Spontaneous Order and Libertarian Social Theory– Charles Johnson
Spontaneous order theories spring from an observation that the vital underpinnings of human social life include largescale patterns of interaction and coordinating structures which seem to emerge without — and indeed, beyond the capacity of — any intelligent designer or designers to construct or manipulate to particular ends. Hayek writes in “Kinds of Order in Society” (1964) that these “spontaneous orders,” unlike orders that result from conscious organization according to a preconceived plan, are “orders of another kind which have not been designed by men [sic] but have resulted from the action of individuals without their intending to create such an order;” (¶ 7) he later characterizes spontaneous order as a “polycentric order” (¶ 9) in which, rather than “creating a concrete preconceived order by putting each individual on a task assigned by authority,” the interconnected but spontaneous actions of individuals produce an emergent order, without putting any one person or committee of people in charge, as “an adaptation to a multitude of circumstances which are known only to the individual members but not as a totality to any one of them.” Examples of spontaneous social orders are familiar and evermultiplying both in economics and in libertarian social theory: the adjustment of market prices to relative scarcities under conditions of free exchange, the convergence on common media of exchange in barter economies, the emergence and maintenance of early networks of paths and roads prior to largescale militaristic roadbuilding.9 Although closely associated with freemarket economics, there is no conceptual reason why the employment of spontaneous order concepts need be limited to narrowly economic topics:10 Hayek viewed the attention to spontaneous orders as the distinguishing feature of all serious social sciences, and frequently mentioned as examples the emergence and ongoing evolution of human languages and writing systems, which, he remarks, “possess an order which nobody has deliberately designed and which we have to discover,” through a science of linguistics. He devotes his closest attention, in later works, to the polycentric evolution of customary principles at common law for resolving disputes peacefully, without recourse to blood feuds or vendettas11 Other examples of spontaneouslydeveloping community norms have ranged from conventions regulating the use of agricultural commons12 to international postal standards to the evolution of common rules for American baseball.13
In the hands of Hayek and other libertarian social theorists, the concept of “spontaneous order” is employed not only as an explanatory alternative to monocentric, “constructive” orders and government planning, but also an alternative normative ideal. Thus, in his contributions to the socialist calculation debate and his responses to Marxist and Fabian worries about the “social anarchy of production,” Hayek invoked spontaneous order to argue that positive social order can emerge without deliberate “social regulation,” and that production without centralized control need be neither blind, destructive, nor chaotic14 — that, in fact, ordinary individuals acting on the dispersed knowledge embodied in price signals could discover opportunities, anticipate future needs, correct allocative errors, and adjust to changing conditions far beyond the capacity of even the most comprehensive aggregate statistics and bestintentioned central regulator or planner. Changing gears from knowledge of production to the production of knowledge, an increasingly popular example with the kids these days is the explosive growth, refinement, and success of the Internet’s knowledge and communications systems in general, and of Wikipedia in particular15 — a project which was directly inspired by Hayek’s remarks on dispersed local knowledge in “The Uses of Knowledge in Society,”16 and which depends on dispersed contributions, emerging consensus, and the polycentric implementation of evolving community norms to inform and shape its articles. In less than a decade Wikipedia has become the world’s largest and most successful encyclopedia even though — or rather, precisely because — its development is driven by the “social anarchy” of contributions and decentralized initiative from millions of volunteer reader/editors, without prior invitation, authorization, screening or direction by any central point of authority.
IV. Three Versions of Spontaneity
A careful rereading of the Myrmidon passage from Brownmiller, with Hayekian lessons in mind, ought to bring out a number of features in her analysis of rape culture that are reminiscent of common characterizations of spontaneous order. Brownmiller holds that rape culture involves some conscious, centrally coordinated campaigns — such as the use of rape as a weapon of war in conflicts between malegoverned nationstates. But her understanding of rape culture crucially depends on the structural effects of widely dispersed actions, which are carried out by a “swarm of men” acting “in anonymity,” rather than by governments or organized bodies of men acting on a centrallydirected plan; this ought to suggest a very clear and direct parallel to Hayek’s characterizations of spontaneous order as polycentricorder, more akin to “organism” than to “organization.” The undirected but systematic actions of the “swarm” of Myrmidonrapists have profound social effects but, because of their very anonymity, “policeblotter rapists have performed their duty … so well … that the true meaning of their act has largely gone unnoticed” (209); just as Hayek characterizes spontaneous orders in such terms as “the unintended and often uncomprehended results of the separate and yet interrelated actions of men [sic] in society.”17 Yet there are of course other ways in which the sexual politics of a rape culture are more reminiscent of “constructed” order and governmental politics than they are of paradigmatic spontaneous orders. Crucially, the dispersed acts that we are discussing are, after all, not free exchanges or willing negotiations, but acts of force, and so any social order emerging from them must be seen as a social order imposed on women against their will, just as the economic or social plans of governments are imposed on the governed without genuine individual consent.18
Both the similarities and the differences from our paradigmatic cases may help reveal two important and interrelated points that I’d like to make about the notion of “spontaneous order.” Both points are easily missed, and often are missed, in extant writing about spontaneous order, in part because of the normative work that spontaneous order does in libertarian arguments for freeing economic exchange from government control. But as valuable and insightful as that normative employment may be, attempting to apply Hayekian categories to Brownmiller’s account of rape culture may help to highlight the fact that market exchange and government intervention, considered as systems of interpersonal coordination, differ from each other along more than one dimension. Perhaps because these dimensions are so often linked with each other in the cases that spontaneous order theorists commonly discuss, they are often treated as inseparable, if not simply conflated with one another. But rape culture, as understood in Brownmiller’s theory, exhibits some of the features of Hayekian spontaneous orders, while seeming in other senses definitely “constructed” and imposed, and considering a case where these features come apart, may help bring out these different senses of, and thus show that the notion of “spontaneous order,” as employed in libertarian writing, is systematically ambiguous. I would in fact argue that the term may evoke at least three different sorts of distinctions, depending on the precise sense of the critical term “spontaneous.”19 “Spontaneous order” can be used to mean a macroscale pattern of social coordination which is:
Consensual rather than coercive (when “spontaneous” means “uncoerced”);
Polycentric or participatory rather than directive (when “spontaneous” means “unprompted”); or
Emergent rather than a consciously designed pattern (when “spontaneous” means “not planned in advance”);
When “spontaneous order” means consensual, rather than coercive, order, coordination is achieved through the free actions and agreements of many different people, contrasted with coordination imposed by using force to compel the participation of unwilling parties. So, for example, when marketable commodities such cigarettes emerge as de facto currencies in barter economies, they are not selected because any authority forces traders to accept them as payment for all debts public and private; it’s because enough people will willingly trade for a smoke that even nonsmokers find it worth their while to accept cigarettes as payment, on the expectation that they could easily make a later exchange with some third party who does smoke, in order to get things they can use.
When “spontaneous order” means polycentric, rather than directive order, coordination comes about through the converging microactions of many players acting independently, rather than deferring to designated supervisors or authoritative decisionmakers or relying on external plans or instructions. In directive orders, coordination happens vertically as players act on the bidding of a recognized personal authority, who takes responsibility for assigning, vetting, and integrating their many microscale courses of action. Polycentric orders, by contrast, depend on participatory or entrepreneurial action: there is no recognized final authority, action is guided by impersonal norms rather than personal deference, and it’s up to individual actors to horizontally coordinate with other participants, to determine which courses to pursue in order to achieve their goals, and so on. Consider the contrast between the directive system of topdown editorial vetting in Encyclopedia Britannica, and the wideopen system of selfcorrection in Wikipedia, which depends on the attention and initiative of an Internetsize pool of potential reader/editors to revise and correct articles. Contributions to Wikipedia are guided by a shared goal of factual accuracy and explicit community norms such as Neutral Point of View20or Citation Needed,21 but rather than being imposed by privileged editors, the interpretation and implementation of these norms, as well as most other questions of vetting and policy, rely almost entirely on the convergent consensus produced by the backandforth among Wikipedians, all with equal power to add, revise or revert changes, and with anyone free to join the fray at any given time.
When “spontaneous order” means “emergent order,” contrasted with conscious design, forms of social coordination emerge from the actions of many different people, acting on motives separate from any conscious desire to effect that form of social coordination. Intentional orders effect social coordination through people acting for the sake of a shared purpose, whatever it may be; in undesigned orders, participants may know nothing about the macropattern emerging from their interlocking microscale actions; or they may be aware of it, but consider it only a side effect — even if a pleasant one — of pursuing a different private purpose. A path breaker whose actions help clear and maintain a road through the woods is not mainly out to help future fellow travelers, or to lay the groundwork for a future highway; she’s out to remove obstacles that block her way from point A to point B. If it helps other people out later on, that collateral effect is just gravy.
For rhetorical simplicity, these three dimensions of spontaneity and nonspontaneity are presented here, as they usually are elsewhere, as if they were simple dichotomies. But each of them must be assessed relative to a level of social organization — an overall polycentric order may contain, as Hayek says, “several nuclei” of centered organization (such as firms or associations in an open market); a consensual process may produce their distinctive spontaneous results because they operate against a backdrop of coercive constraints (when cigarettes are consensually adopted as de facto currency in a prison economy, the adoption is at one level consensual; but of course it is profoundly shaped by pervasive, coercive constraints on possession and exchange, which forbid most uses of outside money). And both the distinction between polycentric and directive orders, and that between emergent and intentional orders, are really differences of degree, with many intermediate shades and borderline cases rather than clean categorical breaks. Participants can exercise greater or lesser degrees of autonomy in selecting and vetting their courses of action; the gap between microscale intentions and the macroscale pattern that results may be more or less wide of a gap.
It is important to understand each of these three distinctions as interrelated but analytically distinct pairs of categories — but they tend to coincide in fact often enough that they might be mistaken as indistinguishable in concept. Notably, when libertarians contrast open networks of market exchange with economic planning and allocation by governments, they are contrasting socioeconomic orders that differ along all of our three dimensions: governmental allocation is legally enforced,coordinated by requirements from a designated authority, and consciously designed to achieve a predetermined set of policy goals; whereas free markets produce organic structures which are the product of consensual agreements, which are participatory in character (without a fixed center of authority), and develop an emergent structure that the parties to the exchange did not consciously set out to create. The question, then, is which of these differences we should treat asdefinitive of spontaneous order. Hayek himself was fairly consistent when he attempted formal definitions of “spontaneous order” — as we have seen above, he defined it in terms of emergent coordination, with a noticeable gap between microintentions and macropatterns, and often recurring to Ferguson’s “results of human action but not of human design.” But Hayek also constantly characterizes spontaneous orders by contrast with “constructed” orders that he defined in terms ofmonocentric authority, simply equating them with “polycentric” orders which are “not made by anybody”22 as opposed to order “which has been made by somebody putting the elements of a set in their places or directing their movements” (1973, 37). His application of the concept in discussing freemarket processes consistently contrasted their healthy, unimpeded ordering process with purposive “interference” or “intervention,” which are both monocentric and paradigmatically backed by the coercive power of government,23 and in general closely linked it with cases where the consensuality of the transaction was at least as important as its emergent properties or participatory context. In any case, both later Hayekian scholars and popular writing have repeatedly used “spontaneous order” indifferently to refer to orders that are “spontaneous” in any of our three senses or all of ’em; or have equivocated between different senses of “spontaneous” from one statement to the next.
Thus, for example, in his chapter on Hayek in Against Politics (1997), Anthony de Jasay passingly characterizes “spontaneous orders” as emergent orders, “an unintended result of human actions directed at other purposes” (121122), when distinguishing them from the results of conscious political activism. But later in the same chapter he directly equates spontaneous orders with consensual orders in order to argue that they have a “prima facie moral standing” which constructed (read, “coercive”) orders lack:
The attraction of spontaneity is both moral and prudential. Though it is not clear whether Hayek saw more than instrumental value in it, he stressed that the elements in a spontaneous order “arrange themselves” rather than being arranged by “unified direction” (1960, p. 160). When the elements are human beings, their property and their choices, nobody’s dispositions are imposed on him [sic] by another’s command. Everybody chooses for himself [sic] what seems to him [sic] the best, given that everybody else chooses likewise. All choices are interdependent, and made mutually compatible by property rights and their voluntary exchanges. None dominates and none is subordinated. This lends the order in question a moral laissez passer, while nonspontaneous orders, constructed by imposing some alternative on the participants by authority or the threat of force, are morally handicapped by their coercive element. If they are to pass for legitimate, they need to show some compensating merit. Spontaneous social orders, in other words, have a prima facie moral standing. Constructed orders must first earn it, or do without. (125126, emphasis added)
But in the following paragraph, de Jasay argues for the “prudential attraction” of spontaneous orders by referring to Hayek’s knowledge problem for constructed orders—to the unique ability of polycentric orders to gather “irretrievably dispersed or latent” knowledge, and so surpass the epistemic limitations of planners, which inevitably hobble the ability to scale up directive orders:
The prudential attraction of spontaneous orders springs from the belief, strongly held by Hayek and fairly well supported by historical evidence, that since the knowledge required for successfully designing a complex order is either irretrievably dispersed or latent or both, the constructed order runs a high risk of being inefficient if not grossly counterproductive. (126)
These two paragraphs are followed by several pages of agonizing over the apparent difficulty that Hayek’s understanding of his own prudential arguments seems to depend on the deliberate enforcement of rules by an authoritative agency; but de Jasay makes no clear distinction, in his worries about “enforced enforcement” mechanisms, between (1) the coercive features of such mechanisms (as opposed to “voluntary conventions” like ostracism they are based on punitive force and paid for with tax levies), and (2) their directive character (as opposed to participatory social sanctions, which are implemented, unprompted, by ordinary people, “enforced enforcement” comes from authoritative orders and “exogenous sanctions” from an “ultimate, sovereign enforcer”). Although each of these arguments depends on a distinct sense of “spontaneity,” de Jasay uses the term “spontaneous order” throughout, without noticing that each argument turns on a distinct and conceptually separable characteristic, and that social orders may be “spontaneous” in any of these senses, with or without being “spontaneous” in the others.24But of course they can. These are simply three different distinctions, and while many examples overlap, the features may come apart even in some of the paradigmatic cases of “spontaneous order.” When barterers converge on a highly marketable commodity as the common media of exchange, they settle on it through iterated consensual exchanges; they also make the exchanges in dispersed interactions without any directing center. Market price adjustments produce emergent patterns that few or none of the individual participants could or did plan for (whether the feeding of Paris or the manufacture of a humble number 2 pencil); these emergent orders are possible because of the harnessing of otherwise irretrievably dispersed knowledge through a polycentric network of consensual exchanges. But orders may be consensual while being (voluntarily) directive; and orders may be polycentric while being to some substantial extent designed. Thus, for example, while the development of Wikipedia is a clear example of a consensual and participatory order — nobody’s forced to contribute; the editing process is wide open to anyone who wants to jump in without waiting for instructions — it is not a strong example of an emergent order. Typically, people edit Wikipedia, at least in part, with the intention of improving the breadth and accuracy of information on Wikipedia — that is, the macroscale success of Wikipedia is a consciouslyadopted part of the microlevel intention.
The second highlevel point to be made about the notion of spontaneous order is related to the point we have made about the equivocal meaning of “spontaneity.” Because the paradigms for applying the concept have generally been cases where a social order is participatory or unplanned, and is also consensual, what libertarians have tended to see in spontaneous order is almost always a macroscale pattern that is freely chosen and where all involved find some mutual benefit from the proceedings. You don’t need topdown command and control to get many important things that libertarians like to use — language, or money, or roads, or Wikipedia…. So, in libertarian vocabulary “spontaneous order” is almost always employed to praise benign orders, especially benign orders that spontaneously accomplish something that government planners cannot do as well, or at all. It is remarkable and wonderful that bottomup forms of social cooperation can so often produce unplanned largescale social outcomes better than could be managed through comprehensive, consciously designed political schemes for social coordination. But nothing conceptually requires that emergent orders need be benign orders. If widely distributed forms of intelligence, knowledge, virtue, or prudence can add up, through many individual selfinterested actions, into a benign undesigned order, then there’s no reason why widely distributed forms of ignorance, prejudice, folly or vice might not add up, through many individual selfinterested actions, into an unintended, malign order.25 So might widelydistributed, microlevel practices ofviolence; since libertarians are centrally concerned with individual freedom from violence and coercion, the possibility our threefold distinction raises of an emergent but nonconsensual order must surely give us pause.