News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.
Cyber security professional here -reliably determining when a computational artifact (file, etc.) was created is *hard*. This is sorta why…
Agreed with the other commentator. It is extremely unlikely that Pangram’s success is due to its cheating by reading metadata.
I see this question as a bit naïve. There is metadata on every document created by a modern word processor…
There’s a simple way to test. Open a pre-2022 essay and copy-and-paste it into a new file.
At the risk of self-advertising:… You claim “AI is unusual in degree, not in kind” and “It is not clear…
Apropos of Sagar’s wish to foist the A.I. industry by its own petard, this article appeared in print in yesterday’s…
I teach both large courses, like Jurisprudence and Critical Legal Thinking (a.k.a Legal Argumentation), and small seminar-based courses at Edinburgh…
Professor Miller, a longtime member of the philosophy faculty at the University of Warwick, was a well-known defender of Karl Popper's philosophy of science and epistemology, and also wrote widely in mathematical logic. He died in November 2024. The Warwick memorial notice is here. Comments are open for remembrances from those who knew Dr. Miller or those who would like to comment on the significance of his work.
(Thanks to Jeffrey Ketland for the pointer.)
Miller's paper, "Popper's Qualitative Theory of Verisimilitude", has been very influential. Popper's appeal to the notion of verisimilitude looked like an attempt to salvage something like confirmation, despite his commitment to falsificationism. Miller (and, independently, Tichy) should that there were serious problems with Popper's conception of verisimilitude.
Recall Popper's words:
"I do not suggest that the explicit introduction of the idea of verisimilitude will lead to any changes in the theory of method. On the contrary, I think that my theory of testability or corroboration by empirical tests is the proper methodological counterpart to this new metalogical idea. The only improvement is one of clarification."
And in spite of Miller's uncovering of a small logical flaw in Popper's reasoning, Popper’s “new metalogical idea” has been enthusiastically taken up by a generation of realist philosophers, including Oddie, Niiniluoto, Kieseppä, Kuipers, Zwart and others.
Perhaps this is why Miller seemed to retain his allegiance to Popper's ideas. He recognized that even when wrong, Popper was the one who was most worth listening to.
David
You are absolutely correct about the legacy of Popper's attempt to articulate a notion of closeness to truth. It is a busy and well-staffed research programme. I tend to think that this was a wrong turn in Popper's development. It seems inconsistent with the core of his falsificationism. I think he should have stuck with what he managed to show in LSD – the canons of rationality only dictate what we should not believe (not what we should believe). (But I am clearly one of those who will write the dissenting opinion on the court of scientific rationality)
I'm a bit curious about what's been said here (which is more about Popper than about Miller). I would have thought that falsificationism was distinct from the idea of verisimilitude. That is, one could defend or endorse a falsificationist theory of method without thereby being committed to the idea of verisimilitude. You'd just be saying that what is rational to accept is the theory that has been most rigorously tested and survived all tests. You reject those theories that have failed tests. You don't thereby commit yoursel to the belief that the theory that survives rigorous testing has a higher degree of verisimilitude than the theories that preceded it.
I have tended to think of the notion of verisimilitude as one that Popper arrived at when he started trying to think through scientific realism and the implications of scientific realism with respect to scientific progress. He wanted to be able to say something about how a later theory might be an improvement on an earlier theory, namely, that the later theory might have a higher degree of verisimilitude than the earlier theory. But that was a later development than the falsificationist theory of method.
Of course, Lakatos did press Popper on the relationship between corroboration and verisimilitude, and argued that Popper needed a "whiff of inductivism" to connect the two notions up. But Popper didn't want to set up a tight connection between increase of corroboration and verisimilitude. He has a footnote in the Schilpp volume where he says something about a realist intuition about the determinacy of the world, but won't go so far as to definite verisimilitude in terms of an increase of corroboration.
Howard
I agree, and I thought that was clear in my comment: falsification is distinct from verisimilitude. Popper, though, did not see it that way (or so I think). His pairing of falsificationism and scientific realism was odd. The idea that our theories always remain hypotheses sits far better with anti-realism. But Popper certainly thought that the drive for the truth is what ensures that scientists will not settle for theories that are merely instrumentally useful (recall his comparison of engineers to plumbers)
Popper, a lifetime realist, argued that the existence of a falsifiable, i.e. testable, theory implies the existence of a reality with which it can clash: “Our falsifications thus indicate the points
where we have touched reality.”
…quoted from:
"Touching Reality"
https://philpapers.org/rec/MERTRG
Popper was a realist. Not a "scientific realist" as that term is currently used.
Specifically on the question of whether Popper was a scientific realist in the way the term 'scientific realism' is currently used, the most significant point of difference that I can see relates to the "no miracles" argument. I don't think Popper ever gave or endorsed that kind of argument for scientific realism. Indeed, I seem to recall that he rejected the demand for the sort of explanation that the realist offers. I think that is a significant point of difference. On the other hand, I suspect that Popper might have held something close to a realist account of theory acceptance. The constructive empiricist would say that a theory is to be accepted as empirically adequate, i.e., true with respect to all observational claims. But I don't think Popper would have said that. I take it that for Popper acceptance of a theory, no matter how tentative, could still have been acceptance as true, i.e., true with respect to both observational and non-observational claims. Of course, he might have hedged in the way that many scientific realists do by saying "accept as true or as having a higher degree of verisilitude than its predecessor". Similarly, a scientific realist might say "accept as true or approximately true".
Howard and David
To bring the discussion back to David Miller, I do not think it is correct to say that Miller merely "uncover[ed] … a small logical flaw in Popper's reasoning". Miller's paper (and Tichy's) are quite technical, but they really take down Popper's theory verisimilitude. Roughly, Popper claimed that one theory(T1) is closer to the truth than another (T2), if T1 entails more truths than T2, and T1 does not entail more falsehoods than T2. Miller and Tichy show that these conditions cannot be met, at least not by real complex scientific theories (theories entailing many claims). I first discovered these papers while reading Psillos' excellent book Scientific Realism: How Science Tracks Truth. Stathis was quite honest about the implications of these papers – there is no clear notion of closeness to the truth. Instead, realists have to rely on our casual intuitions about closeness to truth. As David Merritt noted earlier, there is a research programme trying to rehabilitate the notion of verisimilitude, but there is not yet a consensus that such a notion has been articulated. So this is David Miller's big contribution.
Leave a Reply