of classes—such as the rule that infant children are not allowed to vote—do not offend us because they do not contradict such a principle, and thus the classes are treated equally in the relevant sense. A racist, however, regards laws barring racial minorities from voting as equal in the rele- vant sense because he regards members of the minority as incapable of voting, as children are. Only by reference to a principle of abstract reason—for example, all adult citizens should be allowed to vote—can we assess this argument. If reason “[can]not stand outside the evolutionary process to evaluate different states of affairs” (Kukathas 1989, 197), then it cannot detect inequalities any more than it can detect injustice. Because traditional views of just and unjust, equal and unequal, are created by historical experience (PO, 166), judges who confine their deliberations to the inner content of the spontaneous order simply cannot detect an inequality when they look at it.
Hayek’s second principle for how a judge might discover that an institution needs reform relates to internal inconsistency: the judge’s role is to render social institutions logically consistent, to “bring consistency into a system of rules inherited by each generation” (MSJ, 40). But this principle also fails. First, like justice and equality, consistency is not a value intrinsic to humans or social institutions, but a desideratum of abstract reason. It is not necessarily “a common ideal shared and unquestioningly accepted by the majority” (CL, 206). It is easy to imagine societies in which inconsistencies in the law or social mores are not seen as objectionable.
Moreover, consistency, like equality, is a notoriously imprecise variable. It is easy for a judge to find, among the infinite number of variables that differ in any two cases, some plausible reason for distinguishing one from the other, especially if the judge must confine his reasons to those with a plausible historical pedigree. Indeed, it may be easier for an unjust social institution to satisfy the consistency criterion than the equality criterion;7 judges in slave states frequently responded to the apparent inconsistency of treating slaves as both property and persons by declaring that slaves were sui generis, which formally eliminates the logical inconsistency, or by treating
6. Although, of course, the romantic tradition of radical collectivism advances precisely the same argu- ment. Such revolutionaries do not claim that truly free and equal exchanges are unjust; they contend (correctly) that the present distribution of property is biased as the result of a history riddled with inequality and injustice.
7. As Anthony de Jasay concludes, “legislating for generality seems to be about as effective, and as constraining, as legislating for the common good” (2002, 184).
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 15
 slaves even more like property than previously so as to render the system more consistent and more horrible (Tushnet 2003, 63–64).
Finally, if internal coherence alone suffices as a guiding principle for institutional reform, this sufficiency strongly suggests a positivist view that justice is nothing more than the internal coherence of a legal and social system, regardless of whether the system as a whole suits the needs and nature of the human beings in it—a position strikingly similar to Hayek’s evolutionary conception of social rules as succeeding through competition (CL, 63). Although, unlike Ludwig von Mises, Hayek never goes so far as to declare that “[e]verything that serves to preserve the social order is moral; everything that is detrimental to it is immoral” (Mises [1962] 1996, 34), the proposition that reformers must confine themselves to “immanent criticism” and “piecemeal tinkering” to ensure only that the legal framework is internally consistent does suggest such a conclusion. This inference is reinforced by Hayek’s ultimate abandonment—despite some conflicting language elsewhere—of the principle of an individual’s inherent right to exist. Instead, he concludes that individual freedom is only a privilege created by society to serve its own needs: “Coercion thus is bad because it prevents a person from . . . making the greatest contribution that he is capable of to the community” (CL, 134).
Having confined himself to endogenous or immanent principles as the sole guideposts for evaluating societal institutions, Hayek is trapped in a variety of relativ- ism that necessarily leads to collectivism: if society is the source of all moral and political rules, then no principle outside of a society’s historical tradition can serve as a yardstick to judge a society’s treatment of the individual. If a particular citizen or judge’s reform efforts are confined to “immanent” criticism of institutions based on generally accepted propositions whose truth value is solely the product of their evolu- tionary success (or on principles logically inherent in those propositions), then we are ironically left with precisely the sort of historicism that Hayek elsewhere condemns ([1952] 1979, 111–139). As Linda Raeder observes, “Hayek’s . . . passionate evoca- tion of the transcendent significance of the person is starkly incongruous with his naturalistic-evolutionary justification of liberal values and principles. . . . Hayek . . . and Burke . . . seem to have subscribed to [a] sort of ‘value-centered historicism’” (1997, fn. 55). Given Hayek’s critique of historicism—and the fact that he published Karl Popper’s book The Poverty of Historicism (Ebenstein 2003, 256)—it seems ironic to pin this label on him.8 Yet Hayek shares with Hegel’s heirs the view that “the
8. Popper anticipated this charge when he acknowledged that his own view of reason “may be said to resemble slightly that of Hegel and the Hegelians, who consider reason as a social product and indeed as a kind of department of the soul or the spirit of society (for example, of the nation, or the class) and who emphasize, under the influence of Burke, our indebtedness to our social heritage, and our nearly complete dependence on it. . . . But . . . Hegel and the Hegelians are collectivists. They argue that, since we owe our reason to ‘society’—or to a certain society such as a nation—‘society’ is everything and the individual nothing; or that whatever value the individual possesses is derived from the collective, the real carrier of all values. As opposed to this, the position presented here does not assume the existence of collectives; if I say, for example, that we owe our reason to ‘society,’ then I always mean that we owe it to certain concrete individuals—though perhaps to a considerable number of anonymous individuals—and to our intellectual
 VOLUME 14, NUMBER 1, SUMMER 2009

16 F TIMOTHY SANDEFUR
 gravest deficiency of the older prophets was their belief that the intuitively perceived ethical values . . . were immutable and eternal” and that the cultural selection process operating on social tradition “creates reason” (PO, 166). Either the values that guide the reformer are universal, objective values logically implied by the “invariable fea- tures” (Hayek [1952] 1979, 134) of human nature—that is, natural law principles that allow the reformer to say “such-and-such evolved social institution is unjust regardless of its historical pedigree”—or these values are conventional, evolving through the social consensus that constitutes them as values. In the former case, a reformer is bound to use reason to discover these values and embed them in conduct (rational constructivism). In the latter case, the reformer is bound by the dictates of History realizing itself through the people’s will.9 When Hayek writes that the “[m]ind is . . . the result of man having developed in society and having acquired those habits and practices that increased the chances of persistence of the group in which he is” (R&O, 17) and that the mind is “embedded in a traditional impersonal structure of learnt rules” (PO, 157), one is reminded of nothing so much as Marx’s claim that man’s consciousness is the product of his material circumstances (Tucker 1978, 4).
Reforming Injustice
Hayek’s most sustained examination of the process of social reform in a spontaneous order comes in chapter 5 of the first volume of Law, Legislation, and Liberty, where he describes the judge’s role. This topic is especially important because the common-law judge ought to have even greater respect than the legislator for the gradual, emergent nature of the social order. Yet Hayek recognizes that judges ought to accomplish just
intercourse with them. Therefore, in speaking of a ‘social’ theory of reason (or of scientific method), I mean more precisely that the theory is an inter-personal one, and never that it is a collectivist theory. Certainly we owe a great deal to tradition, and tradition is very important, but the term ‘tradition’ also has to be analysed into concrete personal relations. And if we do this, then we can get rid of that attitude which considers every tradition as sacrosanct, or as valuable in itself, replacing this by an attitude which considers traditions as valuable or pernicious, as the case may be, according to their influence upon individuals. We thus may realize that each of us (by way of example and criticism) may contribute to the growth or the suppression of such traditions” (1966, 225–26). But this position seems like familiar rationalism with its individualistic and reductionistic conception of the social contract. Popper’s acknowledgment that tradition must be “analysed into concrete personal [that is, individual] relations” reflects an understanding that if individual personal- ities really are “socially constituted,” then their individual consciousness—their very status as individuals, both to themselves and to others—would also be a social artifact, and thus to appeal to their individuality would beg the question. One cannot separate out their individuality from their other qualities if they really are the creatures of social and historical—that is, collective—contexts and then argue that notwithstanding the fact that their consciousness is the creature of a collective, nevertheless they are individuals in some “deeper” sense. The very existence of such a “deeper” sense—the ability to analyze collective relations into concrete individual relations—contradicts the premise.
9. Of course, Hayek parts company with Hegel over the latter’s view that History reveals the Spirit of the People through a process governed by laws that humans can apprehend here and now. But he does not deny the process Hegel posits, only that a person can grasp these laws in their complexity. Yet even here the difference may not be so clean-cut. What Pierre Hassner says of Hegel’s views is strikingly reminiscent of Hayek’s: “The political constitution of a people is a result of its spirit (mind); thus it is dangerous to impose on a people a constitution constructed a priori. Political forms can only be spoken of historically; they can be judged only in relation to the extent of consciousness of freedom with which they are associated” (1963, 635).
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 17
 results in some cases by changing the way rules work, at times even by declaring laws unconstitutional. Judges should undertake “deliberate efforts” to “improve the existing system by laying down new rules,” but these new rules are not so much utterly new as “unarticulated” principles already inherent in the social structure. The judge may also “modify” the law by articulating principles that “when articulated, [are] likely to receive general assent.” The judge therefore should engage in “piecemeal tinkering, or ‘immanent criticism,’ to make the whole more consistent both internally as well as with the facts to which the rules are applied” (R&O, 100, 118).
Here we return to the frame-of-reference problem: tinkering is “piecemeal” only if seen from afar. Observed close at hand, such tinkering takes up the entirety of a legal opinion and may radically change a community’s social structure. Was the U.S. Supreme Court’s decision in Lawrence v. Texas (539 U.S. 558 [2003]), invali- dating a Texas law against private homosexual acts between consenting adults, a radical, constructivist alteration of the law, or was it piecemeal tinkering? Moral condemnation of homosexuality is certainly an ancient, evolved custom, a tradition whose purpose many cannot justify or understand and around which social institu- tions have grown spontaneously. This condemnation is what Hayek elsewhere calls “part of a moral tradition of the community, a common ideal shared and unquestion- ingly accepted by the majority” (CL, 206). Simply to declare unconstitutional many states’ laws that are based on this common belief and to do so in the name of a rationalistic conception of individual liberty are good reasons from a Hayekian per- spective to condemn the decision as a radical alteration of the social order, and, indeed, conservatives have repeatedly employed Hayekian grounds for criticizing courts that have ruled in favor of gay rights (see, for example, Goldberg 2004). Yet at the same time many, if not most, Americans regarded the law at issue in Lawrence to be (in the words of one of the dissenting justices) “uncommonly silly,” and such laws seem to have been seldom enforced. Moreover, such laws clearly conflicted with the Constitution’s protections of individual liberty. The justices who declared the Texas antisodomy law unconstitutional could argue convincingly that they were only articulating previously unarticulated but logically inherent principles of American law in terms that, in fact, have received general assent, thus acting well within the spontaneous process by which social mores evolve. It is not surprising, therefore, that others have used Hayekian grounds to praise the advent of gay marriage (Rauch 2004, 160–75).10
Evolution, of course, is not a thing, but a process, or rather an aggregation of processes undertaken by the particular entities that do the evolving. It consists of the nonrandom selection of traits that change for reasons that in the aggregate can be described as evolution, but in each individual instance are not dictated or constituted by evolution. Evolution is an excellent way to picture the dynamic natural world, but
10. Lawrence, of course, was not about marriage, but that difference does not affect my point.
 VOLUME 14, NUMBER 1, SUMMER 2009

18 F TIMOTHY SANDEFUR
 it cannot instruct any particular lion to eat or not eat any particular antelope: if the lion does eat the antelope, that action will be one step in the process of evolution, but the same will be true if he does not eat it. “Respecting evolution” therefore cannot serve as a normative guide.
To use a different analogy, a random genetic mutation may give a bacterium a reproductive advantage, but that mutation is a random one, and its success or failure depends not on “the principle of evolution,” but on its own reproductive fitness. Moreover, at any moment, a population of bacteria will contain some with radical mutations far outside the norm for that population. Most of these bacteria die off, of course; viewed in the aggregate, all of these outliers, both successful and unsuccess- ful, will be rightly described as falling within the overall process of evolution, no matter how extreme. The same is true of common-law evolution. Theodore Burczak is wrong to claim that Hayek “does not recognize that common law judges may also act according to their subjective, theory-laden perceptions of just out- comes” (2006, 78); in fact, Hayek explicitly contemplates judges doing so. But if a judge makes a radical or extreme pronouncement to which voters respond with outrage, both that outlying “mutation” and the reaction against it can ultimately be described as part of the overall process of evolution. Both the judge and the outraged public may plausibly claim to be acting on Hayek’s advice about the benefits of spontaneous order.
Thus, what Amy Wax writes about Burke and Michael Oakeshott is equally applicable to Hayek: the “particularism” of his views of social change “would appear to offer little hope of developing an all-purpose heuristic for assessing specific reforms.” Such “general precepts . . . as giving the status quo the benefit of the doubt and eschewing rationalism do not get us very far in resolving particular ques- tions.” The principle of spontaneous order simply does not provide “a clear roadmap for reform . . . [or] a checklist of specific criteria for classifying changes as good or bad” (2005, 1075, 1089; see also Schwarzchild 2005, 1118).
An inherent conflict therefore exists between Hayek’s vision of the judge as a reformer—piecemeal or not—and his view that social institutions should evo- lve spontaneously without employing rationalized abstractions about justice.11 By “plac[ing] judges’ conscious effort to make rules at the heart of the development of the law of liberty,” writes John Hasnas, Hayek
injects an intentional element into what he otherwise claims to be a process of spontaneous legal evolution. But intentional action is purposeful action;
11. It may be that Hayek’s error here lies in the belief, expressed elsewhere in his work (for example, MSJ, 87), that one can act “outside the market” or take a perspective outside of the economic order ([1944] 2007, 71). If such a position were possible, it might make sense to use “respect for spontaneous order” as a normative guide, like when a nature photographer refuses to rescue a wild animal in distress. But the markets or the social order as a whole obviously cannot be looked upon that way because every human choice—whether to intervene or not—is automatically part of the order, and other people will base their choices on it, building up a self-referential spontaneous order around that choice as well.
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 19
 a judge cannot make a rule without some conception of the purpose the rule is to serve. And because Hayek is supposed to be describing a sponta- neous process, there is no higher human authority to assign the judge the normative value he or she should seek to advance in creating the rule. Therefore, to make a rule, a judge is necessarily required to make a norma- tive choice. He or she must personally decide what normative end the rule should advance. (2005, 103)
Hayek, in his efforts to free himself from this trap, simply falls again into the problem that constructivism lies in the eye of the beholder. He writes that “law as we know it could never have fully developed without such [constructivist] efforts of judges, or even the occasional intervention of a legislator to extricate it from the dead ends into which the gradual evolution may lead it . . . [y]et it remains still true that the system of rules as a whole does not owe its structure to the design of either judges or legislators” (R&O, 100, emphasis added). If even radical alterations in existing law qualify as part of the spontaneous order because the system “as a whole” is the product of social evolution, then it is difficult to imagine an alteration of rules that will not be exonerated of the charge of constructivism. As Kukathas concludes, “if ‘reason’ must be viewed as merely an aspect of the development of social order” in Hayek’s system, “not only does it become impossible to distinguish spontaneous processes from constructed organi- zations, but the very idea of criticism and social reform becomes illusory” (1989, 104).
In The Road to Serfdom, Hayek offers a telling analogy: “The attitude of the liberal toward society is like that of the gardener who tends a plant and, in order to create the conditions most favorable to its growth, must know as much as possible about its structure and the way it functions” ([1944] 2007, 71). The gardener, however, is not himself a plant; he is not constituted by the garden, and his conception of the proper form of a garden is not dictated by the spontaneous growth of the plants he tends. Instead, the gardener stands outside of the garden with an exogenous idea of how it ought to look, and he rationally constructs it, prudently allowing plants to grow in some ways and pruning back others. Putting aside other problems with this metaphor (as noted in Block 1996, 341), it is clear that the antirationalist conception of a gardener working entirely within the system, confined by endogenous rules or the inchoate principles implied by those rules, does not work either for gardeners or for involved citizens. Internal coherence cannot suffice as a guiding principle for a reform-minded judge because either it is an abstract, exogenous value, or it is defined so broadly that, like “spontaneous order,” it can apply to any course the judge chooses, depending on the observer’s perspective. Hayek’s judge cannot recognize a social institution as needing reform and cannot reform it when he does recognize it. Hayek cannot, as he desires, escape the disturbing conclusion of antirationalism in general: that whatever order
VOLUME 14, NUMBER 1, SUMMER 2009

20 F TIMOTHY SANDEFUR
prevails should be perpetuated, regardless of objective principles of justice. His
conception of social reform is therefore illusory.
Possible Boundaries Between Spontaneous and Constructed Orders
We have seen that Hayek does offer a candidate for a principle that might meaning- fully guide efforts to distinguish between constructivism and spontaneous order: the requirement of internal consistency. This principle is attractive to him because of his emphasis on “immanence”; consistency seems to be an organic guideline for reform from within the system. In fact, however, as we have observed, it is still a top-down ordering principle based on the abstract conclusion that the system ought to be consistent in the first place—an idea not itself clearly endogenous to any social system.
In “Kinds of Order in Society” (1964), Hayek seeks to clarify the differences between kinds (or, perhaps more precisely, different levels) of social order. He makes little sustained normative argument in this essay, however. Instead, he sets out two categories of orders: organizations, in which the parts are deliberately arranged in relation to one another with a conscious aim in mind, and organisms, the self- generated spontaneous orders resulting when a group of entities react in a regular manner to a given stimulus.
To illustrate an organism, Hayek uses the analogy of iron filings reacting to a magnetic field: the order is created by the fact that individual bits of iron respond in a predictable manner to the presence of magnetism. He is doubtless correct here in describing self-ordering processes, but this analogy and others have no normative content and notably provide no grounds for arguing against the use of coercion to impose an order. Coercion, as we have noted, can be an ingredient in spontaneous- ordering processes just as surely as in anything else; the “tradition” of holding one’s hands up when a robber sticks a gun in one’s face is surely a regular, predictable response, rooted in long historical experience, and forms a kind of spontaneous order. But is it to be preferred to the spontaneous order that we know of as mutually agreed economic exchange? Hayek’s description cannot guide our choices about how to regulate interpersonal relationships.
Moreover, his distinction between spontaneously ordered organisms and deliberately created organizations is nowhere near as clear as he suggests. As the article progresses, Hayek explains that these two kinds of orders are in fact deeply intertwined. Even where “the conduct of individuals which produces the social order is guided in part by deliberately enforced rules, the order is still a sponta- neous order.” And this condition, he continues, is the case even where the rules are enforced by a deliberately created organization, such as the state. The reason that individual conduct in such a situation can still be called a spontaneous order is that it is not coordinated “according to a preconceived plan.” Even where
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 21
 individual activities are coordinated according to a preconceived plan, Hayek can still characterize them as part of a spontaneous order simply by taking a step back and looking at the big picture. An organization such as the Salvation Army, the Boy Scouts, Microsoft, or General Motors is organized by rules enforced by the organization and according to a deliberate plan for attaining a particular end. The elements are consciously arranged in consideration of certain deliberate objectives. Seen from within, therefore, these firms must be characterized as deliberately constructed orders. Yet in “Kinds of Order” Hayek steps back and regards them as elements in the overall, spontaneous, “polycentric” order: “free system[s] . . . have many organizations (in particular, firms) as their elements . . . [and] require an organization to enforce obedience to (and modify and develop) the body of abstract rules which are required to secure the formation of the spontaneous overall order” (emphasis added). Thus, here, as in his later book Rules and Order, Hayek resorts to the tactic of showing that the system “as a whole does not owe its structure to [conscious] design” (R&O 100, emphasis added). This tactic explains why Hayek employs the term polycentric order: he is describing the interaction of social “elements”—which can be consciously designed and deliber- ately or even tyrannically organized—at such a degree of generality that any result can be described as the spontaneous and unplanned.
Another principle might more effectively distinguish spontaneous from con- structed orders: the presence or absence of coercion. An action might be said to qualify as part of spontaneous order only if it is implemented in a noncoercive manner, whereas if the action is undertaken coercively, particularly by the state, it qualifies as constructivism. The Shakers who choose to sit and walk a certain way or the corporate director who offers a health care plan to employees are therefore acting within the spontaneous order, whereas a lawmaker who imposes a government- run health care plan on a society without regard to the participants’ consent is a constructivist.
This answer, however, raises two problems. First, as we have seen, Hayek’s conception of spontaneous order is fundamentally neutral with regard to the pres- ence or absence of coercion. Social institutions and habits grow up around coercive institutions all the time. The businesses that sell refreshments to people standing in line at the post office on April 15 and the market for accounting firms that help people finish their taxes at the last minute are spontaneous orders, even though they have sprung from the coercive institution of income taxation. Even the “tradition” of holding up one’s hands when a robber sticks a gun in one’s face is an undesigned, evolved social tradition. In “Kinds of Order in Society,” Hayek explicitly contem- plates a “spontaneous” order that results from people’s reactions to coercive rules they are “made to obey.”
Second, Hayek is barred from appealing to the coercion criterion if he is going to remain within the immanent criteria for social reform. Adopting and consistently implementing the principle of noncoercion would require a radical alteration of
VOLUME 14, NUMBER 1, SUMMER 2009

22 F TIMOTHY SANDEFUR
 existing “grown” social orders—the very “hubris of reason” (R&O, 33).12 Of course, Hayek did not shy away from condemning the use of state coercion, which he emphatically opposed. Nevertheless, in his work he seems not to recognize that his opposition to coercion is itself a moral principle derived by reason from the nature of man (Rasmussen and Den Uyl 2005); he seems to regard it as the absence of a rational moral conviction—a habit arising from skeptical restraint rather than a logi- cally defensible belief in a noncoercive society. The importance of freedom, he writes, “rests mainly on the fact that the development of custom and morals is an experi- mental process” (MSJ, 57). Thus, Hayek seeks to characterize his opposition to coercion as an immanent value, arising from respect for spontaneous order itself. But this position would mean that invoking the absence of coercion as a principle for demarcating the boundary between spontaneous order and constructed order would beg the question. If one believes in eliminating coercion out of respect for society’s “experimental process” (MSJ, 57), then one has no basis for eliminating coercion in a society that decides to experiment with the use of coercion.
Hayek cannot commit to noncoercion on the basis of principles arising outside the system of inherited rules—natural law, for instance—without appearing to em- brace constructivism. In addition, doing so would shift the focus of the argument away from the spontaneity and efficiency of the organic social order and toward more familiar arguments about the justice or injustice of coercion itself. Although Hayek refers in one place to “the rule that nobody is to coerce others in order to secure for himself . . . a particular income” (MSJ, 95), he does not explain where this alleged rule comes from. It is certainly not to be found in most social traditions, which are replete with examples of the opposite, from slavery to the welfare state. This rule is not immanent or endogenous; it is an abstract, not to say revolutionary, principle of political philosophy derived from Enlightenment rationalism.
Viktor Vanberg rightly observes that in general Hayek’s criterion for the “nor- mative standard” of “constructive reform” appears to be “the notion that institutions are beneficial if, and to the extent that, they benefit the persons living with them” (1994, 187). The problem, however, is not that Hayek is “not totally unambiguous on this matter” (187). Rather, it is that this criterion, like “justice,” is a conception of the good—an exogenous philosophical premise on which a political actor relies when
12. Barnett, by contrast, does employ a distinction based on coercion when he distinguishes “centralized” and “decentralized” ordering. Centralized ordering is formed by “delegating to some subset of persons or associations in a society the authority to regulate the conduct of other[s]” (1998, 45). This principled distinction enables him to discuss more precisely the relationships between the two kinds of orders he envisions (57–61). “Centralized ordering”—what I have called “pouring the cement”—is, Barnett acknowledges, “absolutely vital to implementing the personal and local knowledge of individuals and associations” (57). The interactions between these two types of ordering allow for the implementation of spontaneously generated ideas, followed by the next round of spontaneous or decentralized elaboration and development. This argument is available to Barnett only because he is not a resolute opponent of “rationalism” and because he incorporates a healthy amount of “abstract theory” (109) into his work. This approach, in turn, is owing to his recognition that “[a]bstract natural rights and rule of law principles exclude wrong answers rather than definitively establish right ones” (110), a proposition that seems to have eluded Hayek (R&O, 21).
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 23
 working within the political world—and not merely an inherited tradition. Hayek’s struggle to push away from rationalism bars him from appealing to these normative criteria and from explicitly embracing or convincingly explaining justice or any other objective standard of political value such as the criterion Vanberg infers from Hayek’s writings.
Conclusion
Hayek’s observations on the nature of social evolution and spontaneous order are insightful and profound when used as a descriptive scheme for understanding how social institutions grow and change, but they do not convert effectively into normative guides. A wise lawmaker or judge seeking to follow Hayekian prescrip- tions is left with a vague and contradictory set of precepts. On one hand, it is unwise to alter long-standing social institutions by implementing rational abstrac- tions of justice, but, on the other, it is necessary to do so because not all social orders that have spontaneously evolved are just ones. If the lawmaker or judge acts to alter a social institution, that action may be condemned as rationalistic constructivism when viewed at close range, but will appear as one experiment undertaken in the whole process of spontaneous order when viewed as part of the bustling dynamism of society as a whole. Spontaneous order is therefore in the eye of the beholder.
This problem is closely related to one in Hayek’s critique of rationalism. This critique requires him to shy away from the conscious implementation of abstract principles of justice derived from reason. Yet without these principles no one can undertake meaningful “piecemeal tinkering” with the social system. Judges may seek to improve the legal order’s internal consistency—assuming consistency can be justi- fied as an “immanent” norm—but this action has no necessary connection with making the society more just or efficient. Although the presence or absence of coercion might help us to distinguish between laudable respect for spontaneous order and imprudent constructivist rationalism, Hayek does not offer this criterion because it, too, would appeal to abstract, rationally derived principles as a guide for political conduct and thereby undermine his critique of rationalism.
In the end, Hayek’s advice to the reformer boils down to “[m]ove ahead, but be careful” (Rauch 2004, 171). This advice may be wise counsel, but it is noticeably lacking in content. Hayek’s observations offer a helpful descriptive model for under- standing social and legal change, but his conceptual distinctions cannot be translated into a prescription for how a legislator or judge ought to act.
References
Barnett, R. 1998. The Structure of Liberty: Justice and the Rule of Law. Oxford, U.K.: Clarendon Press.
VOLUME 14, NUMBER 1, SUMMER 2009

24 F TIMOTHY SANDEFUR
————. 2005. The Questionable Value of Restatements. The Volokh Conspiracy, at http:/
volokh.com/archives/archive_2005_02_06-2005_02_12.shtml#1107872845.
Block, W. 1996. Hayek’s Road to Serfdom. Journal of Libertarian Studies 12: 339–65.
Boudreaux, D. 2006. Hayek’s Relevance: A Comment on Richard A. Posner’s “Hayek, Law, and Cognition.” New York University Journal of Law and Liberty 2: 157–65.
Burczak, T. 2006. Socialism after Hayek. Ann Arbor: University of Michigan Press.
Dawkins, R. 2004. The Ancestor’s Tale. New York: Mariner Books.
De Jasay, A. 2002. Justice and Its Surroundings. Indianapolis, Ind.: Liberty Fund.
Dewey, J. 1935. The Future of Liberalism. Journal of Philosophy 32: 225–30.
Doherty, B. 2007. Radicals for Capitalism: A Freewheeling History of the Modern American Libertarian Movement. New York: Public Affairs.
Ebenstein, A. 2003. Hayek’s Journey: The Mind of Friedrich Hayek. New York: Macmillan.
Goldberg, J. 2004. Misusing Hayek: Jonathan Rauch’s Formulation Doesn’t Work. National Review Online, August 20, at http:/www.nationalreview.com/goldberg/goldberg 200408200923.asp.
Hasnas, INITIAL. 2005. Hayek, the Common Law, and Fluid Drive. New York University Journal of Law and Liberty 1: 79–110.
Hassner, P. 1963. G. W. F. Hegel. In History of Political Philosophy, edited by Leo Strauss and Joseph Cropsey, 628–56. Chicago: University of Chicago Press.
Hayek, F. 1960. The Constitution of Liberty. Chicago: University of Chicago Press.
————. 1964. Kinds of Order in Society. New Individualist Review 3, no. 2. Available at: http:/
oll.libertyfund.org/index.php?option=com_content&task=view&id=1269&Itemid=280. ————. 1967. The Results of Human Action but Not of Human Design. In Studies in
Philosophy, Politics, and Economics, 96–105. Chicago: University of Chicago Press.
————. 1973–79. Law, Legislation, and Liberty. 3 vols. Chicago: University of Chicago Press.
————. [1952] 1979. The Counter-revolution of Science. Indianapolis, Ind.: Liberty Fund.
————. [1944] 2007. The Road to Serfdom: The Definitive Edition. Chicago: University of Chicago Press.
Kukathas, C. 1989. Hayek and Modern Liberalism. Oxford, U.K.: Clarendon Press.
Mises, L. von. [1962] 1996. Liberalism: The Classical Tradition. Irvington-on-Hudson, N.Y.:
Foundation for Economic Education.
Ogus, A. I. 1989. Law and Spontaneous Order: Hayek’s Contribution to Legal Theory.
Journal of Law and Society 16: 393–409.
Petsoulas, C. 2001. Hayek’s Liberalism and Its Origins: His Idea of Spontaneous Order and the
Scottish Enlightenment. London: Routledge.
Popper, Karl. 1966. The Open Society and Its Enemies, 5th ed., vol. 2. Princeton, N.J.: Princeton
University Press.
Posner, R. 2005. Hayek, Law, and Cognition. New York University Journal of Law and Liberty 1: 147–65.
 THE INDEPENDENT REVIEW

SOME PROBLEMS WITH SPONTANEOUS ORDER F 25 Raeder, L. 1997. The Liberalism/Conservatism of Edmund Burke and F. A. Hayek: A Critical
Comparison. Humanitas 10: 70–88.
Rasmussen, D., and D. J. Den Uyl. 2005. Norms of Liberty: A Perfectionist Basis for Non-
perfectionist Politics. University Park: Pennsylvania State University Press.
Rauch, J. 2004. Gay Marriage: Why It Is Good for Gays, Good for Straights, and Good for
America. New York: Times Books.
Schwarzchild, M. 2005. Marriage, Pluralism, and Change: A Response to Professor Wax. San
Diego Law Review 42: 1115–123.
Starbuck, D. 1999. Latter-Day Shakers. Archaeology 52, no. 1. Available at http://www.
archeology.org/9901/newsbriefs/shakers.html.
Tucker, R. 1978. The Marx-Engels Reader. 2d ed. New York: Norton.
Tushnet, M. 2003. Slave Law in the American South. Lawrence: University of Kansas Press.
Vanberg, V. 1994. Hayek’s Legacy and the Future of Liberal Thought. Cato Journal 14: 179–99.
Wax, A. 2005. The Conservative’s Dilemma: Traditional Institutions, Social Change, and Same-Sex Marriage. San Diego Law Review 42: 1059–103.
Acknowledgments: Many of the ideas in this article were inspired by conversations with Donald J. Cooper of the Central Michigan University Center for Charter Schools. Others were inspired by discus- sions at a 2005 Socratic seminar on Hayek sponsored by Liberty Fund. I am also indebted to Jason Kuznicki of the Cato Institute for the Karl Popper citation.
}}