On a passage in Hume's Dialogues

During the early modern period (ca. 1600-1800), there are three main ideal-typical cosmogenic positions one can encounter in debates in the republic of letters: (i) a powerful mind ordered the cosmos and then governs it providentially; (ii) the cosmos had no beginning and exists eternally and necessarily; (iii) the cosmos, including its origin, is governed by chance. Of course, in practice, some subtle variations within the ideal type were possible.

The first position was embraced by (Christian) Deists and Theists alike and often traced back, fairly or not, to ancient Stoicism. But it was also compatible with a standard version of Cartesianism in which God is like a cosmic billiard ball player, and, in creation, breaks symmetry by setting the homogeneous universe in motion in one direction or another. Some variants thereof were probably the majority view within the mainstream of the scientific revolution despite the protestations of Spinoza, Hume, and Diderot (who all attacked providence explicitly). And (i) was also the position that was safest to embrace. This dominance was opposed by Spinozists (viz. ii) and neo-Epicureans (viz. iii).

I don’t mean to suggest that the dominant view was only embraced in virtue of its compatibility with religious authority (and political power). It had genuine argumentative strengths.

The Epicurean position had trouble explaining the law-governed order that we find in the heavens. There is really no Epicurean argument that gets us to, say, Kepler’s laws as a feature of planetary motion. Of course, during the eighteenth century an Epicurean may be tempted to claim that Kepler’s laws are only an ad hoc feature of our solar system. But this looks question-begging. By the nineteenth century that argument also turns out to be false empirically.

That it was question-begging was clear by 1734, when Daniel Bernoulli offered a celebrated argument against the likelihood that the solar system was the effect of mere chance. Some of its evident properties suggested that it was the effect of a common cause. Of course, that argument alone could not settle the debate between those that thought the common cause was a natural mechanism, as Bernoulli thought, or God (as neo-Newtonians thought).

However, by the start of the eighteenth century (by which time Cartesianism was in clear retreat) it was also clear that the Spinozist was also struck with a kind of immanent critique: there was really no sufficient reason to explain the particular variety we observe in nature. Why, on Spinozist grounds, do we observe so many symmetry-breaking features in the solar system? Why do the laws of motion have the particular regularity that they do, and not some other? The Spinozist has no resources to explain this, except to pound the table that nature takes on all possibilities.

To avoid confusion, appeal to sufficient reason occurs throughout the eighteenth century by the very best natural philosophers who embraced determinism (see Van Strien’s paper). And, in fact, Newtonians who rejected Spinozism (and hard determinism) also tended to embrace sufficient reason. So, a principle of sufficient reason was not itself thought problematic.

Obviously, the debate over providential order didn’t just center on the heavens. For example, in a famous (1710) article in the Transactions of the Royal Society, John Arbuthnott argued that over time there existed an observed stable, albeit slightly unequal sex ratio of male and female births also was an argument for providential design (and also an inductive argument against polygamy). So, as sophisticated readers probably know (say from Eliott Sober), there is a pre-history to Darwinism lurking here. Cathy Kemp has argued there is good reason to assume that Hume was quite interested in Arbuthnot’s argument.

Okay, so much for set up. A twitter post by Mauricio Suarez on a lecture by John Dupré showed a picture of an excerpt from two paragraphs from Part VIII of Hume’s posthumously published Dialogues Concerning Natural Religion. I quote the two full paragraphs:

Suppose, (for we shall endeavour to vary the expression) that matter were thrown into any position, by a blind, unguided force; it is evident that this first position must in all probability be the most confused and most disorderly imaginable, without any resemblance to those works of human contrivance, which, along with a symmetry of parts, discover an adjustment of means to ends and a tendency to self-preservation. If the actuating force cease after this operation, matter must remain for ever in disorder, and continue an immense chaos, without any proportion or activity. But suppose, that the actuating force, whatever it be, still continues in matter, this first position will immediately give place to a second, which will likewise in all probability be as disorderly as the first, and so on, through many successions of changes and revolutions. No particular order or position ever continues a moment unaltered. The original force, still remaining in activity, gives a perpetual restlessness to matter. Every possible situation is produced, and instantly destroyed. If a glimpse or dawn of order appears for a moment, it is instantly hurried away and confounded, by that never-ceasing force, which actuates every part of matter.

Thus the universe goes on for many ages in a continued succession of chaos and disorder. But is it not possible that it may settle at last, so as not to lose its motion and active force (for that we have supposed inherent in it) yet so as to preserve an uniformity of appearance, amidst the continual motion and fluctuation of its parts? This we find to be the case with the universe at present. Every individual is perpetually changing, and every part of every individual, and yet the whole remains, in appearance, the same. May we not hope for such a position, or rather be assured of it, from the eternal revolutions of unguided matter, and may not this account for all the appearing wisdom and contrivance, which is in the universe? Let us contemplate the subject a little, and we shall find, that this adjustment, if attained by matter, of a seeming stability in the forms, with a real and perpetual revolution or motion of parts, affords a plausible, if not a true solution of the difficulty. D8.7-8

The speaker is Philo. (At this stage of the argument Demea still thinks of Philo as a fellow critic of Newtonian natural design ably defended by Cleanthes.)

That “matter were thrown into any position, by a blind, unguided force” is a rejection of final causes (and so the position associated with providence). However, that that there is a ‘first position’ is usually taken to be a rejection of Spinozism, which assumes an eternal universe. So, an eighteenth century reader may assume that in the first quoted sentence, Philo is about to develop a defense of the Epicurean position.

However, it becomes clear fairly quickly that Philo is developing a neo-Spinozist cosmogeny. It’s Spinozist because it embraces a species of determinism (very Spinozist) that has not infinite (as Spinoza would claim), but indefinite amount of time to do its work and take on all possible permutations (“Every possible situation is produced.”)

In this cosmogeny order arises out of chaos/disorder through the fixed operation of ‘blind’ natural forces. This position (here) I tend to associate with C.S. Peirce’s clever reinterpretation of synthetic philosophy writing in the Monist in 1891. So, it’s nice to see Hume anticipates Peirce.

In fact, it is Spinozist in character because natural forces seem immanent in matter here (according to Philo the ‘actuating force’ continues ‘in matter’). This is a position that Diderot had developed in D’Alembert’s Dream (as I first learned from Charles T. Wolfe.) It’s not impossible that Hume was familiar with it before he produced his final draft of the Dialogues. But a version of this view can also be found in Toland’s (1704) Letters to Serena, which was celebrated enough that it was a named target of Clarke’s (1705) Demonstration of the Being and Attributes of God. Toland himself distanced the view from Spinoza, but few (certainly not Clarke) were fooled.

It’s worth noting that Kant hit on a similar argument on the development of order from disorder in the Universal Natural History of the Heavens of 1755. In fact, when he first introduces it he explicitly associates it with Epicureanism: “If the planetary structure, with all its order and beauty, is only an effect of the universal laws of motion in matter left to itself, if the blind mechanism of natural forces knows how to develop itself out of chaos in such a marvelous way and to reach such perfection on its own, then the proof of the primordial Divine Author which we derive from a glance at the beauty of the cosmic structure is wholly discredited, nature is self-sufficient [selbst genugsam], the divine rule is unnecessary, Epicurus lives once again in the midst of Christendom, and an unholy worldly-wisdom [Weltweisheit] treads underfoot the faith which proffers a bright light to illuminate it.” (I have argued that in context, it becomes clear that Kant distances the position he adopts from Epicureanism.)

Hume would have been unfamiliar with Kant’s work because this work did not circulate until much later. I mention Kant here, to suggest that after Bernoulli’s prize essay, there is a clear argumentative path for neo-Spinozists to embrace on the arguments of the General Scholium (which defend providence).

Hume’s real contribution (as presented by Philo) is in the second paragraph. Now, there is an simple empiricist way to read Philo’s claim as being strictly false. It’s simply not true that heavens preserve ‘an uniformity of appearance.’ But Philo recognizes this, after all he states “Every individual is perpetually changing, and every part of every individual.” And, in so far as our solar system is an individual, we have no reason to expect it to endure by itself. In addition, Philo is surely (!) familiar with the precession of the equinoxes something Newton had tried to explain in Principia and subject of eighteenth-century controversy. I return to that below.

That ‘the whole remains, in appearance, the same’ presupposes a kind of ideal observer from the perspective of eternity (an infinite intellect). That the whole remains, in appearance, the same can only be true if what this ideal observer perceives is something like a stable ratio of motion and rest in the universe (as Philo’s argument seems to presuppose). This is an echt Spinozist or neo-Spinozist position. But the problem is, as Leibniz famously discerned, it’s not obvious whether in Newton’s cosmos total quantity of forces are so preserved.

It’s possible that Philo is not so up to date. And that he has in mind what Plato called the ‘great year.’ This seems to be the idea that all the heavenly motions eventually return to the same position (a form of eternal return that is usually associated with Stoicism, but can also be discerned in some Platonic works if you squint at Timaeus and Laws in the right way). The reason why I mention this option is because of Philo’s strange invocation of the forms. He clearly does not mean Plato’s forms here, but what he does mean is not so evident. (And I will not solve that today.)

Somebody may impatiently remind me of Hume’s celebrated proto-Darwinian argument that Philo had articulated just before (at D8.6). Philo’s argument there can be traced back to Empedocles (in a report of Aristotle’s Physics, book 2, part 8). Philo’s idea is that what we observe is itself an effect of a process in which, after sufficient time, unstable “forms” are permanently winnowed out. Once established, the process sustains order “itself, for many ages, if not to eternity.”

That is to say, Philo’s cosmogeny provides the outlines of a refutation of theism/deism with a mixture of neo-Spinozism (nearly infinite time, immanent forces of matter that are stable, etc.) and neo-epicureanism (a selection process with an air of chanciness that explains chaos —> order, etc.) But it has itself sufficient controversial moving parts that it can still be challenged from a providential perspective.

*