News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.
I respond to this report here https://jasonstanleyantifascist.substack.com/p/on-the-philosophical-muddle-that
For what it’s worth, I agree. I have, with some trepidation, given it a bunch of my papers and asked…
Looking at some of the AI-amplified cranks on philpeople, it is hard to agree with the assessment of this article.…
Perhaps a useful distinction here is between replacing philosophers and changing the conditions under which philosophical work gets done. LLMs…
Conditional on the hypothesis that journal editors sometimes desk reject high-quality papers written and commended by world experts on a…
Some users seem to have conflicting impressions about the (absolute level of) abilities of current iterations of the LLMs. But…
One thing I appreciate about Greco’s piece is the very helpful distinction between instrumental and intrinsic value, which he lays…
I found this review of Timothy Williamson’s latest book by philosopher Daniel Greco illuminating, and it raises some deep issues about methodology in analytic philosophy. Curious what informed readers think about all this.
I strongly agree with Brian that Greco’s review is illuminating. Among other things, I think that Dan is exactly right to emphasize the tension between Williamson’s earlier seminal discussions of evidence (in _Knowledge and Its Limits_ and _The Philosophy of Philosophy_), including evidence in philosophy, and the treatment of “data” in this latest book.
At one point in his review, Dan refers to “some differences in emphasis” between the earlier and later views, but I think that this understates things. As I understand it, outright inconsistency between the earlier work and the later work is avoided only by insisting on and giving a great deal of weight to the hitherto unrecognized distinction between “evidence” and “data.” In my judgment, Chapter 7 of TW’s _The Philosophy of Philosophy_, “Evidence in Philosophy,” contains, among much else of value, a compelling critique of philosophers’ tendency to (as TW puts it there) “psychologize the evidence”. (In part for this reason, I often recommend that chapter as the single best thing I know of on the topic of evidence in philosophy, and I’ve taught it multiple times in graduate seminars over the years.) As someone who continues to find that earlier critique compelling, I find the suggestion in the new book, that the relevant notion for philosophy is a thoroughly psychologized notion of “data” not at all compelling.
A final thought about this—as Dan notes, Tim is no fan of hyperintensional distinctions. But even though there is again no literal inconsistency here, at a minimum I find it extremely odd to suggest that we jettison the hyperintensional distinctions, including ones that seem to robustly figure in the sciences and in everyday life, while giving so much weight to the alleged distinction between “evidence” and “data,” an alleged distinction that as far as I can tell plays no genuine role in the sciences, in everyday life, or in philosophy.
I agree that this book seems to walk back or at least soften a couple of things Williamson has previously defended, especially with the introduction of the word “data” as a non-factive alternative to “evidence”. Williamson has been a staunch defender of an externalist view in epistemology, and his claim that evidence is simply the same as knowledge (and thus implies its correctness) was, to my mind, part of this agenda. One “perk” of this position is that it rules out a version of the skeptical argument that claims that our evidence is unreliable. But now, it seems, he seems to allow that this argument can simply be rephrased in terms of “data”. (From the internalist perspective, this might be taken as a confirmation that such a non-factive vocabulary is ultimately needed.)
It’s also interesting that he is worried about “overfitting”. Williamson’s stated goal in “Modal Logic as Metaphysics” (and implicitly elsewhere) was to take our judgments (which he preferred not to call “intuitions”) and to use logic to find out if they are consistent. If they are inconsistent, he (roughly put) suggests to find the closest consistent version and accept that instead. For this reason, he rejected the KK-principle, but otherwise this method is aimed at a fairly close “fit”…
Another review of a book by Williamson on a similar theme (restricted to ‘overfitting’ in the study of conditionals) which may be relevant is Daniel Rothschild’s review of Suppose and Tell (https://academic.oup.com/mind/article-abstract/132/525/208/6414892?login=false).
I find the charge of losing predictive success by simply adding the ‘data’ as further parameters in one’s theory a very rough description of what is going on in philosophy of language and linguistics, even for hyperintensional theories. Such frameworks often still play the game of theoretical virtues. The difference is that such theories take robust patterns of ‘data’ to speak to extensional adequacy of a theory rather than mistakes.
I agree that we do often make mistakes and correct them even in the case of language use, but such mistakes can often be detected as deviations from competent language use (local performance mistakes). What does not happen (and what Williamson seems to do) is that a robust pattern of use is labeled as a mistake which should not be predicted by a theory and should be attributed to wayward heuristics which may have developed for one reason or another.
I also doubt that the same methodology can be applied to detect the ‘mistakes’ in metaphysics, epistemology and philosophy of language, since presumably the data in each case looks completely different. I am more convinced that the robust patterns of meaning and use which deviate systematically from the simplest classical analyses proposed for that fragment of language *is* data to be taken seriously and explained by way of the theory itself rather than superadded heuristics, whereas I am not sure if this is true in the case of metaphysics, since the linguistic data is much easier to replicate with subjects who are not vested in the success of a theory than the intuitions marshalled in favor of one view in metaphysics rather than another.
Excellent review, though I think “there is no inconsistency” should have been the conclusion emphasized in Greco’s discussion of potential tensions in Williamson’s methodology rather than a prefatory comment.
Small thing: “intentionalist” should be “intensionalist.”
Medium thing: ‘Similarly, if “Superman flies” and “Clark Kent flies” are necessarily equivalent, then they express the very same proposition. So, if propositions are the objects of belief, anyone who believes that Superman flies believes that Clark Kent flies.’ This is not (only) a consequence of intensionalism, but merely of Leibniz’s Law; relatively fine-grained Russelian theories of propositions also have this consequence.
Great review, obviously. But one thing which I would have loved to see Greco consider (though if nothing else space constraints obviously made this impossible) is the theoretical virtue of agnosticism. Compare: a geometer schooled in Euclid but ignorant of Reimann might think that she should be agnostic about the truth of any geometric theory, bc (i) she thinks the Parallel Postulate is dubious, and so Euclidean geometry is dubious too, but (ii) she knows of no other plausible geometric theory, so (iii) the right thing to do is to go agnostic about whether any geometry is true, at least until she’s worked out the details of a PP-free geometry. By analogy: a metaphysician schooled in intensionalism might think that she should be agnostic about any worlds-based account of (e.g.) truth conditions, bc (i) she thinks that there are *obviously* different propositions which are true in exactly the same possible worlds, and so intensionalism is false, but (ii) she knows of no other plausible worlds-based theory, perhaps because she buys the criticisms which Williamson gives of hyperintensional views, and so (iii) the right thing to do is to go agnostic about whether any such worlds-based account is true, at least until she’s worked out the details of a new view. (Obviously my stipulation about what intensionalism is committed to in [ii] was an oversimplification: intensionalists, forgive me.)
Also, just a side question for Daniel K.: is your first conditional true? ‘1 + 1 = 2’ and ‘2 + 2 = 4’ are necessarily equivalent, at least in the sense that, necessarily, one is true iff the other is; but no fine-grainer thinks that they express the same proposition.
The conditional is Greco’s. My objection was to the specific choice of example, which is in effect a generic Frege case.
Leave a Reply