Responding to AI Summaries of Popper's Critics

I asked the Gemini 2.5 Pro (Preview) AI:

Today, what are the most popular texts criticizing Karl Popper's epistemology?

Let's go through Gemini's answer. I think this will be reasonably useful because the AI is trained on many statements people wrote about Popper and about these critical texts. I expect these answers to provide a reasonably representative mixture of what many philosophers have written.

Identifying the "most popular" texts criticizing Karl Popper's epistemology today involves looking at works that remain highly influential, are frequently cited, and continue to shape discussions in the philosophy of science. Based on these criteria, here are some of the most significant texts and influential lines of criticism:

1. Thomas S. Kuhn - The Structure of Scientific Revolutions (1962): This is arguably one of the most impactful books in the 20th-century philosophy of science and offers a profound challenge to Popper's views.

• Main Criticism: Kuhn argued that Popper's emphasis on falsification as the primary activity of scientists misrepresents much of scientific practice. Kuhn introduced the concept of "normal science," where scientists work within an established "paradigm," solving puzzles rather than constantly trying to refute the paradigm itself. Falsification, for Kuhn, is more characteristic of "scientific revolutions" (paradigm shifts), which are rarer and more complex social and psychological processes than Popper's logical schema suggests. The "Kuhn-Popper debate," particularly prominent in the collection Criticism and the Growth of Knowledge (edited by Imre Lakatos and Alan Musgrave, 1970), highlights these differences.

Popper didn't say that falsification is the primary activity of scientists. He said scientists need conjectures and refutations, not primarily refutations. Popper emphasized this point, even titling a book Conjectures and Refutations.

Popper's goal wasn't to observe what scientists do in practice then accurately document it. He used abstract, logical reasoning to analyze how science could, should, and couldn't possibly work. Responding about what the primaries activities of scientists are or aren't is not very relevant to Popper's main ideas.

The "normal science" and "scientific revolutions" issue is basically about sociology, psychology and human behavior. If it's correct, that wouldn't make Popper's logical analysis of epistemology incorrect. What people are like today (or even their unchangeable human nature) is a separate issue from the logic of knowledge creation.

These claims don't challenge Popper's main, logical points about epistemology. They're a side issue from what I wanted to defend or discuss.

1. Imre Lakatos - "Falsification and the Methodology of Scientific Research Programmes" (1970): Published in Criticism and the Growth of Knowledge, this work attempts to provide a more sophisticated version of falsificationism that addresses some of the criticisms leveled against Popper (including Kuhn's).

• Main Criticism (and development): While a student of Popper, Lakatos acknowledged the naivety of "dogmatic falsificationism." He proposed that scientists work with "research programmes" consisting of a "hard core" of theories protected from refutation by a "protective belt" of auxiliary hypotheses. A research programme is progressive if it leads to novel predictions and degenerating if it only makes ad-hoc adjustments to anomalies. This implicitly criticizes Popper's simpler model by showing how theories can be legitimately shielded from falsification, at least temporarily.

Popper's model was focused on a different thing. Popper focused more on the logic of abstract epistemology, not on the behavior of people. The behavior of people is more complex and is also a different topic. Popper's focus on logic not psychology or behavior was even in his book title, The Logic of Scientific Discovery. Again, this criticism a bit off topic and doesn't actually try to challenge Popper's main epistemology claims.

1. Paul Feyerabend - Against Method (1975): This book presents a radical critique of the idea that there is a single, universal scientific method, including Popper's falsificationism.

• Main Criticism: Feyerabend advocated for "epistemological anarchism," arguing that major scientific advancements often occurred precisely because scientists violated established methodological rules. He contended that any prescribed methodology, including falsification, would stifle scientific progress. His famous slogan is "anything goes."

This doesn't actually claim that Popper's epistemology methods are wrong or don't work. It just claims that there might exist other working methods. None are mentioned here. There is no mention here of any defense of the particular methodologies that Popper said don't work (like induction or justificationism).

I don't recall Popper saying there couldn't possibly be any other methods. If Feyerabend actually knows of any other viable methods, that would be interesting, but that isn't mentioned here. I tried following up with Gemini:

Short answer. Did Feyerabend defend any specific methods besides Popper's as viable?

It said:

No, Feyerabend famously argued against the idea of any single, universally prescribed scientific method, including Popper's, being uniquely viable.

Instead, he advocated for methodological pluralism, famously encapsulated in his phrase "anything goes," suggesting that scientists should be free to use whatever methods, rational or irrational, best suit their research and lead to progress, rather than being constrained by one specific set of rules.[1] He didn't champion another specific method over others as the single correct one.  

So Feyerabend said go ahead and do anything even irrational things. I guess he also was in favor of using illogical methods. This strikes me as either silly or not really talking about the same things as Popper. Methods that, as a matter of logic, cannot work ... cannot work. So don't use them. What would make some sense is using methods that people don't know how to explain in words or logically defend in hopes that they have differences from the methods that have been logically refuted. Human behavior is often quite complex so even if some elements of it are logically wrong, progress could be made because other elements are better. But this is again getting into issues of human behavior instead of the abstract logic of science and epistemology.

1. Hilary Putnam - Various Essays (e.g., "The 'Corroboration' of Theories" in Philosophical Papers, Vol. 1, 1975): As detailed previously, Putnam raised significant objections to Popper's epistemology.

• Main Criticism: Putnam focused on the untenability of Popper's rejection of induction, arguing that if corroboration provides no reason to believe a theory is more likely to be true or reliable for future predictions, then science becomes practically irrelevant. He also emphasized, in line with the Duhem-Quine thesis, that theories are not tested in isolation, making straightforward falsification problematic.

Saying not to reject induction because you don't like the consequences is a bad argument. You can't figure out what's true based on what you'd like to be true.

Also, Popper didn't think science is irrelevant. Putnam is taking some extreme, exaggerated version of Popper's position, then finding he dislikes it, then concluding that Popper must be wrong and we can stick with some standard views despite Popper's criticisms of them.

You don't need any alternative in order to refute something like induction. Again, all of these criticisms strike me as not really focused on logical analysis.

Being unhappy about Popper's alternative doesn't make it false, and it's also weird to be unhappy over conclusions Popper himself didn't reach and denies.

Regarding testing theories in isolation: that's an important issue which Popper worked on, other thinkers worked on, and which could still be worked on more. It's a difficulty that comes up for Popper and others, which we could still figure out better ways to handle, but which isn't a decisive refutation of Popper's epistemology.

1. The Duhem-Quine Thesis (articulated in works by Pierre Duhem and W.V.O. Quine, e.g., Quine's "Two Dogmas of Empiricism," 1951): While not a direct critique solely of Popper (Duhem predates much of Popper's relevant work), the implications of this thesis are fundamental to many criticisms of falsification.

• Main Criticism: This thesis asserts that a hypothesis cannot be tested in isolation because empirical tests of a hypothesis require numerous background assumptions (auxiliary hypotheses). Therefore, when an experiment yields a result contrary to the hypothesis, it's not clear whether the fault lies with the hypothesis itself or one of the auxiliary assumptions. This makes conclusive falsification of a single theory difficult, if not impossible.

Yes. Popper discussed this. His methods can be used despite these issues because they don't rely on conclusive proof of anything. All our knowledge is tentative and fallible.

1. Larry Laudan - e.g., "The Demise of the Demarcation Problem" (1983) and Beyond Positivism and Relativism (1996): Laudan famously criticized Popper's central project of demarcating science from non-science (or pseudo-science) using the criterion of falsifiability.

• Main Criticism: Laudan argued that the demarcation problem is a pseudo-problem and that no single criterion (including falsifiability) can successfully distinguish science from non-science in all historical and practical contexts. He pointed out that many well-established scientific theories have been unfalsifiable at certain points, while some pseudoscientific theories can be formulated to be falsifiable.

These texts and the arguments they contain remain central to understanding the limitations and challenges to Popper's epistemology. They are standard readings in most philosophy of science courses and continue to inform contemporary debates about scientific methodology, rationality, and the nature of scientific progress.

Popper's demarcation criterion is a terminology suggestion related to his commentary and criticism of some thinkers like Marx, Freud and Adler. It's not a very important issue to his epistemology. Science can use many types of criticism, including empirical tests. If Popper was wrong about terminology and some scientific claims can't be empirically tested, that's OK because non-empirical criticism would still work.

My conclusion is that none of these six criticisms actually sounds like relevant, logical criticism of Popper's core epistemology claims. They're mostly either indecisive points (like about some things being difficult) or off-topic points (mostly about human behavior rather than the logic of how knowledge can and can't be created). These points mostly, even if they are correct, aren't a threat to Critical Rationalism. If anyone knows that any of these AI summaries are wrong, or knows of any more relevant criticisms, please let me know.

When I've looked at Popper criticism papers in the past, and books advocating rival schools of thought, including pro-induction textbooks, my general experience has been similar to this: most of it misses the point and is more clearly off-topic than wrong. Many of the criticisms of Popper are things I could concede while still liking Popper and getting a lot of value from his ideas that weren't criticized. And a lot of the positive claims that contradict Popper, like about induction, are too vague to engage with directly. I can respond with meta-criticism about how it's vague or doesn't address key questions, but that is less satisfying than actually being able to respond about epistemology. If anyone knows of some good materials to engage with and/or has enough knowledge about induction and is open to debate, let me know. I've been looking for many years.

I've tried making guesses about induction to add details. I've tried branching (they could mean X, Y or Z, so let me respond to all three). I've tried filling in blanks, writing my own summary of induction, etc. I haven't found this an effective way to enable writing critiques that inductivists accept. I, like Popper, have been accused of attacking a straw man. I think I could pick any of the six texts Gemini listed, criticize it, and be accused of attacking a straw man or dumb text by some inductivists and told that I should have responded to some of the good inductivist literature instead. I don't want to attack a straw man; I don't actually want to guess what inductivists think or fill in blanks for them; I don't want to put words in their mouth; I want them to write it down so I can respond to quotes. I've never been able to find anything suitable to engage with that addresses the key questions necessary for me to write Popperian criticism instead of either asking clarifying questions or filling in blanks myself. I've also asked over a dozen inductivists to provide a realistic, detailed, step-by-step example of using induction, but none of them have ever been willing to try, and I've never found that in literature. Nor have I ever gotten a list of steps to do induction that I could follow. So I've failed to find any abstract induction theory with enough information to engage with well and I also failed to find specific examples or steps for induction. That leaves me wondering what exactly persuaded anyone to be an inductivist.