On Schmitt, Vermeule, and Liberal Faith

A year and a half ago (April 2022) I reiterated a point that I attributed to Jill Gordon and that should be a truism for any careful reader of On Liberty: that J.S. Mill did not defend the idea that the market place of ideas would inevitably lead to truth or social harmony. I observed that this leaves puzzling how the phrase and idea came to be associated with his name and a kind of (providential) liberal faith. I then went on to trace the origin of this idea to Carl Schmitt, who holds it at arm’s length, but seems to offer it as a sincere analysis of Liberalism’s self-understanding.

It turns out I was not the first to notice this. Mea culpa! A few years earlier, in Spring 2019, Adrian Vermeule published an essay, “Liberalism and the Invisible Hand” in American Affairs Journal, 1 (Spring): 172–97. He shows that in a number of works, Schmitt offered the same analysis of such liberal faith in what Vermeule calls ‘invisible hand systems’ (or IHS).

Unlike me, Vermeule thinks, alas, that Schmitt gets liberalism right. In fact, on Vermeule’s account (and this seems very plausible to me), Schmitt thinks that liberalism is non-trivially constituted by this liberal faith in invisible hand systems.

According to Vermeule,

The key hallmarks or notes of liberalism’s invisible hand systems are these: (1) individual agents engage in decentralized interactions, often but not necessarily competitive, such that (2) “as a result of human action, not of human design” (3) some system-level good or goods emerges. Condition (2) implies that the emergence of the system-level good(s) need not be part of the intentions of any of the individual agents, who may be pursuing strictly self-interested aims. Condition (3), on the other hand, does not necessarily imply that the goods that emerge are the same self-interested goods for which individuals were aiming. Liberalism does not promise that participants in its systems will attain the desires of their hearts, only that the quest for individual goods will cause aggregate—but not genuinely common—goods to emerge.

Before I get to my substantive disagreements with Vermeule, first a terminological point. Vermeule repeatedly implies that this is Smith’s view of the invisible hand. As I document in chapter 10 of my 2017 book (which has an obsession with systems in Smith), this is not what Adam Smith meant by the phrase ‘invisible hand.’ Smith reserved that phrase for short-term unintended outcome processes where the agent should/could have known better. In order to avoid confusion, when I am discussing Vermeule’s position, I will use IHS.

Having said that, I do not mean to deny that Smith also articulated something like processes captured by Vermeule’s (1-3); in fact I anticipate Vermeule’s analysis of IHS and call it Smithian Social Explanations. In Smith the IHS is an analytical tool in what we may call his social theory. (It’s not his model with which he tried to develop empirical knowledge.)

But even in Smith’s social theory it is worth noting that Smith thought that during an IHS process individuals would eventually cotton on and if they benefitted from it, this would produce or entrench a kind of lock-in. (The idea is not original to Smith, we see it in Hume’s account of the origin and continuation of justice, which relies heavily, perhaps too heavily on this idea.) In Smith’s version of a IHS, it’s not like all the action occurs behind people’s backs.

I actually think Vermeule agrees with my last sentence. For, Vermeule allows that hybrids of (1-3) are possible, and he has interesting things to say about them. But because he sets them aside, so will I. So far so good. (It also follows that for Smith, people can defect from a IHS process.)

Now, Schmitt and Vermeule agree that according to the liberal conception the emergent goods that follow from an IHS process is good (harmony, truth, etc.). For Vermeule, this is, in fact, key to liberalism’s self-understanding:

Having made both obedience to the divine and public-spiritedness into matters of private choice, however, liberalism must then find arrangements for pursuing social aims that do not constrain the beliefs or motivations of autonomous individuals in those ways, arrangements that are robust to self-interested individual behavior and, by a kind of substitute miracle, transform those selfish motivations into social goods. (emphasis in Vermeule)

At this point, I want to make a historical observation. That Schmitt understands liberalism this way is not entirely surprising, although what he says about Condorcet and continental liberalism in Political Theology is weird. As I have noted before, the view originates in a kind of satirical (and misleading) criticism of Smith. Smith himself thought that IHS could lead to defective outcomes if they were not well-governed. (I’ll cite the passage below.)

But it’s true that in the late nineteenth century, roughly from 1850 to the 1890s, there were influential liberals who did think this way. Frédéric Bastiat and T.H. Greene are key examples, and if one treats Bosanquet as a liberal he surely counts, too. In general, liberals who endorse something like the substitute miracle are influenced by Hegel, Rousseau (or both) or some version of traditional (unity from diversity) republicanism. Hobhouse’s The Metaphysical Theory of the State is a lovely take-down of it.

But again, while Mill sometimes sounds like he believes in the substitute miracle, he certainly did not think in an untrammeled market-place in ideas leading to truth. That’s just a bizarre reading of On Liberty. I actually think Vermeule agrees with me on this because he carefully hedges his bets, “In the empirical version, usually associated with John Stuart Mill, the claim is that truth tends to emerge on average and in the long run from a suitably open process of discussion.” (emphasis added) What Mill does seem to think — as Vermeule notes Schmitt satirizes it in Crisis of Parliamentary Democracy — ordered (parliamentary) discussion will generally lead to better national decision-making if lots of other constraints are in place—read Selinger’s book on Parliamentarism. (This may fall under what Vermeule calls a ‘hybrid’ IHS.)

Interestingly enough in Schmitt and in Vermeule, the IHS process gets conflated with another purported liberal commitment, that is, “Checks and balances—international and constitutional.” Vermeule sets the international case aside, and focuses on the internal one, that is, “the balance of constitutional powers.” That’s a shame because he would have immediately noticed a problem.

The international balance of power is, however, not a liberal commitment. (It’s a realist one.) As Foucault makes clear in The Birth of Biopolitics (and I am pretty confident he had Schmitt’s mistake in mind), the international liberal project (of federalism [in Kant and Bentham] and parliamentary union [in Smith]) is an attack on the balance of power doctrine. While surely, again, some Victorian and cold war liberals have availed themselves of balance of power analysis and strategy instrumentally, it was rejected by people like Cobden and Bright who held on to the original liberal faith in arbitration, federalism, and peace.

It is worth reiterating that commitment to federalism (national and international) is integral to liberalism, which from the start understands mercantile imperialism and violence as its true enemy. Hume (in “Idea Of a Perfect Commonwealth”) and Smith (in book 4 and the closing pages of Wealth of Nations) argue for it independently of their other other commitments. [I mention this because Vermeule treats Madison’s embrace of federalism as an extension of Smith’s argument for religious dis-establishment (which Vermeule treats as an instance of an IHS argument because moderation is supposed to follow from it). This is astonishingly weird history of federalism.]

As I noted yesterday, and amply documented by William Selinger in his excellent book on Parliamentarism (which is respectful of Schmitt), informed readers immediately recognized that Montesquieu’s account of the balance of powers (which presumably was derived from Bolingbroke) is highly misleading as a description of the actual English system. And inspired by Selinger, I explored some of the ways Hume (and Smith) criticized it. As Selinger shows, nineteenth continental century liberals viewed a system of checks and balances as an imperfect and dangerous alternative to what he calls parliamentarism. My own view is that the very idea of ‘checks and balances’ is a-liberal. But I recognize that it has become associated with liberalism, and I will not re-litigate the point here.

Now, at this point a reader may suspect that all I am doing is nit-picking. So let me get to the heart of the matter. Vermeule thinks that liberals defend IHS in virtue of the higher order system-level or emergent goods and because liberal can thereby square their desire to avoid imposing on individuals yet still pursue indirectly their desired collective goods. And if liberals did believe this, they are open to fatal objections:

The mechanisms adduced to yield the system-level emergence of the relevant goods prove—for systematic and not merely happenstance reasons—far more fragile than their liberal proponents suggest. Yet the need for general justifications for liberalism makes admission of this radical fragility threatening. The consequence is a marked tendency to liberal fideism, to ever-more fervent insistence on the scope and power of the liberal invisible hand, even in the face of mounting evidence to the contrary.

Notice that on Vermeule’s account the defense or a commitment to IHS within liberalism is from a normative perspective consequentialist in character. And if IHS is intrinsic to liberalism then, indeed, if these emergent/indirect consequences are not remotely reliably produced by IHS, then liberalism is in trouble. As my initial remarks on Mill suggest, I don’t think individual liberty was ever defended in terms of purported truth it would produce.

Now, as my wording above suggests, and against the implied wording of Vermeule throughout his essay, I don’t think Adam Smith ever defends the IHS in this normative way. (And as I note elsewhere, Smith criticizes Hume’s account of justice, which does defend an IHS like that, as too refined and philosophical.) But I do agree that especially influenced by some of Hayek’s remarks some liberals defend IHS in the quasi-providentialist and as the Marxist would say, ‘naturalized’ fashion Vermeule describes. So, I think it’s fair that Vermeule focuses on Hayek, although it is weird that he turns Hayek into an informational general equilibrium theorist. (This feels a bit like using hocus pocus to confuse uninformed readers.) But let me leave it to the Hayekians to defend Hayek (who is misrepresented by Vermeule).

The important point is this: there is no need for liberals to defend the deployment of IHS in this way. Recall that part of the intended function of the IHS is to let individuals do their own thing constrained by a harm principle (etc.). (Vermeule grants this.) That just is the normative defense of IHS. But that can’t be the whole story.

As I noted above (and this point is already explicit in Locke of all people), before an emergent high order by-product of an IHS gets locked in as a kind of robust emergent phenomenon/good individuals notice such effects, too. Nudge-ers excepted, liberals do not treat humans as unthinking molecules. During the development of a possible IHS process, individuals can then still individually or through talk collectively decide to (use game theory) to defect from or reject the IHS. (When we’re dealing with an imposed IHS this is called ‘civil disobedience.’)

Now, obviously, the decision to step back from or derail an emerging IHS process involves coordination and opportunity costs. So, if the first order benefits of de-centralized interactions within an emerging IHS are very high, it’s possible that individuals in society will accept (what are known as negative externalities or tragedy of the commons) that follow from the IHS.

Vermeule may grudgingly grant the last two paragraphs, but still feel that I have missed his point. For now it looks like IHS as a whole are without normative defense.

Before I address it, I reiterate that the liberal defense of IHS is not about the emergent propertym its about individuals pursuing their own ends (including the right to make mistakes). If there are no enduring (what Vermeule calls) ‘self-benefits’ to participants, the liberal has no interest in defending a IHS. (Somehow, Vermeule has lost track of this fact.) This is why liberals routinely have second thoughts about legalized prostitution, gambling, and even markets in alcohol and tobacco.

Even so, there are consequentialist liberal ways of defending IHS that Vermeule ignores. Here are three (although one I slipped in already before):

First, alternatives to some IHS have far worse emergent properties. I actually think this is Mill’s defense of a fairly wide scope for unregulated free speech in On Liberty. You don’t want to ban what he calls vituperative speech outright because one may well end up with an absence of political freedom or other bad political side effects. Let’s call this the liberalism of fear defense. (This usually carries the day as Mandeville already noted about prostitution in Amsterdam.)

Second, an IHS is a mechanism to combat concentrated power and prevent monopoly agents. I think this is, for example, the ordo-liberal defense of free-markets (that, historically, is a response to Schmitt). Let’s call this ORDO inspired defense, “the dispersed power” defense. An IHS is all about the interactions of dispersed agents.

I should say that neither defense is full-proof. Vermeule quite rightly notes that sometimes IHS produce consequences that destabilize their own functioning and that then run into what he calls the The public choice problem (that is, “agents within the market framework will have every incentive to undermine or subvert it, through self-interested action aimed at manipulating the framework itself.”) I agree with his diagnosis and I would also add that it gives the appearance of permanent liberal crisis. (The preference for thoroughgoing federalism is originally, in part, meant to defuse the public choice problem, but it does not eliminate it.) As my regular readers know, the current liberal inability to respond to this problem is due to a kind of anti-politics that has crept into twentieth century liberalism (recall here; and here).* My own current work on the liberal art of government is prompted by it (and see below).

Third, and to reiterate, an IHS may produce emergent properties that turn out to be welcome. Nobody really imagined the technological and communication revolution that would followed the break-up of the ATT/Bell labs monopoly and that allows you to read my exchange with Vermeule at your own convenience far away from any physical library. This actually is more important than you think (let me explain).

Let me wrap up with two related points. First, throughout his essay, Vermeule pretends as if liberals promote an IHS as an indirect way to pursue the common good. He is right that historically most liberals were indirect perfectionists. (And I like that kind of liberalism myself.)

So, for example, at the start of Wealth of Nations, Smith writes, “It is the great multiplication of the productions of all the different arts, in consequence of the division of labour, which occasions, in a well governed society, that universal opulence which extends itself to the lowest ranks of the people.” (emphasis added.) It is pretty clear that Smith here defends an emergent property that is the effect of an IHS. But notice, too, that for Smith — and liberals as such — to secure these emergent benefits also requires much wider structure: society needs to be well governed. This suggests that in so far providence plays any role, its human agency at the institutional and political level that does a lot of the hard work by liberalism’s own lights.

Second, Vermeule kind of implies that the common good that liberals pursue indirectly through an IHS is to be thought of similar to or a substitute of the common good of the sort that Aristotelians, Platonists, communitarians, and Integralists pursue. (See here: “Those goods, according to liberal doctrine, will not be directly secured in virtue of authoritative commands, in virtue of ordinances of reason for the common good promulgated by one in authority, but instead will emerge indirectly, at the systemic level, out of the freely chosen acts of individuals themselves.”) So, it looks like liberals merely disagree with other approaches over the means by which the common good is pursued. And then the liberals are rather naïve to place their faith in a very imperfect mechanism.

What’s amazing to me is that somebody can spend a whole essay making fun of Hayek and Hayekian defenses of IHS and miss the cardinal insight of Hayek that was also articulated by Frank Knight and Walter Lippmann: that whatever common good, if any, is produced by an IHS it will be rather surprising. It is something that is invented by the process. It simply cannot be the kind of good that is a “genuinely common good” by the lights of the tradition. (Again, Mill was not so foolish to think that free speech would inevitably lead to truth or social harmony.) Prospectively a commitment to an IHS is simply a jump in the dark, it is (to echo Star Trek) a journey into the unknown.

Liberalism surely, then, relies on a kind of faith. But it is a faith in individual and collective intelligence: that in different contexts we’ll figure things out; and that we don’t need a higher power to recognize what’s good for (each of) us.

*Vermeule also diagnoses another purported problem: the depletion of social capital. But this strikes me as empirically unsupported (and also a failure of imagination).