2.2 Polycentrism
Although for simplicity’s sake it is useful to analyze the idea of a public moral constitution for a society, the moral constitution of the Open Society is actually characterized by a variety of sets of rules, regulating different areas of social life, different types of problems, over different areas. And often the same society will be characterized by competing sets of rules, followed by different parts of the population (see further §IV.4.2). As I have emphasized, the rules of our shared social life are tools to address a variety of shared problems; different problems require different tools, and their effective solutions require that they range over groups of varying sizes. The group that shares a problem, and so requires a rule to assist in its solution, will vary from problem to problem. The important work of Elinor and Vincent Ostrom, drawing on decades of research in a wide variety of social contexts, has shown how often informal social rules with an attendant practice of responsibility can be effective for a particular population to address its specific problems.85
Understanding a society as composed of a wide variety of social networks, addressing different problems at different levels in different networks, yields two insights into the institutions of the Open Society. First, as different people come together to solve their particular problems (say, about the use of a certain water supply or safety in their neighborhood), their shared concerns in solving this problem provide a robust basis for wide agreement on these particular rules or institutions. Even though their perspectives may be deeply diverse, they forge common small public worlds and common classifications when, as an interacting group, they confront problems that all participants appreciate. An atheist and an Evangelical may share neighborhood watch duties. In cases such as these, the very nature of the local problem helps produce convergence on common ways to contend with it. Those who would say that our differences are so great we cannot share any commonly endorsed rules conceive of the problem too abstractly. When we face commonly perceived problems that require coordinated action, those with deeply diverse outlooks are apt to devise and abide by common rules.
Second, for many matters within the same society, different social networks may gravitate to competing rules: witness the social networks of vegetarians, religious citizens, feminists, and libertarians. As we live among each other, we simultaneously live in different, and often competing, small public worlds: the common classifications of feminist social networks are distinctly different than the small worlds of libertarian networks (even though some may be members of both).86 Each cluster of perspectives forges its own common world—which is inherently richer and deeper than our general public world—while also offering to the public at large competing small social worlds, often seeking to attract other perspectives into its networks. In this way, the competition between the networks can be seen as ways to search for thicker, more substantive rules that might, from the bottom up, be taken up by larger numbers in the society. Recall the difficulties with the idea of small-scale social experiments (§II.4.1). Given the small number of such experiments on any specific problem, it is always difficult to determine the causes of the experiment’s outcome, and so whether its lessons can be transferred to other contexts. In contrast, in a polycentric system, competing networks can explore, in the same social circumstances, different rules; innovations can spread throughout a population as different perspectives take up what they see as a better social rule, perhaps displacing ones that are less suited to the group’s participant perspectives.87 This points to a model of normative change—and from the perspective of the group, normative improvement—in the Open Society.
Like most insights—that the moral constitution of a society may be transformed by gradual switching to a better, competing social rule—this one can enlighten us about much, but we must also recognize its limits. Many parts of the moral constitution have been changed by this process: the astounding changes that have occurred in Western countries over the last fifty years regarding rules regulating sexual relations is a striking example. However, other social rules require something like a simultaneous change of belief and/or attitudes among members of a social network. If a rule is held in place by a network of beliefs about what others expect one to do (for example, a norm among prison guards that all guards expect other guards to beat inmates and will ostracize one who does not), individuals will be reluctant to depart from what they see as firm social expectations. Here collective reasoning and/or some sort of legal intervention may well be critical (see §IV.2.6).88 There is no single analysis that accounts for all the dynamics of rule change in the Open Society. Moreover, we should not forget that, as the Ostroms’ analyses of polycentrism stressed, these diverse networks of organization must be united under global common rules when their small worlds collide.89 Still, the polycentric insight is that we should not make the mistake of equating the entire moral constitution with this overarching framework that applies to all.
2.3 Liberty, Prohibitions, and Searching
2.3.1 The Principle of Natural Liberty. It is a platitude that liberty is fundamental to the moral constitution of the Open Society: after the work of Mill, Popper, and Hayek, what more can be said? As Mill in particular so decisively demonstrated, an open, dynamic society must be based on freedom of thought and speech. If different perspectives are to communicate their insights, and hopefully enrich each other, an open society must secure the conditions for freedom of thought and ensure that the channels of communication are kept open. The Open Society is not a mere standoff or compromise between opposing perspectives, but a forum in which, while ignoring or even disparaging some, a perspective can gain from others. And just which others have something to teach one—which of their discoveries are discoveries one can make use of—is itself always a matter of discovery.
All this is as familiar as it is fundamental. There is no point in once again rehearsing these important insights. I believe, however, that we can say something rather less familiar yet enlightening about the place of freedom in the Open Society.
Rawls notes that although there are, in principle, an indefinite or infinite number of possible moral (or legal) rules, a moral (or legal) conception that seeks to guide behavior supposes a principle of closure: given such a closure rule, the system of rules can be complete and provide a full guide to behavior.90 Drawing on this idea, John Mikhail identifies one closure rule as:
The Principle of Natural Liberty: Whatever is not prohibited (and this includes the non-performance of specific acts) is permitted.91
On this closure rule, Alf consults the system of rules and determines whether his ϕ-ing is prohibited by some rule in the system; if it is not, then he is free to ϕ. This closure rule is intimately related to the standard formulations of the Principle of Natural Liberty that are to be found in liberal writings.92 For example, Stanley Benn’s grounding principle of morality is that those who are simply acting as they see fit are under no standing obligation to justify their actions to others (while those who interfere with others’ actions are under an obligation to justify their interference).93 As Benn says, “justifications and excuses presume at least prima facie fault, a charge to be rebutted.”94 If Alf has no justificatory burden he is permitted to act without justification—Alf has no charge to rebut, no case to answer. Thus on Benn’s view Alf is free to act until, as it were, he runs into a moral rule that regulates what he is doing. The moral code does not having a standing requirement that Alf be able to identify a permission to ϕ; it only requires that he not ϕ when so doing is prohibited (a requirement to ϕ can be understood as a prohibition on failing to ϕ). Thus, in the interpersonal context, Betty is not warranted, as it were, to call him out for ϕ-ing unless she can point to a prohibition that he has flouted. Benn’s point is that merely acting does not itself invoke the context of justification, because, according to the Principle of Natural Liberty, Alf need not, as a matter of course, cite a permission to act before acting; the context of prima facie fault arises only when Betty can point to
a prohibition that his ϕ-ing appears to flout.
The Principle of Natural Liberty implies a basic asymmetry between action and objection to action.95 It is clearly unnecessary that, as a matter of course, Alf possesses a justification for ϕ-ing in the sense of showing that morality allows it; however, the claim that he should be stopped from ϕ-ing because it runs afoul of morality implies that there is a prohibition such that ϕ-ing is prohibited, and so the rule must be identified. Thus we can get from the closure principle to a moral objection principle:
The Interpersonal Principle of Natural Liberty: Alf need not possess a moral justification for ϕ-ing, unless ϕ-ing appears to run afoul of a prohibition. A morality-based objection to Alf’s ϕ-ing thus must be based on a justification of the form “Rule R prohibits ϕ-ing.”
This, I am convinced, was Benn’s basic point: it is Betty, the moral objector, not Alf, the actor, who has the onus, as a first move in the conversation, to cite a morally relevant reason for her stance. It is in this sense that in a system of natural liberty there is indeed a basic moral asymmetry in favor of voluntary action over moral objections to it.
Thus understood, the core of Natural Liberty is a closure principle. Now logically, a closure principle is not required for a system of rules. If all action types were specified, then theoretically each type could be binned into the prohibited or the permitted. Depending on the number of action types and the capacity of the learner, this list could be internalized by rote. In the case of human moral rule systems, however, it seems that the list of action types is vast and cognitive resources limited.96 Real human reasoners can learn only a relatively modest set of the moral rules, and they do not possess the cognitive resources to infer all the possible applications of the set of rules. Without a closure rule, humans deliberating whether to ϕ may simply be unable to determine the status of ϕ-ing. Such a system would lack decisiveness. A decisive system of rules, for any action ϕ, always allows, requires, or prohibits ϕ (with the “or” being exclusive). Formally, decisiveness is a consistency condition—for any action ϕ, the moral system is single functioned, providing one and only one answer.97 If the moral system is not single functioned in this way, it is in principle impossible that a person can always comply with it. This is obvious if the system fails to be single functioned because, for some action ϕ, it declares that in the same circumstances ϕ is both prohibited and required. Of course the system may yield “prima facie” or “pro tanto” contradictory judgments, which need to be weighed, and, as a practical matter, it may be very difficult to do the weighing. But these complexities are all consistent with the requirement that the system is not, in principle and all things considered, “overcomplete”:98 that is, inherently inconsistent. Another failure of decisiveness, undercompleteness, may seem less worrying. Here, in a context in which ϕ is a feasible option, the system does not generate any of the judgments: ϕ is required, ϕ is prohibited, or ϕ is permitted. Such undercompleteness also makes it impossible to comply with the system: if the system is undercomplete among “ϕ is permitted, ϕ is required, ϕ is prohibited,” one cannot always act in accordance with these rules, for the rules do not entail any instructions. One cannot go ahead and ϕ in this case, for that would be to suppose that ϕ is permitted, and that is what the system is incomplete about. And one cannot simply refrain from ϕ-ing, for the system is incomplete about whether ϕ is required.
As we have seen, a critical task of the moral constitution is to secure shared empirical and normative expectations within a group that allows coordination of behavior. To achieve coordination in cooperative contexts the rules must guide our expectations, so that one can anticipate whether a person must, must not, or may ϕ, and there is also a shared understanding of when one may demand that another ϕs (or refrain) and when one appropriately blames others for what they have done. A moral constitution that lacks decisiveness will thus to some extent fail to perform the very job that we require of it. Of course, every set of rules falls short of decisiveness in important ways: in some contexts the rules may by unclear, and perhaps multiple rules that imply inconsistent directives may apply to a single case. We shall see in §IV.4 that such imperfections can actually help make a system of rules more flexible. Nevertheless, the system of rules that makes up the moral constitution must generally be decisive if it is to play its expected role in coordinating the moral life of a diverse society. Note that decisiveness and closure rules are supposed by plausible systems deontic logic: “ϕ is not prohibited & ϕ is not required” imply “ϕ is permitted.”
2.3.2 The Residue Prohibition Principle. So (i) the presumption in favor of Natural Liberty can be interpreted as a closure principle, and (ii) a decisive social morality requires a closure principle. But it is not the only closure principle. Many appear to reject the Principle of Natural Liberty. Ranier Forst, for one, explicitly does so, insisting that only plausible “presumption” in favor of liberty is one of “equally justifiable liberty.”99 Contrasting to the Principle of Natural Liberty is:
All Liberal Liberties Are Specifically Justified: In a liberal society Alf only has a liberty to ϕ if he can provide a justification to all other persons.
This would capture Forst’s insistence that a person has a liberty only when this is an equally justifiable liberty. The All Liberal Liberties Are Specifically Justified Principle has a weak and a strong reading. On the weak reading this principle is very close to the Principle of Natural Liberty supposing we add a proviso: the Principle of Natural Liberty, as a closure rule, is itself a justified part of the moral system. The Principle of Natural Liberty would then be a justified closure rule, and Alf’s ϕ-ing is justified, as it is an action of his falling under this blanket justification. On this interpretation we might say “one is always justified in doing something that is not prohibited.” This, as it were, mimics the Principle of Natural Liberty, but it is not equivalent to Benn’s Interpersonal Principle of Natural Liberty, since Betty can indeed always challenge Alf, but he would always have an answer on the tip-of-his-tongue, that he has a justified presumption in favor of his action.
The stronger, and I think more interesting, interpretation is that Alf must always be able to show that ϕ-ing comes under some specific rule or principle that accords him a liberty to ϕ. Some have argued that a liberal society is not based on a general commitment to liberty, but to a justification of specific liberties that have value to us.100 We could, perhaps, have a liberal society that strongly values freedom in the form of a public moral constitution comprising an extensive list of permissions—a system of moral rules according to which one is permitted to act on one’s perspective, is permitted to do with one’s property as one pleases, is permitted to deliberate with others about superior answers to common questions, and so on. If this were our liberal public constitution one would be free to act if and only if one could cite a liberty (i.e., permission) to do so. Now as Mikhail notes, this would imply an alternative closure rule:
Residual Prohibition Principle: Whatever is not permitted is prohibited.
If this is the closure rule, a person is prohibited from ϕ-ing unless the system of rules permits ϕ-ing. Here we would see the opposite of the interpersonal dynamic that Benn points to: Alf is going to ϕ, and so Betty could quite intelligibly claim that he must be able to cite a permission to if he is not to act wrongfully. In an obvious sense there is always an onus on a rule-following actor to cite a permission before acting, to show that the Residual Prohibition Principle does not apply.
2.3.3 Moral Learning and the Two Closure Principles. Do people actually employ such closure rules when they act on deontic systems? Under what conditions do learners employ one or the other? Do they have to be taught, as one of the rules in a system, or do rule learners close the system themselves? In a recent series of experiments Shaun Nichols and I have explored these questions and have found that learners do indeed infer closure rules.101 In one experiment, for example, participants were told:
There is a Farm w
ith squeaky mice, and all the mice are supposed to follow The Rules of Mice, written in a book. The Farm has four barns: Red, Blue, Yellow and Green.102
Some participants were given two prohibition rules:
The Tyranny of the Ideal Page 24