Talk:Falsifiability

I propose to undo the last edits by IP 140.78.5.108

The current article does not hide that many philosophers had a different view than Popper and that, even recent philosophers think that the criticisms of Lakatos, Kuhn, etc. were valid. If this is not clear in the paper, I think it should be made clearer. However, the words "claimed", "thought" and similar words in expressions such as "Popper claimed" and "Popper thought" are not neutral at all. They suggest in Wikipedia's voice that Popper was wrong. Wikipedia must present the view in a neutral manner without suggesting it is wrong. Added texts such as "since Newtonian physics does not deny that there could be forces acting on the apple that are stronger than Earth's gravity" even argue in Wikipedia's voice why Popper was supposedly wrong. I always say that there is a need to understand the sources, in this case Popper, before we can write a Wikipedia article, because, otherwise, there is a risk that we attribute the incorrect view to a source when we criticize it using other sources. This is a subtle point: a criticism often, especially in Popper's case, attribute incorrectly a view to the source that is criticized. These criticisms are kinds of strawman fallacies. In these cases, it should be made easy for the readers to see that it is a strawman argument. The misunderstanding is that falsifiability is a logical criterion, not dependent upon the actual laws, except for the material requirement. This is indeed a very subtle and difficult point to understand. Not surprisingly, the IP did not understand it and, because of that, has misrepresented Popper's view. Instead of reading the article to understand that Popper used the basic statement "an apple that moves from the ground up to a branch and then starts to dance from one branch to another" to make clear that falsifiability has nothing to do with whether Newtonian physics denies or does not deny that possibility, the IP presented his personal argument that Popper is wrong because Newtonian physics does not deny it. It's so discouraging. The article explains it so clearly and yet some readers miss the point. I do not think the fault is in the article. It is a difficulty intrinsic to the concept of falsifiability. The difficulty is that falsifiability is practical, about practical experiments, because of the material requirement, but at the same time it has nothing to do with whether or not the law actually contradicts real observations. In the example of the apple, the material requirement is respected, because it is actually possible to measure the position of an apple at different times. The fact that Newtonian physics does not deny that there could be forces that make the apple dance in this way is irrelevant. Referring to this is clearly a strawman argument. It would be relevant if falsifiability was about the possibility of methodologically proving the law false, but this is impossible and it is not required for falsifiability. Clearly, Popper knew it. For some reason, Lakatos could not abandon the idea that falsifiability is about methodologically proving the law false and he argued that it was impossible using arguments similar to the one presented by the IP. Of course, we cannot methodologically prove the law false, for example, because Newtonian physics does not deny that their could be forces that explain the movement of the apple. In a way, falsifiability is a very simple criterion. One simply needs to look at it at the purely logical level (except for the separate material requirement): one needs only to consider the language in which the laws is formulated and whether one can, in that language, build a statement that contradicts the law and yet respect the material requirement. Nothing more than that. No consideration of whether or not we can actually conclude methodologically that the law is false. Dominic Mayers (talk) 17:20, 5 March 2024 (UTC)

@Dominic Mayers: I think you have justified your proposed reversion well enough that you can go ahead and do it. If the IP editor wants to come and argue against you, they can. The IP edits as written are vulnerable to your rebuttal. However, there may be a more interesting argument in Stove (see web version here: chapter 5, section 4) that that the IP editor missed, which is (if I understand it correctly) that Popper asserted that certain statements are falsifiers of Newtonian physics, but Popper did not make a valid deductive argument showing that the statements are falsifiers, and the statements could be considered merely logically independent of Newtonian physics. In this regard, Stove especially takes umbrage at Popper's claim that "almost any statement about a physical body which we may make [...] would contradict Newtonian theory" (emphasis is Stove's). Stove calls this "mere bluff". Biogeographist (talk) 20:29, 5 March 2024 (UTC)

Yes, I believe that Popper in some texts has followed others such as Lakatos in speaking of contradictory universal laws, but the context was the possibility of initial conditions. I don't think it is fair nor interesting to make a big deal about the fact that some times, perhaps often, they neglected to make explicit that context. It's the contrary: it is Stove's rejection of that context that is not natural. He is logically correct, because logically it is possible that the universe is empty and, therefore, the appropriate initial conditions are not possible. However, these philosophers were excellent logicians and it is not serious to attack them at that level. On the contrary, Popper was very careful at the logical level in the analysis of his criterion. If Stove would have argued like that with Popper and Lakatos they would have raised their eyebrows and ignored him, because, of course, they were aware of this logical fact and it is not serious at all to assume that their philosophies fall apart because they would have failed to take into account that kind of logical facts. I decided to read Stove's article How Popper's Philosophy Began (1982), published the same year as "Popper and After: Four Modern Irrationalists". The paper makes some valid points in its first pages, but it starts to really misrepresent Popper with the sentence Consistency with every observation-statement, if it is a defect at all in a theory, must be a very trifling one. For it is a property which Marxism, for example, shares with such superlatively good theories as Newtonian physics, ... He misses the point that unfalsifiability => no predictive power, which is certainly a defect. His arguments against falsifiability, when applied to predictive power is that no theory (including Newtonian physics) has predictive power and, therefore, lack of predictive power cannot be a defect. But, of course, Newtonian physics has predictive power and if it did not, it would not be useful. Stove is really misunderstanding the distinction between the necessary material requirement and the too strong methodological requirement of knowing facts or laws to be true. Popper has always been aware that there is no way to be sure that a contradictory observation statement is true, because errors are always possible in measurements. He was also aware that, even if they are no errors in the observations, they can be made compatible with a law by introducing some explanations for the apparent contradictions. Popper just generalized and discussed more formally an argument that is also needed to justify the merit of predictive power, but which we don't bother to give in the case of predictive power. I think it is important to present an argument for the merit of predictive power, even in the presence of methodological problems (errors in measurements, ad hoc explanations, etc.) and this is what Popper did for the more general falsifiability criterion. Stove missed that point. Dominic Mayers (talk) 00:10, 6 March 2024 (UTC)

The lede

I've done some reparagraphing of the lede just for clarity. But it brings out that the sentence about "some philosophers such as Deborah Mayo" is out of place in the lede. Obv., Popper's views have critics, but this does not seem to be the place to cite Mayo in particular. This material could be eliminated from here, or there could just be a reference to the fact that there are critics of Popper's claims. The detail can come later. The point is that the lede paragraphs are meant to provide a summary/overview about the significance and substance of a topic. Adding details that don't match that description will come across as jarring to the reader and may mislead about what is most important. Metamagician3000 (talk) 08:47, 15 March 2024 (UTC)

Just to say that Popper's view has critics would be pointless and even misleading. The readers must be able to make their own judgment when they read the criticisms and be able to see them for what they truly are. This need for a context does not mean that they do not belong in the lede. Even the confusion that exists in the literature about falsifiability is an essential part of the subject. I know that it is a very difficult task, because the readers could easily get confused, if they expect to be simply told what falsifiability is. When criticisms are presented, they might think that it is because falsifiability really deserves them. But the other options, saying only that criticisms exist or completely ignoring them, seem unacceptable to me. The article must explicitly and clearly consider the confusion that exists in the literature and this aspect is important enough that it must be present in the lede.

There are two aspects in Popper's philosophy regarding falsifiability. The first is that it presents the very common view that predictive power and the existence of experimental tests are key features of science. Popper formalized and generalized this view with the falsifiability criterion. The second is that there are still today many philosophers that criticize the criterion with arguments that apply also to predictive power and the existence of experimental tests. For example, they might not realize that the Duhem-Quine problem says also that no law has true predictive power, because we can never be certain about these predictions. Because he understood very well the Duhem-Quine problem and similar problems, Popper provided a criterion to resist them. This is exactly why he insisted that falsifiability is a logical criterion. If it was only the non educated public that misunderstood this aspect of Popper's philosophy, I would say that you are right, but those who still apply arguments against Popper's philosophy that also apply to predictive power are professional philosophers. Mayo and Stove are examples. I am not saying they reject predictive power as a fundamental criterion, but I think they miss what Popper attempted to do (and succeeded in doing so in my view). He wanted a clear criterion that covers the natural notion of predictive power and resists the Duhem-Quine problem and other problems that he refers to as methodological problems without pretending that these problems have a solution at their own level. He accepted that there are no solutions to Hume's problem, to the Duhem-Quine problem and to other methodological problems. He understood that the Duhem-Quine problem takes a step back on his separation logical versus methodological and correctly says that the methodological problems are still there. He never denied that. He even gave his famous swamp analogy: "the bold structure of its theories [the logical side] rises, as it were, above a swamp [the methodological side]. It is like a building erected on piles". This misunderstanding of Popper's philosophy is a significant part of the literature on falsifiability. It seems to me that, on the contrary, it must also be a significant part of the article and this must be reflected in the lede.

I do appreciate your concern, however, because presenting this situation is a very difficult challenge, but I do not think it is a challenge that can be avoided. Dominic Mayers (talk) 17:15, 15 March 2024 (UTC)

Headings

@Valjean: Your edit summary said that a blank line after a heading is "default" in Wikipedia, but that's not true: per MOS:BLANKLINE, a blank line after a heading is optional. Please refrain from making misleading statements about Wikipedia guidelines in your edit summaries. Thanks, Biogeographist (talk) 20:28, 19 March 2024 (UTC)

I didn't say anything about guidelines. When one adds a new section, the software automatically adds that blank line, and it helps me and others with poorer eyesight while editing. (It saves time and sometimes has prevented me from making serious editorial mistakes.) It does no harm to you or anyone else, so why make it harder for other editors? Why "do harm" by deleting the blank lines? You are literally going out of your way to make editing harder for some editors. That's not a sin of omission but of commission. That's not very collegial. If I hadn't explained it to you, it wouldn't be a big deal, but I did, yet you chose to actively act in an uncollegial manner. SMH. -- Valjean (talk) (PING me) 03:59, 20 March 2024 (UTC)

Patrollers should not give lessons regarding rules of collegiality to other editors unless they have very good justifications to do so, which is not at all the case here. Such an attitude does not contribute to create a nice working environment. I dislike a bit the extra blank lines, but I am much more concerned with the negative effect of this kind of attitude than I am with the extra blank lines. The MOS:BLANKLINE rule is clear: "with one blank line just before it; a blank line just after is optional" (emphasis is mine). Dominic Mayers (talk) 04:39, 20 March 2024 (UTC)

@Valjean: Stop saying it is "Wikipedia's default". That is not true. The "New section" function on talk pages is coded to add a blank line after a heading, but I can choose to use or not use that button; its behavior is not "Wikipedia's default". (I don't even see that button on article pages, only on talk pages.) In fact, the blank line does cause harm because it makes less content fit within the editing window, which means one has to scroll more while editing, which makes it harder for me to edit. So there are usability issues whether one includes a blank line or omits it, so stop acting as if your position is righteous and as if I am the evil sinner (when you are the one who unilaterally decided to impose your preference on the article—I was just restoring the article's longstanding and justifiable convention). Don't go around misleading less experienced editors by falsely claiming that your personal choice is "Wikipedia's default". Biogeographist (talk) 12:45, 20 March 2024 (UTC)