Let me conclude my brief commentary of Christopher I. Beckwith’s intriguing (and controversial, see also here) Greek Buddha: Pyrrho’s Encounter with Early Buddhism in Central Asia, following up on yesterday’s post on the alleged relationship between Buddhism and Pyrrhonism (and, indirectly, other Hellenistic philosophies, including Stoicism).
We are now going to tackle the phenomenon of the Greek Enlightenment, to which Beckwith devotes chapter 4 of his book, and which he provocatively subtitles “what the Buddha, Pyrrho, and Hume argued against.” (On David Hume and his take on both Pyrrhonian skepticism and Stoicism, see here.)
Beckwith begins by drawing a similarity between the ancient problem of the criterion and the modern problem of induction in philosophy, most famously associated with Hume. The problem of the criterion, as I mentioned in the previous post, consists in how exactly do we answer questions like “What do we know? How are we to decide in any particular case whether we have knowledge?” And Beckwith’s central thesis is that the problem was introduced to the Western world by Pyrrho, who in turn got it from the early Buddhists. Given that Hume expressly considered himself a “Pyrrhonian,” one can easily see the connection.
Hume’s problem of induction is a serious one in modern epistemology, and there is a consensus that not only it hasn’t been solved, but it is, in fact, unsolvable, just like Beckwith insists.
The problem can be stated as follows: on what can we ground our claims to empirical knowledge? (Remember that Hume was one of the British empiricists, as opposed to rationalists like Kant.) If we say that we derive knowledge from previous knowledge plus new observations — which is the way science works — then we are saying that induction is based on previous inductions (plus fallible observations). But this is circular, since we are trying to justify induction by way of inductive reasoning.
One could then try to ground induction in the other major mode of reasoning, deduction, where one starts from first principles, known as axioms or assumptions, and derive knowledge from there. This, however, won’t do, since when it comes to empirical matters (as opposed to abstract mathematics or logic) our assumptions are themselves empirical in nature, which means that they are justified inductively — and we are back to the charge of circularity.
This, argues Beckwith with some reason, is analogous to the problem of the criterion: “In order to know if the chosen criterion [for knowledge] is correct, we need to use another criterion. But it too has the same problem: it demands yet another criterion. And so on, ad infinitum. It is therefore impossible to have a criterion of truth.”
What then? Beckwith stands with Hume on this: the only possible conclusion is that there cannot be such thing as perfect human knowledge, and that all we will ever achieve is different degrees of reliability of our notions about the world.
This is true, as far as it goes, but it seems to me to be a far more modest claim than Pyrrhonian skepticism: Pyrrho counseled to abstain from judgment altogether, while Hume does not. The Scottish philosopher was too pragmatic for that, and even though he had shown that human knowledge is inherently fuzzy and provisional, he then immediately added that that’s the best we got anyway, and we need to go on with our everyday business, as well as with the business of science.
Accordingly, I teach my students that the problem of induction is best understood as a reminder to exercise the virtue of epistemic humility: never claim that we know something for certain, because there is no such thing as certain human knowledge.
Why is this relevant to Buddhism? Because “this brings up the question of why the Buddha expressed his insight in the Trilakṣaṇa. Was he, too, expressing a veiled criticism of the dominant belief system known to him — Early Zoroastrianism or Brahmanism, or perhaps both — which like Christianity focused on an all-creating divinity and the perfection of things?”
Maybe. That’s speculative enough, but Beckwith goes further in the direction of rejecting the very idea of perfection: “for anything to truly, absolutely exist as a discrete thing, it must have a permanent inherent identity, which therefore cannot change, and as a result causation — which involves change — is impossible. The Buddha rejects this absolutist-perfectionist view. And as Hume shows so well, causation is all around us and a fundamental feature of our mental processes.”
(Incidentally, on the equally complex issue of causation, which actually led Hume to formulate his problem of induction in the first place, see here.)
I’m not sure that quite follows, but before closing I want to go back to the Stoics, who had ongoing disputes with the Skeptics, especially of the Academic (post-Plato) variety.
To compare the above to the Stoic take, consider this explanation of Stoic epistemology, from Dirk Baltzly’s article in the Stanford Encyclopedia of Philosophy:
“Unlike the Platonic tri-partite soul, [for the Stoics] all impulses or desires are direct functions of the rational, commanding faculty. This strongly monistic conception of the human soul has serious implications for Stoic epistemology and ethics. In the first case, our impressions of sense are affections of the commanding faculty. In mature rational animals, these impressions are thoughts, or representations with propositional content. Though a person may have no choice about whether she has a particular rational impression, there is another power of the commanding faculty which the Stoics call ‘assent’ and whether one assents to a rational impression is a matter of volition. To assent to an impression is to take its content as true. To withhold assent is to suspend judgement about whether it is true. Because both impression and assent are part of one and the same commanding faculty, there can be no conflict between separate and distinct rational and nonrational elements within oneself — a fight which reason might lose. Compare this situation with Plato’s description of the conflict between the inferior soul within us which is taken in by sensory illusions and the calculating part which is not. There is no reason to think that the calculating part can always win the epistemological civil war which Plato imagines to take place within us. But because the impression and assent are both aspects of one and the same commanding faculty according to the Stoics, they think that we can always avoid falling into error if only our reason is sufficiently disciplined.”
It seems to me pretty clear that the ancient Stoics were wrong on this, while Plato got closer to the mark. But what we can do as modern Stoics is to mediate a compromise very much along the lines of the one that Hume himself worked out in his own essays on Skepticism and Stoicism (again, see here). We can agree with the Skeptic that human reason is indeed fallible, but also accept the Stoic account of a distinction between impressions and assent, as well as the Stoic idea that human judgment is perfectible, in the sense that it can be improved upon with proper training. This, seems to me, leads to a reasoned rejection of extreme Pyrrhonism. And indeed, even the late Stoics sought a more sensible position: “Is it possible to escape error altogether? No, it is impossible: but it is possible to set one’s mind continuously on avoiding error.” (Epictetus, Discourses IV.12).
How Buddhism deals with this sort of issue is a different matter for another post (or, better, a more qualified author — though it is interesting that the relevant Internet Encyclopedia of Philosophy entry states that “the Buddha’s epistemology can be interpreted as a middle way between the extremes of dogmatism and skepticism.”).