Talk:Falsifiability/Archive 8

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Given the importance of the distinction between falsifiability, the logical criterion, and the non-logical criterion "ability to be proven wrong", it makes sense to have an article dedicated to Falsificationism, which considers and rejects the latter criterion, but is an important concept independently of falsifiability. Of course, this is not a WP:POV-fork, because it's not the same subject at all and both articles will duly refer to the other—no point of view will be hidden or made less important. This would also help a lot to reduce the false expectation that many readers have when visiting Falsifiability, because the article will have a disambiguation tag at the top such as:

The article Falsificationism could have a similar disambiguation tag at the top, but this is less important. The article itself will, of course, refer to falsifiability. Dominic Mayers (talk) 17:34, 15 December 2021 (UTC)

Another one of your dust bunnies you wish to prop up on a pedestal with Wikipedia? You want to make yourself famous or something? Not only does it not meet Wikipedia's standards for notability, it's not even a thing. Please try to make a case that it meets Wikipedia standards for notability. https://www.onelook.com/?w=Falsificationism&ls=a The only entries are wiki. Circular. Please stop trying to clutter up Wikipedia with trivia and your personal windmills (garbage).

--Doug Bashford — Preceding unsigned comment added by 2607:FB91:7905:9C86:D801:40FF:FE2B:D43B (talk) 10:18, 16 December 2021 (UTC)

Ignoring the false allegations about me, I found a good suggestion here regarding how to proceed for the article: we must provide sources for the subject of falsificationism. Falsificationism is the attempt to provide some formal rules (Lakatos call them inductive rules) to justify the methodological side of science. It's called falsificationism, because the approach acknowledge Popper's point that verification is not possible logically and, therefore, there is no point to expect verifications methodologically, only falsifications can be hoped for. The problem is that rigorous falsifications without the help of methodological decisions (adopted by convention among experts) are also impossible and there are other problems. This is the subject and it is obviously verifiable in the literature. Thanks for the suggestion to list some reliable sources on the subject. I will do that. Dominic Mayers (talk) 13:54, 16 December 2021 (UTC)

One thing to keep in mind is that "falsificationism" is also used to refer to Popper's philosophy, despite the fact that Popper rejected entirely that he had anything to do with Lakatos's falsificationism and did not want to be associated with it. The subject of Falsificationism will obviously not be Popper's philosophy, because it would be a duplication of Critical rationalism. In particular, this means that it's pointless to look at the popularity of the term "falsificationism", because it can mean different things. Dominic Mayers (talk) 14:25, 16 December 2021 (UTC)

List of sources

A point that came out of my search for sources is that "Falsificationism" is closely related to "Falsifiability". Essentially, Falsificationism is an analysis of the application of Falsifiability on the methodological side with a focus on making it rigorous. Because such methodology can hardly be formalized in terms of rules, Falsificationism is typically a criticism. It's not hard to understand the point of view that what is important is the application on the methodological side and if it cannot be made rigorous, then falsifiability has failed. However, this is only a point of view and it cannot be stated in Wikipedia's voice. The other point of view is that falsifiability is a logical criterion that is very useful despite the fact that the methodology is hardly rigorous, because this methodology is nevertheless rational, because science is rational even if we cannot formalize it in a fixed system. Here are the sources that I found:

• Lakatos's The Methodology of Scientific Research Programmes: Volume 1 is of course a primary source on the subject. It's the source that define what the thing is.
• Chalmers's What Is This Thing Called Science? has three chapters on falsificationism. It starts by explaining the logical criterion and its usefulness. Next, it presents different expectations of formalization or rigours at the methodological level and criticize them. This second part is the subject of Lakatos's falsificationism. The first part (which is shorter) is Popper's falsifiability.
• Feyerabend's Against Method contains a criticism of Lakatos's sophisticated falsificationism. Basically, he says that it suffers from the same issue as naive falsificationism. This corroborates Popper's position that the methodology "can hardly be rigorous".
• Thornton's Popper, Basic Statements and the Quine-Duhem Thesis mentions "Popper was aware that there is a significant disparity between precision of the logical analysis of statements contained in his demarcation criterion and the complex, heterogeneous nature of actual scientific practice" and next discusses the problems of falsification in falsificationism.
• Garcia's Popper.s Theory of Science: An Apologia has a chapter on falsifiability that mainly discusses methodological issues, which is the subject of falsificationism.
• Smith's Philosophie of science and its relevance for the social sciences has 4 pages on the subject. It explains what is explained in the current section on falsificationism. By the way, the author says "Lakatos agreed with Kuhn that Popper was wrong in emphasizing falsification as the demarcation criterion between science and non-science". This opinion of Lakatos, which is somehow endorsed by Smith. obviously cannot be stated in Wikipedia's voice. Popper insisted so much that both corroborations and falsifications are important in science. Verifiability does not mean that we must include this kind of confusions in Wikipedia's voice.
• Chapman's Language and Empiricism: After the Vienna Circle endorses the point of view that Popper was originally a dogmatic falsificationist. Even nowadays, this view advanced by Lakatos is wide spread. I provide this source as an example. Of course, Popper was outraged by this view and completely denied that he had anything to do with any kind of Lakatos's falsificationism. Yet, Wikipedia should not hide any point of view, especially not when they are wide spread, but it should not be presented in Wikipedia's voice and the other point of view should also be presented.
• Ploch's Metatheoretical problems in phonology with Occam's Razor and non-ad-hoc-ness clearly presents the view that Popper never was a dogmatic or naive falsificationism and also never believed in rigorous falsifications.

These are the first that I saw in my shelves. Also, while looking for these sources, I realized that we have already a large section on falsificationism in the article. Are we questioning the verifiability of the section? Dominic Mayers (talk) 20:56, 16 December 2021 (UTC)

After all that, you are the only one who knows what the hell you are talking about. Except I understand you are trying to change the subject. You have still not addressed Wikipedia standards and values. Whether anything you said is true, or verifiable, or not, is irrelevant. It's trivia and not notable. Therefore it still does not meet Wikipedia standards. Please find your hobbyhorse elsewhere.--Doug Bashford. — Preceding unsigned comment added by 172.58.37.57 (talk) 23:53, 21 December 2021 (UTC)

Doug Bashford said: After all that, you are the only one who knows what the hell you are talking about. Speak for yourself, Doug. Biogeographist (talk) 03:02, 22 December 2021 (UTC)

I have been watching the talk page, but I don't have an opinion about whether to split off "Falsificationism" into a separate article. Here is another source: "Confirmationism vs. Falsificationism" (1977) by Nicholas Rescher. It presents confirmationism and falsificationism as ideal types exemplified by followers of Carnap and Popper respectively (not Carnap's and Popper's whole works, but just the parts relevant to this distinction). Rescher is not concerned with the "logic versus methodology" distinction. For Rescher, confirmationism and falsificationism are two ideal types or emphases in scientific epistemology, and he argues that a synthesis of the strengths of both emphases is needed, as phases in a dialectical process of knowledge generation. I don't mean to imply that Rescher is "right"; I am only adding his perspective to the conversation here. Rescher's perspective is a little like the view that "falsificationism" is a label for Popper's philosophy, except that Rescher is abstracting an ideal type from Popper, and not doing a detailed exegesis of Popper's philosophy. Biogeographist (talk) 03:02, 22 December 2021 (UTC)

I just read the section "Confirmationism vs. Falsificationism" of his book Dialectics: A Controversy-Oriented Approach to the Theory of Knowledge. Rescher defines a falsificationism that is very much in line with Lakatos's falsificationism. He also refers to Popper insistence on falsifications, but he does not mention that Popper does that only in the context of a rigorous logic. As you say, he does not consider the distinction logic vs methodology. The methodological side in Popper's philosophy uses as much, if not even more, corroborations than falsifications: the construction of the edifice of science depends on corroborations, because falsifications when observed only destroy. But, his criticism applies very well to Lakatos's falsificationism (the role of falsifications in the methodology) and it could be included as such. My only concern is that we might not be faithful to the original intention of the author, because the section is part of a chapter that is centred on "The disputational model of scientific inquiry". For me, the interesting connection seems to be between Critical rationalism, which is really Popper's methodology (not falsificationism), and the "disputational model of scientific inquiry", because these two views of science appear related. We can certainly look whether some reliable sources made the connection. Dominic Mayers (talk) 18:22, 22 December 2021 (UTC)

The second paragraph of Rescher's introduction to the book explicitly mentions the similarity between Rescher's model and Popper: "Since the days when Hegel reestablished dialectic as a central theme of modern philosophy, various schools of thought have viewed this discipline in very different ways: as part of the causal pattern of historical development (the Marxists), as a branch of philosophical ontology (contemporary German neo-Hegelianism), as a sector of rhetorical tradition (the "New Rhetoric" of Chaim Perelman and his associates), and as a way of systematizing the testing process for scientific theories (Karl Popper and his school). The present study of dialectics has kinship with the last two of these approaches." So Rescher knows that what he calls falsificationism is not equivalent to Popper's overall mature philosophy (critical rationalism). Popper comes up in several other places in the book. Nevertheless, Rescher's association of falsificationism with "followers of Popper" does reflect what Popper is famous for: falsifiability, not confirmability/corroborability.

You said: He also refers to Popper insistence on falsifications, but he does not mention that Popper does that only in the context of a rigorous logic. Rescher addresses the logical aspect but not in this section of the book: chapter 4 is on "dialectical logic", which I guess is a kind of paraconsistent logic for the dialectical process? (Correct me if I'm wrong; you are the logic expert.) Rescher is very aware of the logical issues, and he knows exactly how his system relates to strictly deductive logic, but the section of the book on "Confirmationism vs. Falsificationism" does not go into the details. Biogeographist (talk) 19:14, 22 December 2021 (UTC)

Rescher wrote as a way of systematizing the testing process for scientific theories (Karl Popper and his school). He must refer to Lakatos, who, despite his different with Popper, is often seen as a popperian. Except for Lakatos, most of his students were against systematic methods, certainly Feyerabend was. Lakatos proposed (sophisticated) falsificationism as an attempt to provide the missing rigour (but Musgrave, Feyerabend, etc. considered that he did not succeed and Lakatos partially agreed). To my knowledge, Popper only referred to classical logic. I am not aware that he flirted with paraconsistent logic. In my view, these are attempts to compensate for the limit of classical logic. I would say that Popper's main point is that we should not try that. For Popper, it's not within the scope of epistemology or the philosophy of science to try to put science within a systematic system, whether it is called paraconsistent logic, inductive logic or whatever. In other words, his solution to the problem of induction is simply that we should not try to solve it, because we don't use these other kinds of logic in practice. We use patterns, which we may call "inductive rules", but there are no real rules there. They are just tools, like statistical tools, etc. that we use whenever they apply given the background knowledge, but the whole process cannot be turned into a paraconsistent or an inductive logic. He was happy with the fact that we could locate where classical logic plays a role in the scientific process and used it as criterion of demarcation. Dominic Mayers (talk) 20:30, 22 December 2021 (UTC)

My personal view on this is that Popper had the correct global approach, which is not to try to attack the big problem of putting all of science within a systematic logic of any kind, paraconsistent or whatever. Instead, he tried to locate where classical logic can play a role within science as a system that has no guarantee at all to succeed (probabilistically or otherwise), but seems to correspond to what we do successfully in practice. He "justified" this practice using evolutionary arguments or similar arguments that provide no guarantee. In fact, if his approach would have been more appreciated, we could have progressed by trying to see a little more precisely what roles logic and rigorous mathematics play within this otherwise non systematic system, still with a modest goal in mind, not trying to make the whole thing rigorous and systematic. Dominic Mayers (talk) 20:52, 22 December 2021 (UTC)

Popper was aware that the evolutionary argument provided no guarantee. This is why he wrote that he did not consider this argument as an essential part of his epistemology. Dominic Mayers (talk) 21:18, 22 December 2021 (UTC)

You said: He must refer to Lakatos, who, despite his different with Popper, is often seen as a popperian. No, Rescher never mentions Lakatos. He does say, in a footnote on page 55: "For a vivid picture of Popperian methodology in the light of a dialectical perspective of consideration see Hans Albert, Traktat über kritische Vernunft (Tübingen, 1968)."

You said: My personal view on this is that Popper had the correct global approach, which is not to try to attack the big problem of putting all of science within a systematic logic of any kind, paraconsistent or whatever. I am not sure how this is relevant to the question at hand, but Rescher is not "putting all of science within a systematic logic". For Rescher, an important reason for dialectical metatheory in epistemology is that logical theory is insufficient. (In fact, as I think we have discussed elsewhere, logic is insufficient for any account of rationality because one has to account for the metacognitive generation and control of the logic, not just the logic itself.) For example, on page 118, the page just before the section on "Confirmationism vs. Falsificationism", Rescher wrote:

We can determine the probative status of the thesis only "in context"—only relative to the historical concreta of its probative background. Its status relative to a course of argument hinges on its antecedents within this argument in such a way that these cannot be put aside: it becomes important to consider not just the content of the evidence, but also the historical development of the evidential situation. The whole course of argumentation must be brought into the assessment of the probative status of the conclusion.

These general considerations regarding the probative structure of dialectical contexts have one particularly significant bearing in the specific setting of a disputational model of natural science. They mean that we can never really assess the probative standing of a scientific thesis outside its historical context—outside the background of the actual course of controversy and discussion from which it has emerged. The real-life sequence of argumentation and debate that has brought us to where we are becomes a crucial factor in the rational assessment of this position. The probative or evidential situation in this domain is context-dependent on the details of the historical background in a way that finds no parallel in the deductive sciences that have often (and mistakenly) been taken as the model of scientific rationality in general.

These considerations go very much against the grain of the positivist philosophy of science. ... Our present position goes against the grain of this view, arguing (on roughly evolutionary grounds) that some issues of historical development are bound to be probatively significant.

It's the job of philosophers to be as rigorous and systematic as possible, but systematicity is not equivalent to logicality nor is it equivalent to an infallible guarantee of anything. The best we can do is the best we can do. Biogeographist (talk) 21:37, 22 December 2021 (UTC)

Do the modifications as you wish, if you think that Popper insisted on falsifications in opposition to corroborations in his methodology as suggested by Rescher. Fighting against a wide spread view is not easy, especially if it seems shared by highly notable philosophers such as Rescher. My point is that, even if it is wide spread, it is still a controversial view and it should not be stated or even implied in Wikipedia's voice. It's fine to attribute this view to Rescher and to many others, but it should be attributed and the fact that Popper said the opposite should also be mentioned very clearly. But I am not fighting this battle anymore. Nevertheless, I must add key sentences: Yes, Popper emphasized the importance of severe tests. Yes, Popper insisted that the only logically meaningful outcome of a test is falsification. Yes, in this manner, it gives an emphasis on falsifications. No, it was not in opposition to the methodological importance of corroborations. Popper insisted that corroborations are as important as falsifications in his methodology. This should be obvious given the fact that falsifications by themselves can only destroy the edifice of science. Dominic Mayers (talk) 05:28, 23 December 2021 (UTC)

Perhaps a good example of a "follower of Popper" who comes very close to Rescher's depiction of falsificationism is David Miller, whose book Critical Rationalism: A Restatement and Defence (Chicago: Open Court, 1994) is currently described by Miller's Wikipedia article as "an attempt to expound, defend, and extend an approach to scientific knowledge identified with Popper". Section 1.2 in that book, titled "Outline of Falsificationism", says (pp. 6–7):

For justificationists a hypothesis has to pass tests, or be confirmed, or in some other way be touched with grace, if it is to be admitted to the realm of scientific knowledge; if it fails these tests, or is disconfirmed, or even if it fails to be confirmed; it is excluded. Falsificationists, in contrast, think that a hypothesis need submit to tests only after it has been admitted to science. If it fails any of the tests to which it is put, then it is expelled, removed from science; if it passes them all, then nothing happens—that is to say, it is retained. The passing of tests therefore makes not a jot of difference to the status of any hypothesis, though the failing of just one test may make a great deal of difference. For justificationists, on the other hand, the passing of tests is quite as important as the failing of tests, for it is precisely this that determines whether a hypothesis is admitted to the body of science. Of course, justificationists need expulsion procedures as well as entrance examinations; for even the most rigorous entrance examination fails to guarantee the quality of a successful candidate. But the expulsion procedures are the sole means that Popper allows for the control of scientific knowledge. He adopts a policy of open admission, subject naturally to the condition that no hypothesis may be admitted unless there is some way in which, if necessary, it may be expelled.

Miller's phrase "a policy of open admission" is a good label for what Rescher finds insufficient in falsificationism: for Rescher, there is not a policy of open admission in scientific epistemology, because there is also a complementary confirmationist aspect to science, which seeks "reasonable" or "sensible" hypotheses that are "more or less consonant" with what is well known about the way things work.

Later in the same section, Miller says (pp. 11–12):

The whole business [of science] can be explained, quite satisfactorily, without any reference to certainty, probability, confirmation, support, reliability, confidence, justification, good reasons, or knowledge.

Again, this statement by Miller summarizing falsificationism (which Miller associates with Popper) is entirely consonant with Rescher's portrayal of falsificationism.

John W. N. Watkins's review of Miller's book accuses Miller of misrepresenting Popper in this regard (though Popper hardly had a closer follower than Miller, so I doubt that Miller's interpretation of critical rationalism can be considered illegitimate). Watkins says:

Since an unwary reader of this book may gain the impression from its many acknowledgements and salutations to Popper that it is essentially a restatement of Popper's philosophy, I may point out that it is very unPopperian in important ways. (1) Popper did not endorse a policy of waving in any old rivals to a currently prevailing theory, provided only that they are falsifiable. He rightly considered it an important feature of his methodology that it enables us to say what kind of theory would be better than the prevailing one, provided it passes tests ([1963], pp. 217, 240f.). (2) A main component of Popper's methodology was his theory of corroboration (see the concluding chapter and appendix *ix of his [1959]); corroborations are what ultimately govern the rational acceptance of theories. This disappears without trace in Miller's 'restatement'. In his index there are six entries against 'corroboration', five of which refer to places where an author is being quoted or reported. The sixth comes in the course of an examination of my [1984]. I had tried to give a fresh answer to the question, 'Why do corroborations matter?' Miller writes: 'The answer is that corroboration doesn't matter' (p. 120). (3) Popper had a horror of anything like rationality-scepticism. He insisted that a theory's being currently the best corroborated, while not justifying the theory, does justify a preference for it over its rivals. He may not have kept these two kinds of justification as separate as he should have done, but his philosophy allows there to be sufficient reasons for accepting one theory as better than its rivals at the present time. Of course, Miller is free to make what changes he likes; but in a book which purports to present 'a sounder and less blinkered appreciation of Popper's work than it has previously enjoyed' (p. x), I don't think he should silently ditch large chunks of it. ... A more accurate title for the book would have been: 'What Remains of Critical Rationalism when the Rationalism is Dropped'.

— Watkins, John W. N. (December 1995). "Book review: Critical Rationalism: A Restatement and Defence, by David Miller". The British Journal for the Philosophy of Science. 46 (4): 610–616. doi:10.1093/bjps/46.4.610. JSTOR 687902.

So there is indeed a prominent follower of Popper who portrays falsificationism as Rescher does, though Watkins disagrees that Miller's account of falsificationism is Popperian. Still, this perspective on falsificationism should not be dismissed as unimportant, given David Miller's stature among critical rationalists. Biogeographist (talk) 15:36, 26 December 2021 (UTC)

@Biogeographist: First, let us clarify a possible very superficial misunderstanding. Certainly, for any kind of reasonable philosopher of science, including any popperian, it's good for a theory that it succeeds in severe tests and, therefore, corroborations are meaningful. It might only be meaningful, because the theory maintains its status after the corroborations, but it is nevertheless meaningful. By analogy, it might be the case, that only hot temperatures have an effect on ice cubes, but that does not mean that cold temperatures are meaningless for ice cubes. This is valid even when falsifications are not statistical. We can go further than that when falsifications are statistical. Again for any kind of reasonable philosopher of science, including any popperian, for methodological considerations, a scientific knowledge must be repeatable and if the theory is rejected once in these repetitions and corroborated all the other times, the corroborations do matter. However, in this case, some background statistical assumptions are needed, but we make this kind of assumptions all the time in science. Miller's point seems to be simply that, unless we rely on strong statistical assumptions, a theory is still not logically confirmed after many corroborations and, therefore, corroborations by themselves make no difference. It would be a very superficial misunderstanding to assume that popperians are not aware of the role of assumptions in statistical tests and the importance of corroborations in these tests. To sum up, the real issue is not whether or not corroborations are important in science. It's completely the wrong debate. Yes, logically, there is an asymmetry, but this is just like the asymmetry between hot and cold in the above analogy. Once this is understood, then the next step is to realize that Popper, just like most philosophers such as Rescher, recognized that logic and certainly falsifications alone are not sufficient to explain globally the progress in science. This is why he emphasized the role of critical discussions. I am not expert in ancient dialectic, but I suspect that Popper's use of critical discussions is much closer to this ancient methodology than to any modern form of paraconsistent logic. It is not a replacement for logic or an alternative form of logic, but a tool that complement logic by unfolding the rationality within us that cannot be formalized by logic. But Popper would emphasizes that by itself this is not an interesting philosophy at all. He viewed his main contribution in the localization of the role of classical logic and rigorous statistical tests within this overall process, which must involve critical discussion. BTW, for Popper, critical discussions do not imply a social phenomena. He often mentioned that a scientific can do that alone by himself, but that it was better done with others. (I mention that, because the idea that Popper associated "truth" with "socially accepted" is another misunderstanding of Popper.) Dominic Mayers (talk) 19:55, 26 December 2021 (UTC)

I personally agree with everything you said, but Watkins seems to be saying that Miller doesn't agree that it's good for a theory that it succeeds in severe tests and, therefore, corroborations are meaningful. Biogeographist (talk) 21:12, 26 December 2021 (UTC)

Miller's view

I just want that we focus on a common objective for this article and possibly a split of falsificationism. The first goal should be to explain Popper's falsifiability and what it accomplishes. The entire article should be centred on that. Even opposite views should be centred on that, in negative manner perhaps, but presenting opposite views helps to clarify each view. Neutrality here does not mean presenting Bunge's, Rescher's, Miller's, etc. views. Popper is often mentioned, but this is not a lack of neutrality, because it's Popper's falsifiability. I don't think we should have sections or even paragraphs on other's views only for the sake of neutrality. Different views should be presented because they are different views about falsifiability and its applications, not because Bunge, Rescher, Miller, etc. are important philosophers with different perspectives. Otherwise I fear the article will lose its focus. This is why, when I read your comments, I am asking myself what part of the article is incorrect or needs an opposite view that would help us with the simple goal of explaining falsifiability and its applications? To achieve this goal is particularly important to not conflate different notions as being the same and not present in Wikipedia's voice what could possibly be confusion. Yes, there is a lot of confusion on the subject in the literature and I believe the article should not entirely hide it, but yet these possible confusions should not be presented in Wikipedia's voice and also we are not forced to present (with attributions) all possible confusions. We need to keep it simple. We do not want to hide important issues, but not all possible confusions are important. Also, what worries me is that looking at possible confusions is without end in this subject. We could keep playing this game for ever, with Bunge, Rescher, etc. and I will get exhausted. Miller's book is a tough one, because, as you know, it is a defence, and therefore he considers many subtle attacks on Popper's philosophy. Before, we can understand his response, we must look at the subtlety of the attack and often the response is simple, but the context was complicated. I don't think that the article should follow this style. Regarding your specific issue, Miller often follows Popper regarding the separation between the logical side and the methodological side, but he does not say it explicitly each time. In this case, it is easy to imagine that Miller has in mind the simple mathematical or logical situation where there is a universal law and a few instances that corroborate the law, without any assumptions whatsoever, just this abstract mathematical situation. Logically, the law still remains unjustified and the only thing that changed is that we know some instances that corroborate it. We can say that the law remains "totally" unjustified, because if we try to give a measure to this change, it is essentially nothing. That's all what he means when he says that corroborations are useless. He certainly does not mean that in the methodological side, in which we always make plenty of assumptions, the corroborations are never useful. I don't think covering this would help people to understand falsifiability. Dominic Mayers (talk) 23:50, 26 December 2021 (UTC)

My motivation for saying what I have said in this section has been to provide information that may be helpful when asking whether to start a separate article on falsificationism and what content that article would include. I didn't make any argument for or against such an article because I myself didn't see very strong reasons for or against a new article. However, if I were pressed to make a decision about this issue, and if the goal of the Falsifiability article should be to explain Popper's falsifiability and what it accomplishes as you said, then it's my opinion that the section on "Lakatos's falsificationism" should be split off into a new Falsificationism article, since I don't think that section does much to achieve what you say should be the goal of this article. However, ideally the exposition of Lakatos's falsificationism would be only one part of the new Falsificationism article, since there are other important views such as that of Miller, who presents his own explanation of falsificationism and repeatedly calls it Popper's (e.g., "falsiticationism—Popper's theory", ibid., p. x) even though Popper said in 1978 that he "tends to avoid" the term; but at the same time Watkins' criticism that Miller's falsificationism is not Popper's view should be mentioned if Miller's falsificationism is mentioned.

You said: Miller often follows Popper regarding the separation between the logical side and the methodological side, but he does not say it explicitly each time. In this case, it is easy to imagine that Miller has in mind the simple mathematical or logical situation where there is a universal law and a few instances that corroborate the law etc. This interpretation is charitable toward Miller, but like Watkins I am not inclined to be so charitable; one reason is that the first sentence of Miller's first chapter says: "The task of empirical science, like that of other investigative disciplines, is to separate as thoroughly and efficiently as it can the true statements about the world from those that are false, and to retain the truths." Leaving aside possible criticisms of this definition, Miller's subject is all empirical science, not just universal hypotheses. In the next paragraph he says: "Unfortunately for such a project, conclusive verification by empirical means turns out, quite trivially, not to be practicable, anyway above the observational level. Universal hypotheses cannot be conclusively verified" etc. This would support your interpretation if the only statements in science were about "the observational level" and "universal hypotheses", but surely empirical science deals with more than these two extremes (here I am thinking of Mario Bunge, who argues for this point in several books), and as Watkins asked in his book review (repeating a question that Watkins had posed a decade before without response from Miller), what should we do "in the all too likely event that we have a plethora of unrefuted and mutually conflicting hypotheses" (p. 613)? The point of all this is to suggest that the criticisms of Watkins are not so easy to brush aside as far as I can see, and that both Miller's portrayal of falsificationism and the objections to it would be worth including in the Falsificationism article.

In any case, the new Falsificationism article would not be about "the ordinary capacity to be proven wrong" (as your proposed hatnote above says), since falsifiability plays the central role in Miller's falsificationism. It would also not be about "the attempt to provide some formal rules (Lakatos call them inductive rules) to justify the methodological side of science" (as you suggested above), since that is obviously not at all what Miller's falsificationism is about. Biogeographist (talk) 01:54, 27 December 2021 (UTC) and 02:46, 27 December 2021 (UTC)

We need articles that have a clear focus, because this makes it easier for the audience and for us, certainly for me. This is why I want a simple and clear focus for this article and I am glad that you are not against this. We should try to do the same for the falsificationism article. I oppose my approach to a different approach that considers a concept identified by a term such as falsificationism and then try to include what different philosophers wrote in reference to this term or concept, but in the context of their philosophy and with main goal to explain their philosophy. So, in this approach, we have a concept vaguely identified by the terms "falsificationism" and then we consider what Miller, Watkins, Bunge, Rescher, Popper, etc. wrote about this term or concept in their perspective. In my view, this is a receipt to create a very hard to understand article with very little central focus, because each of these philosophers have a completely different perspective. It's a very difficult task to write a simple article while having to consider such a broad range of perspectives. We will have a much simpler article by considering a question that a particular philosopher tried to answer. If the work of this philosopher was considered important, we can expect that other sources will be centred on this problem. I am not saying that we do not consider what other philosophers have written, but now, the big difference, is that it's this specific question, not the term that guides us when we consider what other philosophers have written: it should be clear enough that the content that we consider focalize on the question, in fact, ideally, to be sure that we focus on the same question, the content should refer to the original philosopher that considered the question. I would not worry about failing to present some view points that use the term and thus apparently refer to the concept, because if it does not address clearly the question, it must address another question and there will be another article where this other question is considered. We can create links between these articles so that we still have a wholeness even though each article has a clear focus. When I thought about the falsificationism article, I thought about the specific question that Lakatos tried to answer with the premises that he used, etc. It has to be made precise, because if we are too vague then Lakatos's question could become "what is science" and this is too large as a scope. Anyway, in particular, if we consider the question that Miller addressed in his book and the question that Lakatos tried to answer, it's immediate that they are not the same question at all: Miller totally reject induction whereas Lakatos was trying to find some inductive principle that would make rigorous Popper's notion of verisimilitude, at the least, it was it starting motivation. Lakatos approach, as is well known, was based on how series of theories based on a same core evolve or degenerate. It did not succeed as a way to obtain an inductive principle, but I think there is enough for an nice and well centred article on the subject. We can certainly mention Miller, but more as a totally opposed view in which Lakatos's problem was considered a false problem. What would be an example of a lost of focus is starting to cover Miller's falsificationism as if it was naturally in the same scope and then do the same for the work of Mario Bunge, etc. In my view, this would create an article that is completely missing a central focus. Dominic Mayers (talk) 03:41, 27 December 2021 (UTC)

That's reasonable. All I would qualify is that if the article is about a particular thinker's perspective then it can't be called "Falsificationism". If it's about Lakatos's perspective, then it probably should be a section in Imre Lakatos or else a new article could be written about "Falsification and the Methodology of Scientific Research Programmes" that also discusses the secondary literature on how the paper relates to Popper's perspective. Lakatos's paper is certainly notable enough for its own article. Then Falsificationism could be a disambiguation page that says something like:

Falsificationism may refer to:

• Critical rationalism, an epistemological philosophy founded by Karl Popper
• Various theories of scientific progress in "Falsification and the Methodology of Scientific Research Programmes" by Imre Lakatos

So effectively there would be no article about "Falsificationism" nor a redirect from Falsificationism to any single article. Biogeographist (talk) 04:16, 27 December 2021 (UTC)

Yes, I had in mind and perhaps even explicitly suggested "Falsificationism (Lakatos)". I personally see the parenthesis as a precision on what Falsificationism means, but I admit that this view of mine ignore Miller's choice to use Falsificationism to refer to Popper's philosophy. I consider that it was a very bad choice of terminology by Miller, because then the same term is used to refer to two essentially opposite philosophies regarding an induction principle. But, even though it's a very confusing choice, I agree that (through its editors) Wikipedia can decide not to ignore it. Some would say that Wikipedia has no choice, but I think that given the high level of confusion that this creates, Wikipedia has the choice. In all cases, certainly this confusion should not be stated or implied in Wikipedia's voice. The distinction, even the opposition, between Lakatos's falsificationism and what we strangely call Popper's falsificationism should be apparent so that the conflation between the two can be clearly attributed, i.e., not at all be a fact. Miller did not conflate the two. On the contrary, he says that he considers Lakatos's work as a "falsification of falsificationism". However, Lakatos famously conflated the two and many followed him. Again, Miller made a terminological mistake: Lakatos could not falsify "falsificationism", because he invented the term and the concept. He should have given to Cesar what belongs to Cesar and used another term altogether for Popper's philosophy. Dominic Mayers (talk) 10:48, 27 December 2021 (UTC)

A disambiguation page might not be needed. A disambiguation tag at the top of Falsificationism (Lakatos) seems enough and Falsificationism could redirect to Lakatos's falsificationism. The Wikipedia's editors that intended to refer to Popper's philosophy would have to make it clear in the text they write and that's should be encouraged. If they don't, the disambiguation tag at the top of Falsificationism (Lakatos) should take care of the confusion very clearly. Dominic Mayers (talk) 11:07, 27 December 2021 (UTC)

No, I oppose redirecting Falsificationism to an article about Lakatos. I am not convinced that Lakatos's meaning of falsificationism is the most common meaning. As far as I can tell, the term is more commonly used to refer to Popper's view. Biogeographist (talk) 13:59, 27 December 2021 (UTC)

I don't really mind about choice of terminology as long as it is useful, not ambiguous and that we use it consistently. So, I don't mind what choice we make, but if we decide that falsificationism cover Popper's falsificationism, then I want to be sure that you understand that it cannot cover the opposite view presented by Lakatos, because this render the term falsificationism meaningless or worst a confusion stated in Wikipedia's voice. It's not because the two meanings are used in the literature that Wikipedia cannot make a choice. On the contrary, Wikipedia must make a choice so that it has a useful terminology. Dominic Mayers (talk) 14:27, 27 December 2021 (UTC)

No, it's not true that Wikipedia must make a choice in this case any more than it "must make a choice" in any other case where there is a term with multiple meanings and a disambiguation page is used or else the multiple meanings are discussed in the same article. My proposed solution in this case is a disambiguation page to accommodate your desire not to discuss the multiple meanings in a single article, but I am not opposed to the latter solution.

By the way, it is not true that Lakatos invented the word "falsificationism". It was used earlier in, for example: Buchdahl, Gerd (March 1965). "A revolution in historiography of science". History of Science. 4 (1): 55–69. doi:10.1177/007327536500400103. "the falsificationist disciples of Professor Popper" (p. 59), "the falsificationists fare no better" (p. 60), "But how does this affect the falsificationist?" (p. 61), "a disciple of Popper, and we might expect him therefore to defend some form of falsificationism" (p. 64), "falsificationism or systematic 'experimental criticism', a variant of Popper's view" (p. 65), "no attempt to offer falsificationism as a methodological panacea" (pp. 66–67). Biogeographist (talk) 15:10, 27 December 2021 (UTC)

Some times I say instead that Lakatos popularized the term and the concept. This is the point that really matters in our discussion. I agree that "falsificationism" is more often used to refer to Popper's philosophy. However, this is not sufficient, because in most cases, perhaps Miller is the exception, when people refer to Popper's falsificationism they actually describe Lakatos's falsificationism. It's untypical that people refer to Popper's falsificationism with an emphasis on the distinction between the logical side and the methodological side and the fact that the whole point of Popper's philosophy is to locate the limited but crucial role of the logical side in the scientific methodology, which uses this rigorous component within rational critical discussions. Instead, they refer to falsificationism to mention its falsification problems. I do not mean only the experimental problems as described in Lakatos's dogmatic falsificationism, but also the limitation of the logic of falsifications (and corroborations) as described in Lakatos's naive falsificationism. The latter is what Rescher does, for example. Therefore, going along with this usage of "falsificationism" appears to me to be a support in Wikipedia's voice for this wide spread confusion. Dominic Mayers (talk) 19:00, 27 December 2021 (UTC)

Some times I say instead that Lakatos popularized the term and the concept. If that were true, then Miller wouldn't use the term so proudly. The truth is that the term has multiple meanings. It's not owned by Lakatos. Biogeographist (talk) 19:22, 27 December 2021 (UTC)

In fact, we already have everything we need for a disambiguation page. It could be created at any time, with no need to wait for a decision about what to do about the Lakatos section that is currently in this article. Here's a revised proposal for the disambiguation page that is feasible right now:

Falsificationism may refer to:

• Critical rationalism, an epistemological philosophy founded by Karl Popper
• Three models of scientific progress in "Falsification and the Methodology of Scientific Research Programmes" by Imre Lakatos
  • Dogmatic falsificationism
  • Naive falsificationism
  • Sophisticated falsificationism

I have boldly created a first version of the disambiguation page at Falsificationism. Biogeographist (talk) 17:43, 27 December 2021 (UTC) and 18:44, 27 December 2021 (UTC)

I maintain my position that saying "falsificationism" may refer to "critical rationalism" seems to be only valid in the case of Miller and incorrect in most other cases and therefore is supporting in Wikipedia's voice a wide spread confusion. The fact that Miller uses falsificationism to mean Critical Rationalism is not sufficient to make it a disambiguation item in a disambiguation page, especially given that the vast majority use this to refer to Lakatos's falsificationism and its falsification problems. Dominic Mayers (talk) 19:10, 27 December 2021 (UTC)

I disagree that that the vast majority use this to refer to Lakatos's falsificationism. What do you think, for example, of the definition of falsificationism here in Oxford University Press's Dictionary of the Social Sciences? This is the more typical definition of "falsificationism" without Lakatosian qualifiers. It's explicitly associated with Popper. Biogeographist (talk) 19:14, 27 December 2021 (UTC)

Please note that the fact that many authors in this vast majority conflate this problematic falsificationism with Popper's philosophy is what makes the problem serious. I haven't look at this entry yet, but if it is precise enough, then it's very likely that, even though it is attributed to Popper, they actually refer to Lakatos's problematic falsificationism. If it is not precise enough, then it's not conclusive and the explicit attribution to Popper is besides the point. Again, this attribution to Popper is part of the confusion that cannot be stated in Wikipedia's voice. Dominic Mayers (talk) 19:18, 27 December 2021 (UTC)

Wikipedia does not exist to WP:RIGHTGREATWRONGS. You may be passionately against associating falsificationism with Popper's philosophy, but many people do in reliable sources, and Wikipedia neutrally reflects those sources. Biogeographist (talk) 19:29, 27 December 2021 (UTC)

No, what you say is only true when a content is not controversial and even if a majority presents a point of view, it can be controversial and it's exactly our case. Note here that Popper is very clear about the importance of a distinction between the logical side and methodological side and that he refers to the methodological side to discuss the falsification problems. Thornton mentioned this emphasis. There is a serious and very real controversy here. Wikipedia's rule is that a controversial point of view cannot be stated or implied in Wikipedia's voice. Now, I have read the entry. It is a kind of entry that if it was further elaborated, it would mention that falsificationism suffers from falsification problems. It is a too small entry to be useful in our current discussion. Dominic Mayers (talk) 19:35, 27 December 2021 (UTC)

So how do you propose to resolve the controversy without treating falsificationism as if Lakatos owned the term? Biogeographist (talk) 19:42, 27 December 2021 (UTC)

You are right that Lakatos does not own the term. It was perhaps used before Lakatos with the sense that was also given to it by Lakatos. So, perhaps "Falsificationism (Lakatos)" is misleading. It was only a way for me to identify the usual sense of falsificationnism in the literature, unfortunately often wrongly conflated with Popper's philosophy. Dominic Mayers (talk) 19:54, 27 December 2021 (UTC)

Falsificationism

My idea would be to use Falsificationism for falsificationism as used by Lakatos, even though he does not own the term. You right that Wikipedia does not have to choose a terminology, but it can if it useful and avoid stating in Wikipedia's voice a confusion. The disambiguation tag at the top of the article would say something like

I changed my position. I consider now that it is simpler to consider that falsifiability is a very special solution among different approaches in falsificationism. It remains very wrong to identify Popper's philosophy with all these proposals that fail to avoid the falsification problems or to conflate it with a philosophy that needs to be complemented by some form of confirmationism. Dominic Mayers (talk) 20:46, 27 December 2021 (UTC)

I still think that to use Falsificationism for falsificationism as used by Lakatos is a bad idea, because any article with that name should accommodate views such as Miller's. However, I could accept a new redirect if it were to a section of Imre Lakatos with a hatnote at the top of the section that says, e.g., "Falsificationism" redirects here. For the approach in Karl Popper's philosophy, see Falsifiability. This is still problematic but is no worse than the previous redirect to a section of Falsifiability, and would advance your stated goal of keeping this article focused on Popper. Biogeographist (talk) 21:55, 27 December 2021 (UTC)

Actually, given that I changed my position, now "Falsificationism" can be formally said to cover Miller's falsificationism, but I maintain that it is a problematic association, because one should not associate a special case to the general case, especially in reference to problems that are seen in all the other cases and only in the other cases. For example, Rescher refers to falsificationism as a philosophy that can be seen in opposition to confirmationism, ignoring that Popper's philosophy has a methodology that covers this needs. Popper's falsificationism is Rescher's falsificationism + critical discussions that support a form of quasi-induction that makes a duality with confirmationism pointless. So, formally, I am not against your request that falsificationism includes Miller's or Popper's falsificationism, but I still maintain that it is done in a misleading way in the literature and this confusion should not be supported or implied in Wikipedia's voice. Dominic Mayers (talk) 22:16, 27 December 2021 (UTC)

I don't think that either Miller or Rescher are being misleading in their usage of the term "falsificationism". You think Rescher is being misleading, and Watkins thinks Miller is being misleading, but I have no doubt that Rescher and Miller would object to being called misleading. They are both well-informed experts who are well-qualified to choose to use the term "falsificationism" as they do. This is not really an issue of who is right (especially since Popper and Rescher have similar approaches to the disputational process overall, whatever terms they may use to refer to the components of the process); it's an issue of which views are important in relation to the term falsificationism. I say that they are all important. And when I say that an article on falsificationism should accommodate views such as theirs, I don't mean that it should present their views in Wikipedia's voice, only that their views should not be excluded. Biogeographist (talk) 22:35, 27 December 2021 (UTC)

In any case, I believe my proposal meets all your requirements. As I explained, I am not formally against that falsificationism is a large category that includes Popper's philosophy. This is a big change in my position. Also, there is no reference to Lakatos in my description of the article Falsificationism and in the disambiguation tag (see above), which was an important requirement for you. Of course, simply by WP:proportion and the fact that the special case falsifiability is already well covered, the article will focus a lot on Lakatos's falsificationism. So, I feel like I had compromised a lot, but I hold strong on the fact that any confusion created by associating falsifiability and Popper's philosophy to the general case and ignoring essential aspects of Popper's philosophy, which is a unique special case, cannot be stated or implied in Wikipedia's voice. The "implied" is important here, because often choice of name, etc. can imply a point of view in a manner that is worst than something stated explicily in Wikipedia's voice. For example, given that Popper's philosophy is a very special case, referring to it as falsificationism is vague and can be used to create a confusion. Dominic Mayers (talk) 22:45, 27 December 2021 (UTC)

Above you said to me: We could keep playing this game for ever. Now I definitely think that you are the one who is playing a game. You proposed to use Falsificationism for falsificationism as used by Lakatos, i.e., to create an article called "Falsificationism" about Lakatos. And now you say you made no reference to Lakatos. You are playing your game and are completely ignoring my point. I hope you're having fun. There is no way I am going to agree with you when you play like this. Biogeographist (talk) 22:56, 27 December 2021 (UTC)

I don't see why you see a problem. I just told you that I changed my mind on two things: (1) Falsificationism can be a large category and (2) the scope of the article Falsificationism does not have to refer explicitly to Lakatos. That should make you happy, but instead you say I am playing a game. I am lost. It's strange because I thought we were moving toward an agreement. Is the problem that I hold strong that Popper's philosophy is a special case and there is no point in referring to it in such a general manner and in general any confusion possibly created in this manner should not be stated or implied in Wikipedia's voice? BTW, I am still maintaining my position that Falsificationism should have a clear focus. But, you might like a different approach where the content is organized in terms of different perspectives: Rescher, Lakatos, Bunge, etc. It's not my way, but if you want to write the article with this style, go ahead. It will be just difficult for me, not only to work on it, but to actually read it. I am a person that is driven by a question given in one perspective and when I work on this goal, I translate every thing into this perspective that applies to the question. Therefore, I prefer read sources that have done the work to adapt themselves to that perspective, i.e., that are centred on the question and its associated perspective. In other words, it's way easier if, let say Rescher refers to what Lakatos has written, i.e, we can see that we are reading something like a discussion between the philosophers and there is one context. For example, when I read Thornton, it's very clear that he read the same Popper that I read. His agenda is to explain Popper. He does not have any other agenda. It's less obvious in the case of Lakatos and that explains why I feel it's a different subject, but still he focused a lot on Popper. I feel like Rescher brings a new subject and similarly with Bunge, etc. because they have their own agenda, perhaps even more than Lakatos did with respect to Popper. Dominic Mayers (talk) 23:08, 27 December 2021 (UTC)

(edit conflict) Either I don't agree with your "special case" argument or I don't understand it. You said your idea would be to use Falsificationism for falsificationism as used by Lakatos. If you're saying that Lakatos is the general case, and everything else fits within that, I don't agree with that. It completely ignores the historical context in which these thinkers formulated their ideas. A single-article treatment of falsificationism would probably have to be an account of different thinkers presenting different interpretations of "falsificationism" in different contexts to address different problems/issues. For example, as you said further above, for Miller Lakatos's problem was considered a false problem. Rescher was partly concerned with problems such as cognitive economy, which was (I guess) not so important for Popper, Lakatos, or Miller. I don't see how all this could be formalized into a logical schema. The intellectual contexts are different, so a historical approach would be appropriate to some degree.

I wrote the previous paragraph in response to this edit of yours before I saw your subsequent additions. I will have to think about how your description of your working style in those sentences relates to what I said in the previous paragraph. Biogeographist (talk) 23:47, 27 December 2021 (UTC)

When I wrote that my idea would be to use Falsificationism for falsificationism as used by Lakatos, I meant that I can see Lakatos just like anyone else that saw the subject falsificationism. He might not own it, but his view contributes to define the subject. I thought it was obvious that, nevertheless, Lakatos's falsificationism cannot be the general case, because I accepted that the general case covers Popper's philosophy. Dominic Mayers (talk) 23:56, 27 December 2021 (UTC)

I'm still not sure that we understand each other, but maybe we do. Let me put it this way: I don't oppose creating a separate article on Falsificationism starting with material that is currently in this article's section on Lakatos's falsificationism. However, since the title of the article is "Falsificationism" not "Lakatos's falsificationism", it should be understood that the article may evolve to include perspectives and problems other than those that concerned Lakatos (especially if I get involved in editing it). How does that sound? Biogeographist (talk) 00:03, 28 December 2021 (UTC)

Sounds good. I might disagree with your style, but I will be happy to try to adjust myself. Regarding your view that different perspectives or agenda are necessary, I cannot say that it's incorrect. For me, the secondary sources that we use should have placed their view within a perspective associated with the subject just as when we have a discussion. If I had to explain a new thesis or a complete new agenda before you can make sense of what I write here, you will find it weird. In the same way, I expect that sources that contribute to a subject are centred on the subject and there should be no need to understand a new perspective or a new agenda. Lakatos had a different agenda, but he criticized Popper in so much details that we can connect the two agendas easily and it becomes useful as a way to explain what Popper's agenda is not. It's not necessarily the case with all "secondary" sources that come with a different agenda or a new thesis. They might be disconnected in an unclear way and the reference that the "secondary" source make to the subject might be like a part of a different subject and any attempt to make a connection would flirt with original research and any implicit connection would also flirt with original research. What I am talking about here reminds me a bit of the paradigm shifts of Kuhn. For kuhn, there is no discussion possible between people having different paradigms. In this analogy, sources that are analogous to different paradigms should be used in different subjects. Dominic Mayers (talk) 00:28, 28 December 2021 (UTC)

I especially agree with this: For me, the secondary sources that we use should have placed their view within a perspective associated with the subject just as when we have a discussion. When we were discussing how to edit Scientific evidence earlier this year, I was trying to find secondary sources like what you described in that sentence, but for the subject of "scientific evidence". (By the way, I am still interested in editing that article, and I collected sources, but never came back to it.) At the same time, in our discussion about Scientific evidence, you argued (if I remember correctly) that some secondary sources may not include all important views, which of course is correct. So one has to use a variety of secondary sources that may frame the discussion among primary sources in very different ways. And how is one to adjudicate among secondary sources? There may be a tertiary source that covers it all, but probably not. At the "tertiary" level some degree of editor discretion will probably be necessary. Regarding your Kuhnian paradigm analogy, Michael Friedman said in response to Kuhn that although some scientific paradigms/frameworks are incommensurable in certain ways, nevertheless paradigm/framework change can be rationally guided at a philosophical meta-level. The analogous meta-level here would be the discretion and discussion of editors guided by Wikipedia's guidelines and policies. Biogeographist (talk) 00:55, 28 December 2021 (UTC)

Also, I apologize for losing my patience above. Biogeographist (talk) 01:10, 28 December 2021 (UTC)

And how is one to adjudicate among secondary sources? The issue is often that the subject is large to start with and it can be addressed using so many different perspectives, agenda or theses. In that case, in principle, we have no other choice as to accept all perspectives that is proposed by editors and weight them using WP:proportion. In practice, it might not be what happens. Instead a few or even one editor come(s) in and pick a perspective and this perspective becomes part of the definition of the scope, i.e., of the subject. If editors try to include content using a different perspective they might meet resistance, because it appears as disturbing the flow and organization of the article. I don't think it is a bad thing. In fact, perhaps the solution was indeed to split the large subject so that every article focuses on a single perspective. It does not mean that the different articles cannot be connected. I don't remember the details, but I do remember that I was arguing in favour of a split and that it was not a POV-fork if the article were correctly connected. To answer your question, my answer is that a natural way is to pick one source not to fix the content, not as being the correct source, but as concrete material used as a reference to define the question, the subject, the perspective. In the case of Lakatos, his book is useful to define a question, a perspective, an agenda, but it did not prevent me to accept that it can be generalized so that it covers Popper's philosophy. In fact, the more I think about it, it's not surprising at all, because Lakatos got his agenda from Popper's agenda. He just did not appreciate how well Popper's solution was connected to the reality of science. I mean, for me, Popper found the correct way to locate rationality in science. Lakatos had the same agenda, but did not appreciate that and he kept looking for rationality where it could not be found. But the agenda, the perspective, etc. were not so different. In fact, some outsiders that look at this from a distance, classify them both as positivists. I don't see what these outsiders see, but I find interesting that some people can put Lakatos and Popper in the same category. So, you see that I have a different approach than you. You suggest a top down approach where we look at higher levels of abstraction. I know some of it is necessary, but I am saying that having a concrete material as a point of reference is also necessary. Yes, higher levels of abstraction are necessary, but they come naturally through discussions and in trying to understand the concrete material, simply trying to describe abstractly what is the question, etc. Regarding the non-commensurability of different paradigms, I also do not believe in this and Popper also did not believe in this. He wrote The myth of the framework to oppose this view. Nevertheless, it remains that some deep considerations of both paradigms are needed to make the connection. Therefore, I am not against different perspectives, but we need to have the willingness to go deep in each perspective to make this connection. It's not against any rule, because it will be our understanding of these sources and we will simply present them in the best way possible without doing any original research, because it will be the actual way the sources are connected. But that requires a deep understanding of the sources and a lot of reading to confirm with secondary sources, etc. a tremendous amount of work some times. Now, add to this that we can keep adding new perspectives and new sources with new agenda, new theses in this manner. It becomes frightening. The alternative approach is to present the different perspectives as being superficially connected, almost as if they were opposed some times, when in fact they are very close, if not almost the same thing. This is not an interesting article for me. This is what I would call an horizontal approach at the expense of a vertical approach with one perspective. It's way better to focus on a single perspective and go deep into it than to do that, only covering each perspective superficially. I admit that my attitude is motivated by a limited amount of energy. Dominic Mayers (talk) 03:51, 28 December 2021 (UTC)

Thanks for your thoughts. I agree with a lot of them. I have read Popper's "The Myth of the Framework", and there is a lot to like in that essay, but it's important to note that what Popper argues against in that essay is a concept of "framework" that is more non-technical than what Michael Friedman was talking about. Popper in that essay was against the relativism of the doctrine that: truth is only relative to a framework (which is a loose and non-technical concept in his essay), and one can only think within a framework, and people have to share a common framework to understand each other, so no rational discussion is possible outside of a framework. Popper was right that all of that is wrong. There is a vulgar interpretation of Kuhnian paradigms that is analogous to the doctrine that Popper is criticizing here. But Friedman tried to fill in what was missing in Kuhn's account of paradigms, which is an explanation of how rational change of paradigms is possible. And importantly, incommensurability in Friedman (and I guess in Kuhn too, but I have not read much Kuhn) is a technical concept that only refers to a property of the presuppositions of scientific theories and what one can do with them. These theoretical presuppositions (paradigms/frameworks) are incommensurable in certain ways, but as Friedman especially emphasized, this does not imply that there is no rational way to compare and change frameworks. Friedman's point is very compatible with Popper's point in "The Myth of the Framework" but in a more technical context.

I definitely feel this: But that requires a deep understanding of the sources and a lot of reading to confirm with secondary sources, etc. a tremendous amount of work some times. That is why I have not returned to work on Scientific evidence; it requires a lot of such work. Biogeographist (talk) 15:48, 28 December 2021 (UTC)

My conclusion is that splitting is good, because it allows each article to be simple and focused. Some times, it's not really a splitting because the other articles in that split are yet to be written and that works too. I would like to see Falsificationism to be a simple and well focused article. Connections with other views that are easy, useful and not controversial can be included where they are useful in the article, even perhaps in the lead. What must be avoided is to imply criticisms, oppositions or supports when that can only be seen after a lot work that make sure it's not superficial. The neutral section "See also" should be used in these cases. Criticisms can be included without doing a lot of work when they come from sources that are centred on the subject with no need to understand a new agenda or a new thesis. A key point is that this will not only make it easier for us, but also for the readers. Dominic Mayers (talk) 16:57, 28 December 2021 (UTC)

Since we have a consensus here that Falsificationism will be an article, I suggest continuing further discussion about the article at Talk:Falsificationism. The lead section in particular could be tricky to work out. Biogeographist (talk) 21:01, 28 December 2021 (UTC)

The article is peppered with footnotes, which gives an impression that it is thoroughly referenced. Well, it is not. I found this problem because I was curious about what is the "searchlight theory". It turned out that the footnotes are in fact further musings, often with no clear relation to the sentences footnoted.

There are two purposes of footnotes: (A) references (B) clarifications or minor pieces of info not critical to overall flow, e.g., etymologies of some terms, interesting but not essential biographical details, etc. This was not the case for the "searchlight" section. Therefore I removed all them and replaced with "citation needed" tags, and you should immediately see that citations are indeed needed.

I also removed pieces that show no relation to "searchlight" neither in text nor in refs cited, i.e., constituting pieces of WP:SYNTH/WP:COATRACK.

I have no interest in further reading here, but whoever works here must review all the remaining "pseudoreferences". Loew Galitz (talk) 21:01, 9 January 2022 (UTC)

@Loew Galitz: I was considering removing the entire section, because, of course, the searchlight theory is Popper's philosophy, which is already discussed in the article as needed to understand falsifiability and more than that is out of the scope of the article. It's not as if Popper proposed a different philosophy with his searchlight theory. The bucket theory is Popper's criticism of some philosophers, which is also discussed as needed in the article to understand falsifiability. The point is that the article needs to be consolidated with a focus on falsifiability. This is also why we plan to make the Falsificationism section much shorter and move the remainder in a main article on the subject. Dominic Mayers (talk) 21:16, 9 January 2022 (UTC)

In contrast, the observation of a single black swan is technologically reasonable and sufficient to logically falsify the claim. On the other hand, Duhem and others said that a definitive experimental falsification of such statements is impossible.

The highlighted part contradicts both the source and wikipedia itself: the very previous sentence is an example of definite falsification. I fixed it ( On the other hand, Duhem and others said that a definitive experimental falsification of an isolated hypothesis is often impossible.), but was reverted. But this is the very essence of Durham's point: a validity oif a hypothsis is often contingent on "auxiliary hypotheses", i.e., often it is meanngless to discuss a hypothesis outside its context. .Please voice youor opinion. Loew Galitz (talk) 00:05, 10 January 2022 (UTC)

If you now understand the distinction between the logical aspect and the methodological aspect, then you can better understand the lead and I can answer your question. The auxiliary hypotheses in Duhem's problem are not a part of the logical aspect of falsifiability. As Thornton says it's very simple on the logical side. If you can state "this single ferrous metal is unaffected by a magnetic field", then that's it you have shown that "all ferrous metals are affected by magnetic fields" is falsifiable. In practice, you need the auxiliary hypothesis that there was no hidden shield that prevented the magnetic field to affect the metal, but this is not an issue at the logical level. This is why I am telling you that by definition the logical side is separated from all the experimental hypotheses. So, the lead is not stating anything incorrect when it says that definitive (methodological) falsification of these (formal) statements is impossible. Definitive (methodological) falsification of these formal statements is always impossible, because there is always auxiliary hypotheses that are required and you cannot tell if is an auxiliary hypothesis or the statement that is falsified. There is no need to say that these statements are isolated from the auxiliary hypotheses, because it is implied by the fact that they are on the logical side. You might think that it's better to say it anyway, but I don't think so. It's confusing at the informal level of the lead to do that and it's not formally required. Dominic Mayers (talk) 02:59, 10 January 2022 (UTC)

tl;dr I concluded that your version is wrong. I may be mistaken. I gave two arguments. Please address them directly without lengthy philosophy. Best with references supplied with quotes from referenced text. Your personal reasoning is irrelevant. Loew Galitz (talk) 04:01, 10 January 2022 (UTC)

I collapsed the part that you need if you don't understand the basic. This contains the answer to your questions. It's philosophy, because falsifiability is a philosophical concept. It's not my personal opinion, but a basic explanation of falsifiability as needed to answer your questions. If you don't understand this answer, I am willing to help. Dominic Mayers (talk) 04:17, 10 January 2022 (UTC)

I don't need answers from your brain; I need answers from sources. I can verify that the sources cited match the wikipedia text (at least we can discuss that), but I have no desire nor obligation to verify your reasoning on the subject, i.e., whether your reasoning is based on solid references. Loew Galitz (talk) 04:35, 10 January 2022 (UTC)

(edit conflict) It's all from the sources. Go read a Wikipedia article about some mathematical result. You will not be able to match the text in the article directly with a text in a source. It's the same thing here. You refer to a text about Duhem's problem. You ask about the auxiliary hypotheses. Popper, Thornton and Chalmers explain that falsifiability is a logical criterion and this is the basic notion required to provide the answer. In fact, the problem with auxiliary hypotheses was also raised by Lakatos. Lakatos's concern and Popper answer is covered in Schilpp's book The Philosopy of Karl Popper. It's your responsibility to understand this if you want to help. Popper's answer is what I explain here. He says that it's a logical criterion. To make it clear, he gave as example the following:

"Newton's theory... would equally be contradicted if the apples from one of my, or Newton's, apple trees were to rise from the ground (without there being a whirling about), and begin to dance round the branches of the apple tree from which they had fallen."

Chalmers uses the following example of potential falsifier of Newton's law: "The brick felt upward when released". The point is that falsifiability is purely at the level of statements and their relations, i.e., at the level of the language. This was explained by Popper to answer the exact question that you raise: what about the auxiliary hypotheses? The answer is that they are not needed and not a concern at the logical level where falsifiability is defined. At the logical level, there is no auxiliary hypothesis, so it's pointless to say that the theory is isolated (i.e., separated from the auxiliary hypothesis). The whole point of Duhem's problem is that the theory in isolation, i.e., separated from the auxiliary hypotheses and other theories in the background knowledge, cannot be falsified, because it might be these other hypotheses and theories that are incorrect. It's not in my brain. It's exactly what the sources explain. Dominic Mayers (talk) 05:01, 10 January 2022 (UTC)

Let me give you an example of your dubious logic on your ferrous example. (A) "All ferrious are affected" (B) "This ferrous one is unaffected" . As you say if the statement B (regardless its truth) is proof of logical fal...lity of A. Now, if you start adding "aux hypotheses" to falsify (A) in practice, you are running right into the No True Scotsman falacy or spherical horse in vacuum issue. So lets go back to swans. And citing the sources. Loew Galitz (talk) 04:50, 10 January 2022 (UTC)

First, it's not my dubious example. It's Thornton's excellent example in Stanford Encyclopedia of Philosophy. Yes, Thornton explains that at the logical level it's simple: (B) contradicts (A) and that shows that (A) is falsifiable. This is by definition of falsifiability. There is nothing dubious here and it's not at all my logic. It's the definition given by Popper, Thornton, Chalmers, etc. Now, as you point out, it's not true that (B) falsify (A) in practice, because there are auxiliary hypotheses, a background knowledge about the apparatus, etc. needed and the whole thing gets complicated. That's true, but this is why the distinction between the logical and methodological side is important and falsifiability is a logical criterion, i.e., defined on the logical side, free from the problems of the methodological side. That's not my logic. It's all in the sources. Dominic Mayers (talk) 05:14, 10 January 2022 (UTC)

The issue about the phrase in question is not about the distinction you mention. It is about whether this piece of wikipedia text faithfully represents the source. Once again, for the third time: I say that the clause "in isolation" is crucial for Duhem's logic, and you seem to have confirmed this right here. Loew Galitz (talk) 05:25, 10 January 2022 (UTC)

Go read a Wikipedia article about some mathematical result most wp articles in math and computer science are shitloads of dung (beyond the lede), precisely for the reason you mentioned, especially the ones which contain proofs of some seminal statemments. Loew Galitz (talk) 05:22, 10 January 2022 (UTC)

No, my point also applies to excellent articles in mathematics and science. In these complex domains, you cannot apply policies without understanding the sources. It's even more true in philosophy. In our case, the content is actually very very close from the sources, but you must realize that the role of Wikipedia is to present in a single short article a vast amount of information available in sources and to extract what is most relevant. It requires an encyclopedic work to achieve that and you cannot apply policies to judge the process without a minimal understanding of the subject. It's not complicated when you understand the subject, but is more complicated than simply applying policies without any understanding of the subject. Dominic Mayers (talk) 05:39, 10 January 2022 (UTC)

I say that the clause "in isolation" is crucial for Duhem's logic, and you seem to have confirmed this right here. Yes, I said it's true that Duhem's problem does not apply to the entire system of statements that includes the theory on the logical side as well as the hypotheses, etc. in the methodological side. However, given that we know from the context that we are referring to falsifications of theories stated in the logical side, then we are in the case of isolated systems. There is no need to insist that it is an isolated system to be correct. On the contrary, it creates a confusion, because it suggests that the Duhem's problem does not apply to all the cases that we are considering. So, it's clearer not to mention uselessly that the systems that we consider must be isolated. The essential point is better conveyed in that manner. Dominic Mayers (talk) 05:57, 10 January 2022 (UTC)

Perhaps what you don't like is that the text says "Duhem and other say ..." and Duhem speaks of isolated systems and we don't in an explicit manner. Perhaps we could write "imply" instead of "say", but "imply" is misleading here, because it suggests that a complex deduction is needed and it's not the case. I think "say" is fine here, because the fact that the systems are isolated (from the methodological side) is clear from the context. My point is that it's not wrong and the essential to convey is the apparent contradiction between (1) the asymmetry that includes (logical) falsifications and (2) the Duhem's problem that says (methodological) falsications are impossible. We want to convey this apparent contradiction in the simplest manner possible without mentioning unnecessarily the fact, which is clear from the context, that the theories or statements considered are isolated from the methodological side. Dominic Mayers (talk) 06:18, 10 January 2022 (UTC)

I can see that you might be frustrated that the text is not verifiable in a mechanical manner that does not require some thinking. Your objective is fine. It will be great that all articles can be verifiable in a mechanical manner that does not require any thinking, but the priority is to have a text that conveys the information in the simplest manner possible. Twisting a correct text so that it can be verified mechanically is not optimal for the readers. The ordinary reader that is learning about falsifiability will not see the problem that you see. Besides, in practice, we always consider systems in isolation. The concept of a totality is purely philosophical. I does not exist in practice: there is always some methodological aspects that are ignored in the description of a system: we never reach the bottom of the swamp in Popper's language. Dominic Mayers (talk) 06:42, 10 January 2022 (UTC)