Jonathan Swift and Pessimistic Meta-Induction

[Aristotle’s ghost] freely acknowledged his own mistakes in natural philosophy, because he proceeded in many things upon conjecture, as all men must do; and he found that Gassendi, who had made the doctrine of Epicurus as palatable as he could, and the vortices of Descartes, were equally exploded. He predicted the same fate to attraction, whereof the present learned are such zealous asserters. He said, that new systems of nature were but new fashions, which would vary in every age; even those who pretend to demonstrate them from mathematical principles would flourish but a short period of time, and be out of vogue when that was determined.--From Jonathan Swift's (1726) Gulliver's Travels. Part Three, Chapter 8. 

Swift let's Gulliver recount through Aristotle's Ghost what we may call a pessimistic meta-induction (see here in the Stanford Encyclopedia). Swift is not the first to produce such a meta-induction I am familiar with several version of it to be found in Montaigne's Apology of Raimond Sebond -- involving induction over multiple sciences  --, and I would not be surprised if it has an ancient Skeptical pedigree. Montaigne wrote in the aftermath of the Copernican revolution so it is no surprise he is sensitive to this opportunity to argue for his version of skepticism. 

We are inclined to read Swift's version as, perhaps, displaying his scientific experience. How could one claim that Newton is mere 'fashion' and founded upon 'conjecture?' In his edition of Swift, Sir Walter Scott, suggest that this passage shows "the Dean understood little natural philosophy."* Scott may correct about this. While one can wonder why another novelist fails to recognize that a writer need not agree with his character, I leave that aside. But Scott underestimates Swift's remarks.

First, note that Swift correctly grasps that Newton had made Descartes's vortex theory obsolete. It's little noticed fact that Newton offered the first quantitative vortex theory and showed that it was incompatible with the empirical evidence (for explanation see this paper--the argument is Smeenk's.) This was by no means the consensus everywhere. (And, in fact, Newton's argument rested in part on a subtle mistake in his treatment of torque recognized by Bernoulli.) But Newton had clearly shifted the burden of evidence; Descartes version was doomed.

Second, it is worth noting that 1726 was also the year of publication of the third edition of the Principia. (So, Gulliver was written in the shadow of the second edition.) Even so, the most controversial empirical results of the Principia (shape of the Earth, universal gravity, the lunar orbit etc.) were still very much contested. These were not settled until more than a decade later in groundbreaking work of Maupertuis, Clairaut, and Euler. While there were many sympathetic defenses of Newton in circulation, that even in the British Isles, Berkeley and Toland had demonstrated that there was space for works that were not uncritical of Newton (along many dimensions).

Now, one may still think Swift should not have said (or let Aristotle's  Ghost say via Gulliver) that the Principia was grounded in conjecture. But that was not an uncommon view (see here for evidence) among the learned who had good evidence to think that the empirical world did not favor Newton's theory and that he violated standards of intelligibility. 

Moreover, the idea that the use of mathematical is a matter of fashion and context specific is one that is key to a series of important arguments by Mandeville developed in his (anti-mathematicist) treatment (see here) of using the incentive structure of medicine to support epistemic claims in his A Treatise of the Hypochondriack and Hysterick Diseases (recall my old posts here and here.) I believe Mandeville's arguments were inserted in the second (1730) edition. If so, then Swift may well have encouraged Mandeville's development of the point. Mandeville's argument is so much more thoroughly developed than Swift's remark that I hesitate to call it an influence. 

Anticipating Kuhn (and Adam Smith), Aristotle's Ghost treats the commitment to a particular scientific paradigm as akin to a religion. And while he does not claim that the acceptance and rejection of scientific theories is entirely due to social context, social context is a key factor. And, in fact, it would be useful to read this passage alongside the critical treatment of the royal society in his description of his visit to Laputa and the floating island, but that's for another voyage, I mean digression.

 

 

*I don't fully get the whole footnote, and I would love for somebody to explain it to me.  

 

 

Comments

[David Hilbert]

Not at all relevant to the point of your post but something you might appreciate that I just ran across:

https://link.springer.com/article/10.1007%2Fs12576-018-00655-4

What I find interesting is the actual sophistication of Swift's attempt to estimate the physiological parameters, not that he got it wrong.

[Eric Schliesser]

Yes, I think that's marvelous. Thank you David.