On Quine's Conservatism, and the separability of philosophy from (technocratic) politics.

A recent blog post (here) by Cyril Hédoin got me to read Jeffrey Friedman’s (2019) Power without Knowledge: A Critique of Technocracy (OUP). The book draws quite deeply on Quine and Lippmann to offer a fresh account of technocracy and its limits (and pays special attention to debates over voter ignorance, the financial crash, journalistic impartiality, etc). In many ways the book is quite iconoclast and wild, and there are parts where the arguments feel dated even though they are profound on their own terms.

In reading Friedman’s book, I was reminded of a comment by Vernon Smith (the Nobel laureate in economics) on Ellsberg in a letter: “in Ellsberg a unifying thread, throughout his career, that found and exposed flaws in decision theory and decision making—from his early scientific papers to his leak of the Pentagon Papers to which he had access at Rand as a coauthor with top secret clearance.” There is a similar tenacity in Friedman, who I would describe as a ‘leftie Lippmann’ with a keen eye for unintended consequences. In future posts I return to his arguments and positions more carefully.

In today’s post, I want to use Friedman’s use of Quine to look at the relationship between Quine’s philosophical views and his conservative politics. In graduate school (Chicago was a stronghold of Carnapianism a quarter century ago), I had heard that Quine had been a Nixon supporter (see here for evidence). Carnap’s red Vienna roots were being re-discovered, and his embrace of conceptual engineering seemed to fit well with the modernist aesthetic of social engineering. By contrast, Quine looked (despite his many Carnapian commitments) stodgy.

In the early days of the philo-blogosphere, Edward Feser once collected quite a few of Quine’s more conservative statements (here). But Feser argued that these had no intrinsic connection to his philosophy (see here for his own restatement of his position). For Feser being conservative is not a bad thing. In fact, in 2009 Feser claimed that “Quine apparently never so much as entertained the possibility that conservative political views might sit uneasily with such a radical metaphysical position.” Let’s call this the ‘separability thesis.’

Below I suggest that some of Quine's philosophical views do support his conservative views (or vice versa). In particular, I suggest that Quine holds a cluster of doctrines that are critical of technocracy, as Friedman’s work also implies.

Now, before I continue: even in quite systematic philosophers one’s metaphysical or epistemological views need not determine one’s political stances. That is to say, it’s philosophy and not ideology. (Friedman has a useful working definition that draws on his Quine-eanism, “An ideology enables” the thinker “to accumulate a knowledge base that confirms its crowning postures, producing dogmatism about them as a side effect.” (p. 312)) Often in philosophy the connection is not so tight. Does Spinoza’s monism determine his social contract theory?, probably not; but one can see how he draws out the connections. Does Spinoza's monism and social contract theory determine his qualified defense of a certain kind of political oligarchy? Not really. (Etc.) Even so, that rather skeptical fideists like Montaigne and Hume simultaneously have conservative leanings can be explained to some degree. That is, their world view and their political leanings support each other. (In Friedman “a worldview… is a bundle of logically related positions, some or all of them tacit, that can have a dogmatizing effect if it becomes the interpretive basis of information gathering (that is, if it becomes an ideology), but can otherwise merely have the effect of creating latent predispositions toward certain conclusions that, in the limit case, are not supported by any information at all.” p. 313)

So much for set up.

Throughout his (2019) book, Friedman uses a whole different variety of underdetermination arguments to undermine the epistemic commitments of much contemporary technocratic-friendly social science. In fact, one of his main insights is that much contemporary technocratic-friendly social science posits either relatively homogenous agents or relatively homogeneous responses to incentives and signs/cues from the environment.

On Friedman’s revisionary view, this commitment is not supported by the empirical evidence and also (and more interestingly) means that social sciences cannot do justice to the causal role of ideas in human agency. For on his view, our existing ideas and beliefs shape our responses in different (and non-random) ways to our environments. In his jargon: we interpret signs in idiosyncratic ways (radical translation begins at home) in virtue of the path dependency of our life trajectories. We are confronted by “one great, blooming, buzzing confusion” and our irreducible posits only offer us a partial slant on (social) reality, “a device for working a manageable structure into the flux of experience.” The first quote is from James (sometimes via Lippmann) and is repeated throughout Friedman’s book. The second quote is from Quine’s “Two Dogmas” (and uses Whitehead’s language).

As noted above, Friedman draws on several of Quine’s commitments and quotes him a number of times. These are mostly from Quine and Ullian’s (1978) The Web of Belief, although he includes “Two Dogmas” in the bibliography and, as hinted just now, there is some evidence he quotes it without always giving citations to it.

Anyway, here are the explicit passages Friedman quotes from Quine and Ullian:

[A] p. 67: Widespread misbeliefs can thrive” because “ignorance of relevant truths is often accompanied by ignorance of that ignorance”—that is, by radical ignorance.—Quine and Ullian 1978, 59.

[B] p. 236: Still, it would seem that the ideas screened in by one’s web of beliefs should be largely consistent with a given posture and should therefore have the effect, over time, of strengthening one’s confidence in its validity as the web grows in size. Thus, Quine and Ullian contend that “what we come to believe derives much of its support from the sheer bulk of past cases.”—Quine and Ullian 1978, 84.

[C] p. 240: as Quine and Ullian put it, “The cause of a belief may have been some unqualified person’s irresponsible remark. It may even have been a misunderstanding on our part of someone’s words, or a subconscious association of ideas. . . . The cause may have gone unnoticed, or have been forgotten; but the belief is there, and by chance it may even be true.”—Quine and Ullian 1978, 15.

[D] pp. 246-247: Rather, one evaluates inbound information as plausible or implausible by virtue of its congruence or incongruence with the totality of the “salient” evidence and reasoning of which one is aware, and the interpretations to which the evidence and reasoning give rise. That was Quine’s point in “Two Dogmas of Empiricism.” As he and Ullian later put it, “It is in the light of the full body of our beliefs” that candidate beliefs “gain acceptance or rejection; any independent merits of a candidate tend to be less decisive.”—Quine and Ullian 1978, 16. I am not suggesting assent to all aspects of Quine and Ullian’s account, such as their claim that a web of beliefs is ultimately grounded in observation statements.

[A*] The ignorance of relevant truths is often accompanied by ignorance of that ignorance.

[A] and [A*] are expressions of so-called true Knightian uncertainty. In Friedman, the context of a [A] is the advanced division of epistemic labor.

[B] describe the path dependency of beliefs and fixation of belief in virtue of it. And [C] describes the potential ungroundedness of one’s path dependent beliefs. In particular, in broad context they are part of an explanation of circumstances where it is rational to defend one’s position even if one lacks good arguments. The context is, in fact, an explanation for a kind of confirmation bias/fixation of belief. Friedman calls the process a ‘spiral-of-conviction’ and, while drawing on Lippmann, he anticipates elements of Kevin Dorst’ account of rational polarization (see especially pp. 241-3)

[D] describes epistemic/evidential holism.

Now, none of these commitments necessarily entail conservatism. However, [B/C] are standard views in the Burkean variety of conservatism. So, this alone casts doubt on Feser’s separability thesis.

In addition, [A/A*] true Knightian uncertainty is standardly embraced by many critics of technocracy. Not all critics of technocracy are conservative, of course. And not all who embrace true Knightian uncertainty are in all respect critics of technocracy. As Oakeshott noted in “Rationalism in Politics” even some rather (classical) liberal free market theorists who claim to reject ‘saint-simonism’ may still embrace other features of technocracy. While Oakeshott has Hayek in mind, Jeffrey Friedman singles out James Buchanan — who was Knight’s student! — on this very point.

Obviously, [D] does not entail anything conservative, but Friedman uses it to interesting effect to undermine technocratic social sciences. More on that soon.

What’s lurking here, however, is that Quine’s supposed “scientism” (Feser’s word) may be anti-technocratic in character. In fact, that Quine is antitechnocratic is supported by some of the very passages that Feser cites.

[E] “Freedom to remodel society, gained by revolution, can be a delicate affair. Society up to that point, if stable at all, was stable in consequence of the gradual combining and canceling of forces and counter-forces, some planned and some not. The new and untested plan shares all the fallibility of the planner, this young newcomer in a complex world. Maybe the new order bids fair to overprivilege a hitherto underprivileged group; maybe it will presently prove to underprivilege all. It is delicate, and delicacy is seldom the revolutionary’s forte. The constraint imposed by social tradition is the gyroscope that helps to keep the ship of state on an even keel.” (Quiddities: An Intermittently Philosophical Dictionary, p. 69)

Some of the other passages quoted by Feser hint at a fondness for empasizing the risks of unintended consequences.

All of this may be surprising because in ‘Epistemology Naturalized’ Quine famously insists that “Epistemology, or something like it, simply falls into place as a chapter of psychology and hence of natural science.” (Ontological relativity and Other Essays, 82)* Notice, that here psychology is not treated as a social science, but a natural science. Quine repeats the point several times on the next page, including in his slogan, “epistemology in natural science and natural science in epistemology.” (83) In fact, when he discusses the three component sciences of psychology — evolutionary psychology, behavioral psychology, and linguistics — he never treats any of them as a social science. Nor does he treat social control as an end. There is an odd detail, in “Epistemology Naturalized” that suggests this may be deliberate; Quine treats Peirce as “the old empiricist” (78) not as the Pragmatist (which would emphasize the use of science).

Now, in “Ontological Relativity,” Quine uses an example from ‘economic theory’ to illustrate the limits of ontological reduction. So, it’s not as if he is wholly uninterested in social science. (In fact, we know he was buddies with Samuelson.)

But there is another suggestive passage that makes me think Quine (writing with my former colleague, Joe Ullian) was a critic of technocracy. It occurs near the very end of their little book (second to last paragraph). In context, they argue that natural selection gives some ground for hoping there is considerable moral agreement. But they realize this cannot justify. They write:

[F] There have been many theories that have sought to provide ultimate grounds for what is morally good or right. Some are religious theories, pure and simple; others have taken human desires or interests as their bedrock. Still others are cast in an abstract vein; Immanuel Kant's, for example, turned on what maxims might admit of universal generalization for all people at all times. Some modern decision theorists are now arguing for a foundation that draws on both the interest theories and the abstract ones; they build on a sophisticated mathematical theory of preference. On the whole theories purporting to offer ultimate grounds for moral appraisal have had their troubles; surely none has commanded anything approaching universal assent.

Here Quine and Ullian are not rejecting technocracy as such. But they are rejecting the technocrat’s favorite tool as providing a foundation or sufficient reason for moral life. Their rejection is characteristically conservative in character, moral life is taken for granted. And because the technocrat’s favorite tool lacks foundations it is denied what we may call a self-vindicating character.

To be sure the passage does not justify attribution to Quine and Ullian whole-sale rejection of the use of modern decision theory. But it does seem to require case-by-case justification. As they put it at the start of their book, “Decisions in science, as in life, can be difficult. There is no simple touchstone for responsible belief.”

---

*See also “if we are out simply to understand the link between observation and science, we are well advised to use any available information, including that provided by the very science whose link with observation we are seeking to understand.” (p. 76)