“The recognition of the insuperable limits to his knowledge ought indeed to teach the student of society a lesson of humility which should guard him against becoming an accomplice in men’s fatal striving to control society — a striving which makes him not only a tyrant over his fellows, but which may well make him the destroyer of a civilization which no brain has designed but which has grown from the free efforts of millions of individuals.”
— Friedrich A. Hayek, “The Pretense of Knowledge.”
A strong libertarian case against centralized interventionism is that market orders are complex, too complex for a single mind to understand in its entirety, and therefore that interventionism is likely to lead to unintended consequences. But, few libertarians apply this same logic to the institutions of governance. Instead, many prefer to take a “hyper-rationalist” approach, believing they are capable of at least designing a pattern of governance, such as anarchy or “the minimal state,” that they think the world should move towards. Are these libertarians suffering from a fatal conceit?
I have made this point before. In “What is Limited Government,” I argue that an optimally limited government — that is, one limited to its comparative advantages — is probably not going to be the one envisioned by most minimal state libertarians (i.e. constrained to protecting property rights). Instead, we would probably see a complex bureaucracy providing for a set of public goods, where the specific elements of the set are contingent on the facts of that particular moment of reality. For example, in one moment the state may have the lowest opportunity cost in providing for roads, whereas in the next it may be the institutions of the market which better provide for roads. After making my principal argument, I posed a challenge: do libertarians suffer a fatal conceit? I want to expand on that challenge, and hopefully draw attention to it.
The institutions of governance arise out of a process of spontaneous order. It’s easy for us to lose sight of that, because, on the surface, the truth seems to be the opposite. Governments are toppled and they are replaced. The institutions that are most visible, such as the division of power at the Federal level, are planned in the sense that their organization is oftentimes deliberate. The United States’ bicameral legislature, for example, was not an accident. Rather, it was designed. Thus, it is not uncommon to see a dichotomy created between markets and government, where the former are “natural” — complex and spontaneous — and the latter are “artificial.” But, taken as a whole, institutions of governance are products of human action, but not of human design.
Even the most “obviously” planned features, such as the division of power in U.S. Congress, or between the executive and the judicial branches, are not so “obviously” designed. This is analogous to the argument that while some of our laws may be codified — hinting towards design —, the truth is that these rules were codified only once they spontaneously developed to a point where humans could communicate them (Hayek ). In the case of our institutions of governance, democratic orders existed prior to the late 19th century; and, they probably existed before they were adopted at the Federal, or national, level. Democratic governance developed at a local level and once a basic democratic system of rules was understood enough to be communicated, the system was implemented on a larger scale. Of course, even institutions at the largest scale continue to develop, even without a plan, after they are created. Spanish democracy, for instance, is probably much more complex today than any of the politicians responsible for it originally intended. And, to a significant degree, “political planners” are aware that some of the holes in their plan will have to filled on their own, or through experience.
One piece of evidence, although I admit that the history of governance is not my area of expertise, is the difference in outcomes between what we can, for lack of alternative terms, call “bottom-up” and “top-down” democracies. Namely, compare the differences in welfare between democratic society in the United States and much of Western Europe and “democratic” societies in the developing world. Democratic institutions are much weaker in, for example, Venezuela and Ecuador than they are in the United States or Great Britain. Venezuela did not go through the same process of political development that did the U.S.; democracy must filter down to the people in Venezuela — who have little experience with local democratic institutions —, rather than up. Even in parts of developed Europe, democratic institutions are not as strong in some countries (e.g. Greece and Spain) as they are in others, because they were largely imposed on local populations, rather than adopted from them. Consider, also, how difficult, to the point of being impossible, it is to build stable democracies in countries we invade.
Even where institutions are supposedly planned, they are often too complex for many or most people to grasp in their entirety. Take James Buchanan’s and Gordon Tullock’s The Calculus of Consent. Buchanan and Tullock, after providing a basic theoretical structure (of minimizing the joint costs of political externalities — when governance is less than perfectly plural — and decision-making — which increase as governments approach perfect pluralism), go on to describe some essentially basic features of U.S. representative democracy. They use game theory to analyze bicameral legislatures, finding that by establishing two houses to a congress, where (ideally) the voting base for each member of each house is different, a representative democracy can increase the number of people it includes, while decreasing decision-making costs. They show how vote trading can increase the efficiency of the democratic process, and how it leads to, at least, Kaldor-Hicks optimal outcomes. These features, however basic to our democracy they are, are non-obvious, and even Buchanan and Tullock must simplify. Now, imagine how complicated our governing system — not just at the Federal level, but also at local levels — is, and how difficult it is to understand as a whole. Knowing it in all its detail is probably impossible.
Theories of ideal democratic government are built on consent. Buchanan and Tullock (1962) imagine perfect pluralism at the time of deciding on a set of rules (a constitution) that governs decision-making (which may be less than perfectly plural). John Rawls (1971) uses the conceptual tool of the original position to argue that all members of society would choose a set of institutions that maximize the position of society’s worst off. Public reason political philosophers imagine a world where reasonable people can debate with each other, compromising over political differences. These are all abstract ideal types of how consensual government works, but none of them are actual representations of reality. The real world is more complex, and consent, especially explicit consent, amongst all members of a society is unlikely.
But, does complexity rule out implicit consent? If we were to understand the current system, and compare it to all other known systems, would we choose to dismantle and replace it? I don’t think so. And, I think that bottom-up, developed democracies, like that of the U.S., are closer to perfect pluralism than many libertarians realize. We feel that this is not the case, because we are so detached from the actual decision-making process. The division of labor is not only relevant for institutions of markets, but also institutions of governance. Those who make political decisions specialize, and the rest of us specialize in our own fields, and good political institutions accurately communicate this specialized, localized knowledge, including our preferences, between voters and representatives. In the same way that decentralized markets can harmonize interaction between heterogeneous and otherwise unconnected agents, so can decentralized political processes.
Do we prefer a system of explicit consent to one of implicit consent? Imagine how small of a community would be necessary for institutions of governance that are perfectly understood by members of that society? The larger the body of people who must be coordinated, the more complex these institutions and processes of coordination. If we were to choose institutions that we explicitly consent to, because they are simple enough for us to understand in their totality, we would have to accept living on a much lower indifference curve. That is, that society would be much poorer than larger, more complex societies. Note, this is the same problem that exists with planned economic orders. Planned economies are much poorer than non-planned economies, precisely because they do not benefit from the same level of complexity and harmonization. Why do we not extend this logic to governance?
An anarchist might respond by pointing out that what s(he) wants is not a planned society, but one where individuals are free to opt-out, or disassociate, so that they can form their own communities. But, anarchist societies are not libertarian societies. The world, taken as a whole, lives in a state of anarchy. There is no strong, binding level of political order above national governments, excepting limited international governing bodies, such as the European Union. This has been even more true historically. Yet, out of this anarchy rose territorially defined nations. The idea that we can wipe the slate clean, so to speak, and allow for small, local communities of largely homogenous agents is not just utopian, but completely detached from the real world. On some level, we already live in the “ideal” anarchic society.
Oh, but “true” anarchy would be governed by some set of meta-rules, such as Robert Nozick’s (1974) utopia or Murray Rothbard’s (1998) ruling ethic. But, remember, in a world of ethical and moral heterogeneity, these meta-rules are imposed (thus the attraction of public reason), and suffer from the same criticism of planning as do governments and markets.
Maybe my problem is with accounts of “what ought to be.” As orders become more complex, our minds become less able to understand these orders, and theories of “what ought to be” begin to suffer from problems of hyper-rationalism. What we think ought to be, once implemented, may turn out to be entirely dystopian, as is the case with socialism. We want to replace what we perceive — and undoubtedly is — a flawed world with an alternative that, in our models, corrects many of the flaws we perceive. But, maybe the real world is gradually moving closer to what we want, through a process of competitive planning — what spontaneous order really is (and distinctions of what is local and what is not tend to be arbitrary) —, and our grand, all-society-encompassing planning can only move us further away. Libertarians should drop their pretense of knowledge, their fatal conceit.