Ontological (dis)proof

Graham Harman, at work on a book on Quentin Meillassoux and in Paris, and reading Meillassoux’s unpublished dissertation L’inexistence divine, remarks on Meillassoux’s “hilarious reversal of Anselm:”

For Meillassoux,(Harman remarks) God is the only entity that can be proven for sure not to exist.

Everyone who knows the ontological proof knows that it was argued against almost from the day it left Anselm’s scriptorium. But Meillassoux is in a fairly select group in arguing not only that Anselm didn’t prove what he set out to prove, but that the opposite can be proven on (onto)logical grounds alone.

It is not quite, however, a set with only one member. Harman’s post reminded me of a 1948 article by John Findlay called “Can God’s existence be disproved?” Findlay is one of the more unjustly neglected figures of 20th-century philosophy, a prototype for today’s belated attempts to “bridge the Continental-Analytic divide.” A younger colleague (and eventual critic) of Wittgenstein, he visited Carnap in Chicago (when Wittgenstein learned of it, he asked not to have Carnap’s name mentioned again lest—they were sitting in an ice-cream parlor—he “lose his milkshake.”) Should these qualifications as an analytic philosopher seem insufficient, Findlay also taught Arthur Prior, the founder of tense (a.k.a. “temporal”) logic, and secured him his first jobs. Yet Findlay also translated Husserl’s Logical Investigations, wrote what was for decades almost the only book on Meinong in English, as well as works not only on Hegel and Kant, but (especially in the Gifford lectures and in Ascent to the Absolute numerous themes from Indian and Chinese philosophy; the epigraph for the second volume of Gifford lectures reads:

How odd
Of God
To choose
The Hindoos.

Findlay’s ontological proof of the non-existence of God comes from early in his career. Any object of religious devotion, he argues, would (having unsurpassable superiority as Anselm’s God does), be such that its non-existence is logically inconceivable. Thus far, Findlay is on orthodox Anselmian grounds. God's existence would then be inescapable, if and only if He is a necessarily existing being. But according to Findlay, there are no necessarily existing beings.

As Findlay wrote in a later synopsis:

Existence…has sense only when general descriptions have an extra-linguistic application; to say that things exist is a roundabout way of saying that their descriptions apply. And…ifwe could give sense to the necessary existence of anything, we should at once make anything we said of that thing empty and unmeaning. For the nature of necessary truths is that they hold whatever the non-necessary circumstances may be, and this means that there could be no saving or redemptive or consoling implications in the necessary existence of a religious absolute. On all counts then, the implications of modern linguistic philosophy as of Kantianism are that religious absolutes are not things that possibly may exist: they are things or putative things which certainly do not exist, since reference to them involves either a violation of the forms or rules of logic. (The Transcendence of the Cave pp 85-6)

Thus, from the definition of God as a necessarily existing being, and the stipulation that there are no such beings, he draws the logical conclusion: God does not exist.

It was indeed an ill day for Anselm when he hit upon his famous proof, (he wrote in his 1948 paper); For on that day he not only laid bare something that is of the essence of an adequate religious object, but also something that entails its necessary non-existence.

The hinge of this argument, it’s easy to see, is in the minor premise: there are no necessary entities. Now bearing in mind that Meillassoux’s magnum opus remains unpublished and untranslated, still one can glean from After Finitude that this is also Meillassoux’s contention. There is of course a difference. Findlay is arguing from Kantian and from Wittgensteinian grounds that the necessity in question is nonsensical: instead of an ex hypothesi necessity, to the effect that Q is necessary because we have posited “if P then Q,” it is a rather an arbitrary stipulation of the necessity of Q all on its own. This necessary existence, divorced from all contingent P’s, is what Findlay says must be barren of consequence for any contingent beings who also happened to exist (say, you and me). This is close in spirit to what Wittgenstein says about all necessary truths being tautologies. Meillassoux on the other hand rejects necessity as such, with the sole exception of (as his subtitle says) “the necessity of contingency.” For Meillassoux, the notion of any necessity is itself contradictory. This is of a piece with his rejection of the principle of sufficient reason, which of course is the move by which Meillassoux means to undo the itch for absolutes, and thus the warrant for fideism (=strong correlationism) (See A.F., p 82).

Now Findlay later came to hold that, as he put it, that

one can work the argument in reverse. If, instead of holding a necessity of existence to be impossible, one holds it to be conceivable, then one is at once able to conclude that it is actual and necessary….[in this case,] the only way it can avoid existence is in fact by being at some point internally inconsistent or otherwise impossible. (ibid, p 89)

This is more or less what Gaunilo had tried to argue against Anselm, in making God analogous to an island, and Meillassoux seems to hold that to treat the Whole as an intra-whole object is to treat it as an island or something like it. As Findlay remarks, the argument depends on whether one sees necessary existence conceivable or inconceivable:

If one’s ideas of the possible are not unclear and muddled (as they always to some extent must be), one cannot merely be dubious about a necessary being: one must either reject it or accept it outright. (p. 90)

The upshot here seems to be, that the only reason one could have for being tentative on this score is being unsure on the degree of clarity in one’s ideas of possibility. So of course it remains of some interest that just as Findlay’s conclusion at this stage in the argument was an “attitude…of tentative acceptance” of God’s existence (he later developed it into a more full-fledged “rational mysticism”), so Meillassoux, in a very different key, also reaches a sort of accommodation. As is well known from his article “Spectral Dilemma;” the assertion of God’s ontological impossibility is (in Harman’s words)

true only if God is identified with the Whole, which Meillassoux does not do. There’s actually plenty of room for the virtual God in his system.

This virtual God, thought rigorously under the rubric of possibility, may come to be someday, and be in this way innocent for all those terrible ills guilt for which (Meillassoux thinks) the most tortured convolutions of theodicy has not delivered Him. Meillassoux’s rejection of the Whole is part of what he takes on from Badiou’s “Platonism of the multiple.” I am not myself convinced by this argument, which is often said to depend upon Cantor, but actually rests on a particular and not at all self-evident reading of Cantor. (Gödel for one, I think, did not hold to it.) But Meillassoux’s weird future-oriented virtual theology remains one of the most provocative aspects of his thought; his science-fiction god is just an index of how provocative supposedly highfalutin’ metaphysics can be.

These parallels between Findlay and Meillassoux are not moves in the idle game of “find the antecedent.” (Neither were my comparisons between Meillassoux and Barfield.) Antecedents there are aplenty, for nearly any pronouncement. What is interesting (and insufficiently remarked) in philosophy is when very different thinkers stumble upon a similar configuration of thought. As John Shade wrote:

If on some nameless island, Captain Schmidt
Sees a new animal and captures it,
And if, a little later, Captain Smith
Brings back a skin, that island is no myth.

(Pale Fire ll.759-762).

Findlay, the semi-Wittgensteinian neo-Platonist, and Meillassoux, the scourge of correlationism, have both glimpsed the same most perfect island, though perhaps respectively (it would seem) from the windward and the leeward side.