Talk:Falsifiability/Archive 7

Where is the rational part in "Critical rationalism"?
The new section explains that not only falsifiability, but also the methodology that applies it, both, are free from the problems of falsification. In doing so, this new section also explains that Popper's methodology does not deny the holism of Pierre Duhem. Yet, Popper continues to emphasize the possibility and usefulness of attempting falsifications and possibly obtaining corroborations, which, he says, must be done in a piecemeal way. In Popper's view, this is where rationality has to be found. But, there seems to be some contradiction, because falsification is to be attempted in a piecemeal way and Duhem holism says that the system as a whole, including the theory that is needed to explain the observation and possibly auxiliary hypotheses, only can be falsified.

It is not reasonable to hope that all readers will happily accept the way to remove this contradiction that is proposed in the current version of the article and was given by Popper: there is no contradiction because the holism theory concerns only justifications (that is, the search for certainty in the form of some kind of inference rules) whereas Popper's view on rationality is not based on reference rules. But, then on what is it based? Popper's answer is very practical. These falsifications lead to useful problems, they steer progress in science. Yet, the article must provide more explanations and consider the connection with statistics, etc.

The difficulty in having people accept Popper's way to remove the contradiction is the same as the one that troubled the logical positivists in the years 1930. But the modern form of this difficulty is even more entrenched, because it appears to be supported by the success of statistic in social science and by the view of knowledge that is often suggested in the field of artificial intelligence. I mean that many people look at statistical rules as rules of induction and, similarly, it is often suggested that artificial intelligence can explain the growth of knowledge. So, in this modern positivist view, Popper had a very good point regarding the importance of falsifications, but he was wrong, so it is claimed, that no induction is used in science. So, what we have here is that positivism is still well alive today. We can speak of modern positivism in a very realistic and concrete manner.

Of course, Popper acknowledged the success of statistic and also of artificial intelligence (to the degree it deserved it at the time). Popper saw in a positive manner the exact same reality as the one that is seen by the modern positivists. However, the meaning of "induction" used by the modern positivists is not the same as the one used in the debates of the Vienna Circle. The argument, if we can call it an argument, of the modern positivists is different and refer to a different notion of induction.

What is the key difference? The key difference in modern positivism is that there is no attempt to propose a model for scientific knowledge that can be used to explain, using reasonable premises, how the totality of our objective scientific knowledge can grow. What Hume found is that any explanation based on inference rules that act on the current knowledge together with observations fail under reasonable premises. Goodman tried to reject one of Hume's premises, namely that inductive rules need standard justification, by proposing a different way to justify induction rules. However, he did not succeed. The point here is that we still do not have a solution to the original problem of induction that is based on justification (i.e. on inference rules). No one except Popper has offered an alternative model of rational growth of scientific knowledge. Popper built upon the holism of Duhem to create his model. He did not oppose it. He especially accepted one key consequence of this holism: there is no rule to pick a theory over another.

Regarding Hume's argument, modern positivists either ignore it or they provide some quick way to supposedly escape it such as a quick reference to Goodman's point that induction does not need justification. They do not provide a global model to explain the rational growth of knowledge. Instead, the focus is on local problems of statistic, artificial intelligence, etc. where the methods work to some degree and these are called inductive methods and they claim that this solves the problem of induction. But, again, no global model is proposed. There is no real explanation.

It is not the purpose of this article to go into this debate more than it has been in the literature, but, of course, there is no need to. There is plenty of sources that discuss that. The point here is that some subsection should be about that. This is necessary before we can start the much needed section about the link between falsifiability and statistics and probability. This new section about the link with statistic is important, because it will show that Popper insistence that there is no inference rules, but yet there is rationality in science, show some light in the current crisis in the field of Statistic (as described by Deborah G. Mayo in her recent book "Statistical Inference as Severe Testing: How to Get Beyond the Statistics Wars" (2018).

Dominic Mayers (talk) 21:31, 19 June 2020 (UTC)


 * wrote: No one except Popper has offered an alternative model of rational growth of scientific knowledge. But Charles Sanders Peirce sketched a pragmatic model of the rational growth of knowledge before Popper. ("Whewell, Peirce, and Popper are all trying to solve the problem of the growth of knowledge; they all treat it as a key problem in the philosophy of science; and in spite of important differences, their answers to this problem share a striking family resemblance." ) Peirce influenced others, such as Nicholas Rescher, who formulated a pragmatic dialectical model in books such as Dialectics: A Controversy-Oriented Approach to the Theory of Knowledge (1977) (which is cited in the "Further reading" section of this article for its treatment of confirmationism versus falsificationism), Methodological Pragmatism: A Systems-Theoretic Approach to the Theory of Knowledge (1977), and Cognitive Systematization: A Systems-Theoretic Approach to a Coherentist Theory of Knowledge (1979). (This is not an exhaustive list of Rescher's relevant books; the quantity of books he has written is amazing.) Incidentally, Donald T. Campbell, who was profoundly influenced by Popper in the 1960s, later reported being influenced by Rescher too. (See, for example: )
 * And: The point here is that we still do not have a solution to the original problem of induction that is based on justification (i.e. on inference rules). Well, there is the meta-inductive justification. This is mentioned in the "Meta-Induction" section in "The Problem of Induction" in the Stanford Encyclopedia of Philosophy, and in Gerhard Schurz's more recent book Hume's Problem Solved: The Optimality of Meta-Induction (2019). But this doesn't seem to be justification "by inference rules" as you call it; rather it seems to be a kind of pragmatic methodological justification, perhaps somewhat like Peirce and Rescher (I am guessing, since I haven't read Schurz—the aforementioned SEOP article groups his meta-inductive approach with pragmatic approaches). Biogeographist (talk) 01:03, 21 June 2020 (UTC) and 01:58, 21 June 2020 (UTC)


 * Yes, these two statements are subjective. Of course, I am sure we can find many inference-based proposed solutions to the problem of induction. I just don't see them as valid solutions, because I believe in Hume's argument. The meta-inductive approach seems a dead end for me, but I haven't looked at it carefully. My first thought is what is the point of a meta rule to find rules, when there is an argument that no rule exists?  The fact that the obtained rules, let's call them the child rules,  might have limited applications do not help avoiding Hume's argument, because at the end the meta rule acts exactly as a rule, if the child rules are to be applicable. It is just a big rule that takes into account a lot of history, but Hume's argument still applies.


 * Similarly, I am sure that many alternative models have been offered. In fact, I see Duhem holism as one such a model. In this case, Popper built on it. But there could be others that are totally different. I suspect Peirce's Model is compatible with Popper's model, because I was told that Popper refers to Peirce in a positive manner.  Anyway, yes, it is subjective when I say that Popper is the only one to have offered an alternative (i.e. not rule-based) model to solve the problem of induction. I just guess that previous models were not sufficiently complete to deserve to be called a solution to the problem of induction. In the case of Peirce, I might be wrong, because I did not look at it. Given that Popper was positive about it,  I would be very surprised that these models oppose each other instead of supporting each others. So, yes, maybe I should acknowledge, as Popper did, that his model is not totally new. His main contribution was falsifiability as a criterion of demarcation and the role of falsifications in a method that is not severed from nature, i.e., that is connected with nature through what he calls expectations and predispositions. The latter being is solution to the problem of induction. The article is incomplete in this regard. The link with probability and statistic would make much more clear how complete was Popper's solution.


 * The link with statistic, brings out more clearly that though there is no rule to pick a theory, there is a way to globally evaluate progress. This is a point on which Popper insisted a lot. It is easier to see concretely what he meant when we consider how Popper's view was used by many in statistical studies. It's well known in statistic that we can not draw definitive conclusions, in particular, because we can not be sure if the data is correct. Popper, following Duhem, says that there is no rule to reject a theory. There is no difference in the case of statistic. Yet, Popper's point (and it might also have been Duhem's point) is that we can still see the  problem when there is a falsification. It means that we can see the progress when new theories come into play and these falsification problems disappear. So, a key point is that the absence of rules to pick a theory over another does not mean that we can not see progress. Popper have never formalized his criteria to evaluate progress, but he described the main ideas very clearly.


 * In the application of falsifiability in the literature, statistic comes into play in two different ways. The first way is when we have all-some meta physical principles such as in the neutrino example. In these cases, often there are parameters that correspond to the "some" part, i.e., the existential part. To transform the meta physical principle into a falsifiable law, the parameters must be given or a statistical method can be proposed to estimate these parameters. This is the first way where statistic enter into play.  The second way is the most obvious way and the most clearly pointed out: Popper explained that an actual falsification is not done by a single falsifying basic statement. The definition of falsifiability requires only a single falsifying basic statement, but actual falsification is different. Popper always emphasized the difference. For an actual falsification, we need a falsifying hypothesis with repeatable instances. Obviously, many have made the link with statistic and there are plenty of examples in the literature.


 * Of course, in the article, I will only use verifiable statements. I consider that in a talk page, it's fine to share our understandings, because it's more practical in that way. We always read the literature in terms of our understanding. Editors can better understand each other if they share their understanding.


 * Dominic Mayers (talk) 02:47, 21 June 2020 (UTC)


 * Just to respond to your first paragraph above: You seem to be thinking of meta-induction as a recursion of induction, but that can't be right, since the circularity would be a prima facie failure to solve the problem of induction. To jump back to Rescher for comparison, Rescher differentiated between "thesis Darwinism" and "methodological Darwinism": the latter is not just a recursion of the former; they are completely different. Without having read Schurz, I imagine the relation of some inductive method to meta-induction is like the relation of thesis Darwinism to methodological Darwinism, since the SEOP section on meta-induction says that Schurz said that an optimal meta-inductive method "gives rise to an a posteriori justification of induction", that is, a justification based in part on pragmatic/empirical results but not based on a priori rules. Rescher and Schurz, whatever their differences, seem to share an emphasis on competition among methods. This is not surprising since Rescher published a book in 1980 titled Induction: An Essay on the Justification of Inductive Reasoning that is cited by Schurz. Rescher, however, explicitly connects the competition among methods with Darwinism in his account of the rational growth of knowledge, whereas Schurz (at least as described in the SEOP article) is more narrowly focused on the problem of induction and not on an explanation for the long-term growth of knowledge; in other words, Rescher's philosophical work has a broader scope than Schurz's if my guess is not mistaken. Biogeographist (talk) 12:59, 21 June 2020 (UTC)

I found a description by Schurz of the relation between meta-induction and Popper, in which Schurz says that Popper's account of corroboration contains a special case of meta-induction, which makes Popper a kind of meta-inductivist:

"Several aspects of Popper's account of theory testing can be criticized. First, only for strict but not for statistical hypotheses is it possible to derive observable consequences by means of deductive logic. [Footnote: To overcome this problem, Popperians have suggested regarding extremely improbable observations as a "falsification" of the respective statistical theory (Gillies 2000, 148ff.; Popper [1935] 2002, chap. II.68). Howson and Urbach (1996, 174) argue convincingly that this view is untenable.] Second, Popper is not right about all methods of discovering hypotheses being irrelevant to questions of justifications. One counterexample is the inductive generalization inference explained in section 1.1: it gives us a method of discovering a statistical hypothesis that at the same time inductively justifies this hypothesis. On the other side, Popper's criticism of methodological induction is correct for the reasons pointed out in section 1.1: inductive methods for extracting general hypotheses from observations exist only for empirical but not for theoretical hypotheses because the latter contain nonobservable (theoretical) concepts.

All these aspects are minor compared to the following fundamental challenge. Even if we grant Popper that discovery procedures are separated from measures of justification and that the derivation of observable consequences proceeds in a deductive way, his claim that the scientific test procedure can proceed entirely without inductive inferences is untenable. The idea of Popper's notion of "corroboration" is, of course, that we should base our future predictions and actions on those theories that up to now have been most successful—that is, theories that so far have been corroborated best (Popper 1983, 65; 1979, section I.9). It follows from this that Popper's "deductivistic" program of corroboration, too, contains in its core a fundamental inductive step, which Musgrave (2002) called epistemic induction. The principle of epistemic induction says the following: if theory T$1$ has been more successful (explanatory and prognostically) than theory T$2$ so far, then it is reasonable to assume, relative to the given state of evidence, that T$1$ will also be more successful than T$2$ in the future. In other words, the success preferences established so far are projected inductively into the future.

Epistemic induction is a special case of what we call "meta-induction" because it does not inductively infer object-level hypotheses about ordinary events but meta-hypotheses about the confirmational success of object-level hypotheses. Meta-induction (in the sense of this book) applies this principle at the level of arbitrary methods of prediction or action."

Biogeographist (talk) 13:59, 21 June 2020 (UTC)

Meta-induction, model for growth of scientific knowledge, Schurz, etc. (for ease of editing)
Before, I go into detail about meta-induction, etc., I feel we must discuss the distinction between credit issues (who first proposed a concept: a method, a  model, etc.) and issues that are directly about a concept, such as whether competing views exist. The latter is in my opinion much more important. The former, i.e., who should get the credit, is important, but it will be sad to render the article controversial because of that kind of issues. Therefore, the style that we use is important. If two authors, say Popper and Charles Sanders Peirce, have overlapping contributions, the style used should make both contributions more important, not one less important in favor of the other. Yet, the article should reflect the actual impact on the literature that each author had in the context of the content that is in question. For example, in the context of the problem of induction, the Stanford Encyclopedia of Philosophy entry on the problem of induction mentions Popper, but not Peirce. Is that mean that, in a similar context, we should not mention Peirce? Not at all. On the contrary, it will be an additional support for Popper's contribution to mention Peirce. The same is true for Pierre Duhem's contribution. So, if we avoid a controversial style, credit is not an issue. This is important, because it is very easy to find controversies in the literature that are not based on the concepts themselves. This article is an example. In these cases, I strongly propose that we use the Wikipedia policy that states that though every thing in Wikipedia must be verifiable, not every thing that is verifiable needs to be included in Wikipedia.

We have a more critical situation when a competing view, say a competing model for growth of knowledge is proposed. By "competing", I mean that the competing model, if it was correct, would render Popper's model obsolete, if not wrong. However, I don't see that such a competing model exists. For example, I just read a review of Schurz's book and immediately, from what I understood, it appeared to me consistent with Popper's model. If that is the case, as I explain above, I am not going to present it as a controversy. I need to check the details, but most likely they use statistic to evaluate the success of methods and that is exactly what is suggested in Popper's model. I am not saying that they are not going further than Popper, but the basic idea of Schurz's solution appears the same. I also have the impression that in one way, it is less complete than Popper's solution. The point is that they argue in terms of picking the optimal method, but that does not provide an explanation for the growth of knowledge. In their own terms "we cannot tell whether induction is reliable or not, but if any prediction method is reliable, then induction is." I would not call this an explanation for the growth of knowledge. Moreover, I think I understand now what you mean by "meta-induction" and it's not what Popper would call induction. My first interpretation of meta-induction was very natural: a rule to find inductive rules, but what Schurz do is not that, if I understood correctly. Instead, it looks very much like Popper proposed. It is explained in that we must assume an axiom that the method works, which means that it leads to progress. Popper also said very clearly that we can evaluate this progress. I have the impression that it is exactly what Schurz does. To say more about this similarity, Popper (and later David Miller) argued a lot that this cannot be turned into rules of induction. In particular, they have written many arguments to show that Baye's theorem cannot be turned into a proof that data is evidence for choosing a statistical hypothesis. Interestingly, a section in Schurz's book provides an argument that says exactly that. I looked at a sketch of the argument and it appears very similar to an argument already used by Popper and Miller. Anyway, the key point here is that it does not seem at all a competing model.

My hope is that I am right about the fact that Schurz's work fits within Popper's model. In that case, I am not going to be controversial and, of course, what is new content from the book will be clear. I have no interest whatsover in giving more credit to Popper than he deserves. The idea of a controversy, if there is a match, is the wrong way to look at credits. On the contrary, it will make the contribution of Popper even greater, if we succeed to give a lot of credits to others in a consistent manner within his perspective.

I would like to share my perspective on how unlikely it is that any new and valid model would make Popper's solution obsolete or beside the point. The great strength of Popper's solution is that it is based on a model that is far from being simple, but yet its details are very natural and match very well with what is done in practice. Popper's model has four parts: the objective knowledge in the form of falsifiable theories, objective knowledge in the form of (temporarily hidden) theories that impregnate observations, the definition of basic statements in terms of these hidden theories and, finally, nature, not in the form of the laws behind the observations, but in the form of predispositions and expectations that are behind our choice of conjectures. This last part is the non objective knowledge. This four-parts model is not trivial at all. Of course, the first three parts are based on the meta-logical notion of truth given by mathematical structures, which was introduced by Kurt Godel and Alfred Tarski in the years 1930 and Popper acknowledges that he was influenced by Tarski,  but this fact does not make it easier to understand for scientists that are not interested in logic at this level of depth. Note, by the way, that falsifiability is very important to make this model useful. Similarly, the fact that the conjectures cannot have any rule to justify them is very important, because otherwise it would clash with Hume's argument.

The reason why I believe that it is very unlikely that one will provide a competing model that will make this model obsolete is that this model, despite its apparent complexity, is very natural. The logical part where we check the theory at the logical level, etc. is very natural. The observational part where we do actual falsifications is also very natural. Even the part where we guess conjectures using expectations or, in Einstein's terms, using our intuition without "logical path" is natural. In fact, the more I look at it, the less I understand why there was even a debate about that. Well, it's the attachment to a justification that was the problem, but we have seen, for example in the case of Lakatos, that people at the end, just cannot reasonably state such a rule. Schurz is another example. He speaks of taking the optimal methods and you said yourself that there are no inference rules. At first, Popper did not expect this resistance. He wrote that originally he did not want to publish his model, in particular about falsifiability, because he felt that it was too obvious. When we look at what it would take to formalize it, we can see that, actually, it is not so obvious. I guess it is both obvious and not obvious. It is not obvious when we try to formalize it, but it is obvious when we simply look at how it matches with what we do in science.

Well, I might be wrong and a competing model that will make Popper's model obsolete might be proposed one day, but it seems that this day has not come yet.

Dominic Mayers (talk) 03:37, 22 June 2020 (UTC)


 * Popper's model has an important place in the history of philosophy and methodology of science, but it was not the beginning or the end of that history. See, for example: On the continued empirical testing of theories of scientific change, see, for example, various references in:  Biogeographist (talk) 23:19, 22 June 2020 (UTC)


 * Definitively, it was not the beginning of the history of models for scientific growth: Duhem's and Peirce's work are excellent examples. It will certainly not be the end of that history too. However, let us not be lost in the vast varieties of models that might have been proposed.  The subject of this article is falsifiability and indirectly methodologies to apply it. So, the only models that I want to consider are those that extend or nicely complement Popper's model and models that are opposed to Popper's model.  The bucket-like  models  are opposed to Popper's model and the article discuss them. By the way, I found that Schurz's work is typical of what is still going on today to support the bucket view of science.  There is no proof, of course, that the bucket view is incorrect. In particular, Hume's argument is based on premises. So, one can always try to deny the validity of one of these premises, just like Schurz did.  I am biased. I see that Hume's argument and Popper's model, both, use very reasonable premises.  However, many philosophers, Lakatos, for example, did not like Popper's premise about the organismic nature of science. My impression is that this covers the whole situation. The only thing that is yet missing in the article is the extension of Popper's model in terms of statistic and how this relates to Popper's view on how we can evaluate progress in science. I doubt very much that there exists in the literature a different approach to extend Popper's model or a model that is not a bucket-like model that is opposed to  Popper's model, but I will have a look at "Theories of Scientific Method: An Introduction" by Nola and Sankey, because you suggest it.


 * Dominic Mayers (talk) 02:17, 23 June 2020 (UTC)

Notes on the first chapter of Schurz's book
Here is an excerpt that describes the inductive rules that will be justified: "The two simplest forms of these inferences are the following.

1. Inductive prediction: r% of all so far observed Fs have been Gs. Therefore, with a (subjective) probability of approximately r%, the next F will be a G—and thus will be predicted to be a G, provided r is greater than 1/2 and F is the total evidence regarding the next observed individual

2. Inductive generalization: t% of all so far observed Fs have been Gs. Therefore, with high (subjective) probability, approximately r% of all Fs are Gs."

- Schurz

Of course, if it was not for the fact that it is previously explained in the preface that by "justifying", it is meant "showing that there are no better rules" (in some sense that has yet to be clarified), I would have immediately concluded that this is doomed to fail. At pages 8-9, we are told that it's a meta-induction rule that is provably optimal, but I guess it's fair to infer from this, as Schurz seems to do, that the rules that are induced using the meta-inductive rule will also be optimal. For now, my thought is the following. Given that Schurz accepts Hume's argument that these induced rules cannot explain the growth of knowledge, I am not sure what's is the point to show that it's the best we can do. It seems to me like a very negative result. I paraphrase: not only they do not work, but they are the best we can hope. That's sad.

My hope is that Schurz's argument avoid this issue by being clever about how the "success rates of all accessible prediction methods" is evaluated. If the evaluation is done in a way that can not be turned into inductive rules, just as Popper suggests, then there is some hope. But then, even without knowing how the success rate is evaluated, it must be that the rules (which are mentioned above and are the ones induced by the meta-induction rule) are not to be used as inductive rules, because that would still clash with Hume's argument. I have some doubts, because this point is so important that it would have been mentioned. Without further qualifications, these rules are to be used to explain growth of knowledge in the usual way. This is in contradiction with Hume's argument. (Well, not really, because the claim is only that they are optimal among the rules that can not work, but that's not a big recomfort.)

Unfortunately, when we read carefully chapter 1, it becomes clear that there is a claim that Hume's argument was escaped. It is the need for a circular justification of ordinary inductive rules that is being escaped, so it is claimed. So, it is a very clever modern positivist argument. It fits in the bucket view of science. We can certainly mention the book in the article. It really illustrates that the bucket view of science is definitively well alive. I can paraphrase the argument of the book as follows: If we accept that the bucket view of science was the best view thus far, then the meta-induction rule tells us that it is optimal to continue to use it. Really, Schurz claims as a fact that the inductive rules were the best rules thus far. On this basis, using the meta-inductive rule, he argues that it is rational to use them. That is his way to escape Hume's argument. Here is the excerpt where he explains that "[T]he justification of meta-­induction generates an a posteriori justification of object-­induction as follows. We know by experience that in our world inductive prediction methods have been more successful in the past than noninductive methods, whence it is meta-­inductively justified to ­favor object-­inductive strategies in the future."

- Schurz

Dominic Mayers (talk) 17:33, 22 June 2020 (UTC)


 * This is not the place to discuss in detail Schurz's book, nor is it the place to park your notes on the book. I only quoted the book to give you a sense of how meta-induction relates to Popper's work. I don't think everything that you've said about Schurz is right, but since I don't think this talk page is the place to discuss his book, I won't respond here. Indeed, I think you should delete this section about his book or move it to your user talk page (you have my permission to delete or move my response too). I will just repeat what I said above: I see Schurz's domain of inquiry as relatively narrow, not the kind of broad theory of knowledge that Rescher elaborated across many books. Biogeographist (talk) 23:19, 22 June 2020 (UTC)
 * The main reason why I read this book and discussed it is because you mentioned it. I did not remember the sentence where you say that you haven't read it, because may be I would have picked a book of Rescher instead. The title "Hume's problem solved" caught my attention. So, I read it and discussed it, thinking that it was important to consider seriously the sources that you mention. So, I am disappointed that you did not find my points about the book useful. In general, i.e., not referring to this book in particular, I think it is useful to discuss sources in the talk page. I do not regret to have read the first chapter and to have written three paragraphs about it, even though I see that it does not correspond to what you expected from it. If you want to delete my three paragraphs, which were in part written for you, go ahead. I have no intention to move anything in my personal talk page. Everything I wrote was in the context of this article and, in this particular case, it was in response to what you wrote. Dominic Mayers (talk) 00:59, 23 June 2020 (UTC)


 * The main reason why I read this book and discussed it: No, you didn't read the book; you just read a few pages, fixated on a few phrases, and jumped to the incorrect conclusion that it is a very clever modern positivist argument that fits in the bucket view of science. Schurz's project is not a continuation of positivism—he wrote an earlier book, Philosophy of Science: A Unified Approach (2013) that showed that he was familiar with the limitations of positivism and of many other positions in the philosophy of science—and your idea that his project must neatly fit into one of only two views of science is nothing more than a facile confirmation of your own preconceptions—Schurz's earlier book, not to mention many other books, present many more than two views of science.
 * As I said above, I only quoted the book to give you a sense of how meta-induction relates to Popper's work, not to propose using the book as a source in this article, so a discussion of this book is not a discussion of a source, unless you are proposing to use it as a source. Biogeographist (talk) 02:31, 23 June 2020 (UTC)
 * In the same paragraph, I explain that I read the first chapter. So, there was no intention to pretend that I read the entire book. You are wrong and most unfair when you say that I just read a few pages, fixated on a few phrases, and jumped to [a] ... conclusion. Actually, I read the entire first chapter in detail, and part of it more than once. I carefully read it. Why are you claiming things that you can not know? Where you there when I read the first chapter. (Now I have to be careful: if I don't put the first every single time, I will be accused.)  Regarding the possibility that my conclusion was incorrect, this is another story. I don't think that it was incorrect, but even if it was, it should not be  a big deal. It can happen. Your feeling seems to have been hurt by the term "modern positivists".  Did you assume that I meant that Schurz defended the exact same thesis as some positivists of the years 1930 and thus was an ignorant? Are you a friend of Schurz? Anyway, let me be very clear that I was impressed by the quality of this first chapter. I saw in there one of the best explanation of Hume's argument. Definitively, my impression was that Schurz was competent.  I introduced the terminology "modern positivists", because there is the justification aspect in common.  Many attempts to explain how scientific knowledge can be justified by some form of induction were made by the positivists. By "modern positivists", I refer to those who continue to want this kind of justifications of inductive rules.  Clearly, from the first chapter, we can see that Schurz believes that the growth of scientific knowledge can be explained (at the least partially) by ordinary inductive rules (their probabilistic generalization) and that he proposes a way to justify these rules. That's all what I meant. Dominic Mayers (talk) 04:03, 23 June 2020 (UTC)


 * I have the book, and just by skimming through it I can see that you fixated on a few phrases in the first chapter and did not accurately summarize Schurz's project. It is not positivist. Not all philosophers who have sought to justify the use of induction are positivists: for example, Mario Bunge is a major non-positivist philosopher who considered some uses of induction to be justified, while being opposed to what he called inductivism, the incorrect idea that all scientific hypotheses are obtained by induction from empirical data. One can use induction for some tasks without claiming that all science is inductive. Indeed, Schurz pointed out that Popper himself used a kind of meta-induction, and if Popper was using a kind of meta-induction without any justification, that is not rational, not worthy of the name "critical rationalism". So Popperians should at least welcome a justification of meta-induction, even if they want to avoid using any object-level induction. But again, using object-level induction for some tasks does not make one a positivist. Even Bunge did not think object-level induction could be entirely avoided, even though the over-all pattern of empirical research is hypothetico-deductive. Biogeographist (talk) 13:00, 23 June 2020 (UTC)

You wrote It is not positivist. I never said that it was. You have not read me. I explained that I don't use the expression "modern positivist" as a synonym of "positivist". I basically explained that by "modern positivists" I mean philosophers who have sought to justify the use of induction. So, basically, you are telling me that Not all [modern positivists] are positivists. as if I was saying otherwise, but I am not.

Bio, I am a bit tired of these discussions. The main criteria that I set for myself to decide if I continue to work on Wikipedia articles (in a given domain) is that I share a minimal understanding with other editors. Unfortunately, it seems that in accordance with your (conscious or unconscious) criteria, you feel that I am biased and don't present what you would consider a balanced view of Popper's view of science. My experience is that these criteria that you have, whatever they are, do not lead to useful discussions. We are going no where. It seems as if you will always find issues and be concerned unless the paper presents an opinion that says Popper was wrong. But, I am definitively not interested in presenting anything that is wrong. I don't see this article as being about Popper's view of science, which parts was wrong and which parts was correct, but about falsifiability and a methodology that applies it. Of course, what is interesting is what appears to be correct given the support it has in the literature and I focus on this. I cite Popper because he was the main proponent, but I am also using other philosophers to support the methodology. For example, even Lakatos position, at the end, supported the proposed methodology, in the following sense that he did not use any inductive rules and, as mentioned by Zahar, he used falsifiability. I would find totally not interesting to present an opinion like "Mr X disagreed with Popper and offered an alternative model", without actually presenting Mr. X's model and the arguments that Mr X provided and the counter arguments that Mr Y, Mr Z, etc. explained, etc. I mean the article should not be superficial and just express the fact that there was many competing opinions without presenting these opinions in a way that allows the readers to make their own judgment. The opinions in themselves without actual content that the readers can evaluate for themselves is just confusing, not interesting.

Also, you make a big deal about the fact that Schurz only argued that induction can be some times useful. But, you miss the point that Popper's view is that induction is never useful. So there is a dichotomy here: it's either some times useful or never useful. So it is either that the bucket view is never valid or that it is valid some times. In Popper's view, in the searchlight view, the bucket view never applies. In fact, Popper is well known for that: someone wrote "he wants none of this" referring to any degree of induction or justification.

The main confusion here is also that many statistical procedures are called inductive procedures even when they are not used as ampliative rules to justify new knowledge. (By the way, I like the fact that Schurz adopted the notion that, in its largest interpretation, an induction rule is any ampliative rule.)  For example, a statistical procedure to evaluate parameters can be a part of a conjectured meta-physical principle. In Popper's methodology, we are certainly free to use statistic in the formulation of a meta-physical principle. Perhaps many would say that, because of that, Popper also accepts induction. But as long as the statistical procedure is not used to justify any law, but only to define a conjectured law, there is no induction. Let's not quarrel about terminology. It's the idea of a justification by inductive rules that Popper rejects, not the use of statistic to evaluate parameters in a conjectured law.

Similarly, consider Mayo's example where she uses many scales to measure her weight. She says that because she uses many scales she can infer that she took more than four pounds. It may seems like induction in the following sense that this means that for other scales it will remain true that she took four pounds. However, behind the scene, there are error models for the scales, etc. It's not induction. If we consider all the background knowledge, we can deductively infer what will happen statistically with other scales. Popper is not denying the great value of all the research that is often done under the label "induction." The article would be confusing and not interesting if it brings out controversies based on terminology. Instead, we want to bring out how all of this fit harmoniously together, how all the research in statistic, etc. fit well under Popper's model.

But I feel that this is an issue for you. It's like, for you, doing this is being biased, not presenting competing views, etc. I am not biased. For example, I sincerely looked at Schurz's book (through the first chapter) to see what the content was. I saw that he is among the philosophers who have sought to justify the use of induction (which I call the modern positivists).

Anyway, it's the last time I argue with you. As I said, my criteria to decide if I continue to work on Wikipedia articles (in a given domain) is that I share a minimal understanding with other editors. Clearly, you feel very much concerned and you are the only other editors here, so I cannot work on this article and related articles anymore. It's too unstable as a situation.

Dominic Mayers (talk) 16:25, 23 June 2020 (UTC)


 * Above you said: I am biased. Now you have said: I am not biased. How am I supposed to respond to such self-contradiction?
 * If you make statements on this talk page that I find to be untenable, why shouldn't I respond? You said: you miss the point that Popper's view is that induction is never useful. No, I know that is what Popper said, and Schurz also said that is what Popper said. Schurz's view, with which I agrree, is "that although Popper's objections against certain understandings of induction brought to light important insights, his more radical claim that empirical science could go without induction is not tenable" (p. 11). Schurz is not alone in that view; even raging enemies of inductivism such as Bunge admit that there is some role for induction in science. This implies that, as I said above, the idea that there are only two views of science, the bucket and the searchlight, is too simple. Scientific research is more complex than that. See, for example, the discussions of induction in Bunge's two-volume Scientific Research: Strategy and Philosophy (1967, revised and reprinted in 1998 as Philosophy of Science) or the multiple volumes of his Treatise on Basic Philosophy on Epistemology & Methodology: in volume 6, one of his many regulative principles of scientific realism is: "Make liberal use of analogy and induction—but always be on the lookout for their limitations" (p. 269).
 * In my view, this article does not put falsifiability in a large enough context of history and philosophy of science. Perhaps someone will rectify this situation in the future. Biogeographist (talk) 17:09, 23 June 2020 (UTC)
 * Bio wrote : I am biased. Now you have said: I am not biased. How am I supposed to respond to such self-contradiction? No need to respond. Simply try to understand the two sentences in their respective context. It's so easy to find contradictions like you just did. I noticed a few times this kind of "contradictions" in the work of great writers. I don't remember where exactly, but I noticed them. Dominic Mayers (talk) 17:17, 23 June 2020 (UTC)
 * It's essentially the same context. Biogeographist (talk) 17:25, 23 June 2020 (UTC)
 * It's not the same context. Your decision that it's the same context is subjective. I would not even take the time to check the exact contexts, but most likely in one case I referred to my effort to consider opposing views, but not superficially opposed, as I explain above, whereas in the other case, I must have referred to the fact that at some point we need to accept or reject premises that are used as starting points in an argument such as Hume's argument (and staying neutral is not practical, just like Escher would say) and we all do that in accordance with our biases. Dominic Mayers (talk) 17:39, 23 June 2020 (UTC)


 * Bio wrote : [T]here is some role for induction in science. ... This implies that, as I said above, the idea that there are only two views of science, the bucket and the searchlight, is too simple. The bucket view is that induction is possible (and thus play a role in the progress of science). In other words,  when we say that some times we can use induction,  we are saying that the bucket view is valid. The fact that something else can play a role in the progress of science is accepted in this view. In the searchlight view, induction is not possible. I would have to agree with you that I have no argument to say that it is the only view in which induction is not possible. The issue regarding the separation in two views, the bucket and the searchlight view,  is not the important issue. The important issue is whether induction is possible (this is the bucket view) or is not possible (this is the searchlight view or something else). This classification might be simple and tautological, but I hope you agree that it is important.


 * Let's consider the important issue. I already clearly explained that we cannot prove that induction is not possible, because it depends on premises that one can always reject.  However, one has to propose an argument for the alternative view and this also requires premises. In the case of Schurz, the main premise that he used is that induction was possible (and successful) in the past.  I am sure that you accept this premise, but this is a personal choice. In a critical discussion, I would be allowed to explain to you why I don't accept this premise. Remember, we are in the context of the growth of scientific knowledge.  So, let us see how induction was in play in the discovery of Newton's theory. Humm,  not really.  Let us now consider how it was in play in the discovery of Einstein's theory of general relativity.  Humm, not really.  We see a pattern here. In order to even be able to conceive that we are in front of an instance of a universal law, we need to first have the universal law. Now, here is the main confusion. One might say that, even though the concept of instances is defined a posteriori, the relation between the laws and the instances are the same as in an inductive rule.  The argument would be that it's the inductive relation that matters and the fact it can be conceived only a posteriori is not important. That's fine, but we must keep in mind that there is no claim that the relationship between instances and the law is not useful. Of course, it is useful. This is what determines if there were falsifications or corroborations. Popper himself says that we must consider falsifications and corroborations when we evaluate progress in science.  Therefore, we must not confuse the usefulness of this "inductive" relationship with the possibility of induction. I used quotation marks to mean that actually the "inductive" relationship may have nothing to do with induction.  The existence of a justification for the law is a necessary ingredient in the concept of induction.  But then, we meet all the problems of falsification, the issue with the confirmation holism of Duhem and also, of course, Hume's argument. These are very strong evidence against induction. Now, let us see what the great philosophers and scientists wrote about that. Einstein clearly said that there is no logical path in science. It did not say, some times, we discover new laws through intuition, but other times, we use induction.  What about Kuhn, Feyerabend and Lakatos. All three did not believe in induction. Feyarabend was very clear about that.  Now, one might counter argue that these philosophers only referred to the process of guessing the law, not the process of justifying the law once it has been guessed. But this is not true, because they were quite clear that it also applies to any way to reject a law after it was guessed. Duhem says it explicitly, because he is aware of the confirmation holism issue. Therefore, I have no idea what is the motivation for Schurz's premise. On the contrary, we have plenty of evidence against it.  Moreover, the obvious usefulness of this "inductive" relationship can easily explain our subjective feeling that induction was successful. But, again, this is a redefinition of what is induction.  In practice, the usefulness of these "inductive" relations corresponds to the Mayo's example that I previously gave and to the use of a kind of induction relationship when we estimate parameters using statistic, and there might be other applications, but none of them are induction.


 * Anyway, you would be right here to say that by arguing again with you, just after I wrote it's the last time I argue with you, I contradicted myself. This is definitively a contradiction between what I said that I will do and what I actually did. But, the spirit remains. I am not sure why I argue. My experience with people is that at this level, there is little chance that you start to have a different attitude. It just never happened in my life. The problem is NOT that you think that induction is possible. I have no problem with this. The bucket view is explained in the article. Both views are there. In fact, I would also enjoy mentioning Schurz's argument in favor of the possibility of induction. Absolutely no problem there. I am happy to let the readers decide for themselves. The problem is that you seem to have a problem. You suggest that the article is biased. It's as if you want a more superficial account that looks like Popper did that, but Duhem had a different view and Rescher offered a new angle, etc. I mean that you are not looking for unity (in which the diversity is still clearly seen, in fact, even better seen), but more focusing on diversity. I suspect that it is important for you that in the style, Popper is only one view among many. This is what you feel would correspond to a balanced article.  For me, this would be a superficial article with no unity.  Given the pressure that I feel, I repeat, that I am not going to pursue this.


 * Dominic Mayers (talk) 18:43, 23 June 2020 (UTC)


 * Thanks for your response. Perhaps I should say that even though I have criticized some of your statements, I do find what you say interesting and I don't disagree with everything you say. I apologize if my emphasis on criticism has been grating. Here is an example of a statement of yours with which I agree: I appreciate your admirable ideal of a textual unity in which the diversity is still clearly seen, in fact, even better seen. One reason why I don't make major contributions to the article myself is that I am not confident that I could achieve that ideal. I don't think you have achieved it either. It would be very difficult to achieve. The editors of  The Cambridge Companion to Popper (Shearmur & Stokes) said that they did not commission a paper on Popper's treatment of induction because "the views of scholars with a specialist interest in this topic are varied but, by now, well entrenched" (p. 4). I imagine that means that the controversy over aspects of Popper's treatment of induction is so intense that not even the editors of that volume trusted any of the best scholars to be able to produce a unified summary in which the diversity is still clearly seen, in fact, even better seen! That is how difficult this problem is. So we should not feel bad that we are not able to do it either. I think you should keep working on the article as long as you think you can keep improving it. I would not want you to stop editing Wikipedia because you think I am harassing you (you didn't say that, but it may be how you feel); that is not my intention. I myself have no plans to edit the article beyond minor copyediting, so you don't need to worry that I am going to impose some kind of epistemic anarchy on the article. Biogeographist (talk) 22:34, 23 June 2020 (UTC)


 * I am glad that you wrote a nice reply, but my decision is already taken. Besides, the kindness of your reply came with a view that was not so encouraging. Maybe after someone else would have made important additions I might get involved again. Dominic Mayers (talk) 00:14, 24 June 2020 (UTC)

A word about Popper
While reading on the subject of falsifiability, I came to know about Popper, the man in his historical context. In Popper's time, for many, religion was not a reference for truth anymore, only a reference for moral, etc. There was a strongly felt need to separate science from religion. For this reason, clear scientific rules were needed by the positivists. Interestingly, even Alder's psychoanalysis and what Popper calls vulgar Marxism could claim to be scientific given the rules that were tentatively proposed at the time. In this context, in contradistinction, in Popper I see a man that was against rigid rules, etc. Rules fixed in advance, lead to lack of understanding, he felt. He saw in science a way to liberate men from fanaticism, a fanaticism that has been responsible for the killing of some friends when he was young. Thus, in Popper's heart, any attachment to rules, to supposedly verified truth, etc. was not wanted as part of science. I think that, for Popper, inductivism, even the supposedly moderate view in which induction has only "some roles" in science (whatever that means), is a remnant of a religious attitude in science. After all, the view that we own a verifiable truth, even probable truth, has definitively a religious connotation.

I would add to this that there might be a strong influence coming from the big investment of our society in artificial intelligence and also in modern medicine and its use of statistic. In both cases, the existence of these rules are important. In the former case, it's the hope of artificial intelligence to discover such rules. In the latter case, these rules are needed to claim that modern medicine provides evidence for truth - they even call it Evidence-based medicine, which for even more legitimacy was generalized as evidence based practices. In both cases, Popper's position is hard to accept.

It's easy to understand why. Popper says that the clashes that exist in an the overall theoretical system (i.e. the three objective parts) that we use to understand our environment never tell us which part of the system is wrong, that is, there is no rule to say whether we reject or accept a theory (in any part of his model). The clashes, i.e., the refutations, only tell us that we made a mistake and that we must try to do better. That's the only rule: conjectures and refutations. Of course, we can see that we do better when the old clashes disappear, but there is always new clashes, perhaps even more. Popper is against any religious attitude in the form of rules of truth, especially when it is disguised as science. The science of Popper is a science of modesty. I see in Popper a modest man that is against the possibility that we can verify truth, but yet explain how we can rationally progress by trying to remove the clashes in science.

But this is a big big issue. It's not surprising that it started in the years 1930 and it is still well alive today. I know that many refuse to see that it's the same issue as in the years 1930, but it is. As it was the case in the years 1930, there are those who want some form of verification (through some form of inductive rules) and those who following Duhem, Popper (and I believe also Peirce), say that there is no such rules. Duhem (and perhaps Peirce) was (or were) the first to recognize the confirmation holism issue, but it is much more precise with Popper. For some reason, Popper had the chance to more directly address the clash between verificationism and non-verificationism, using critical rationalism. Dominic Mayers (talk) 22:15, 24 June 2020 (UTC)


 * I see what you appreciate in Popper, and I appreciate it too, especially the emphasis on fallibility and rejection of infallibility in epistemology (and rejection of the authoritarianism that can accompany infallibilist epistemology). Infallibilist epistemology that expounds "infallible rules", and "infallible confirmation", should be rejected. But fallible rules and fallible confirmation, together with fallible falsification, are practical and necessary. (The complementarity of fallible confirmation and falsification is a theme in Rescher's book Dialectics cited in the article.) Today mainstream epistemology is fallibilist. "Almost all contemporary epistemologists will say that they are fallibilists", as the IEP article on fallibilism says. Gerhard Schurz in Hume's Problem Solved says the same: "Classical foundationalistic epistemologies demand that the basic beliefs be epistemologically certain or necessary (see Dancy 1985, chap. 4.1). Most contemporary epistemologists [including Schurz] reject the infallibility requirement as too strong" (p. 28).


 * This widespread acceptance of fallibilism in epistemology is due to many thinkers who developed fallibilist epistemology over the 20th century from Peirce onward, if not before (though Peirce coined the term), not only Popper. So we appreciate Popper for his contributions but we do not have to accept Popper's whole philosophy as a "package deal" to have a fallibilist epistemology, and we should not feel the need to be Popperian and defend Popper's philosophy as a whole and at any cost in order to promote fallibilist epistemology. Indeed, I propose that we will have a stronger fallibilist epistemology the more diverse are our sources (and consequently the more diverse is our pool of arguments and evidence, when those arguments and evidence complement each other and increase the comprehensiveness of our epistemology), and the more we can show how falsifiability and falsification fit within this context of fallibilist epistemology that is much larger than Popper's philosophy alone.


 * By the way, there are many important resources within recent argumentation theory which Popper did not utilize, perhaps partly because that field only started developing late in his career. The resources of argumentation theory are indispensable for opposing infallibilist and authoritarian applications of science and technology, because infallibilism and authoritarianism are at least as likely to arise in informal argumentation about science and technology as in formal methods. (Rescher's work in the aforementioned book Dialectics is helpful in this regard, among many other works in argumentation theory.)


 * Inductivism, the idea that all valid universal propositions are inductions, or even worse that all knowledge is a result of induction, should be rejected, for reasons made clear especially by Popper. But there is still a small role for induction, which should not be wrongly associated with inductivism as a "package deal" (much less should it be associated with infallibilism—there is no "slippery slope" from induction to authoritarianism!). I have already provided some references from Mario Bunge about what it means that there are some roles for induction in science. If you don't have access to those sources or don't care to search through them, I'll provide some quotations from Bunge, who by the way rejected inductivism and religion just like Popper. Here is Bunge on the (limited) legitimate role of induction in science, from his book Chasing Reality:

"Induction is neither everything nor nothing. It occurs in low-level (empirical) generalizations as well as in the confrontation of theoretical predictions with relevant data. This confrontation involves some statistical processing. But there is no such thing as probabilistic inductive logic (or probabilistic epistemology), because propositions, not being random items, cannot be assigned probabilities except arbitrarily; and also because hypothesizing is not a rule-directed activity but an art.... (p. 7)

The vast majority of philosophers have been remarkably laconic on inverse problems, with the sole exception of the induction, or Data → Hypotheses, problem. Since this is an inverse problem, it is likely to have either multiple solutions or none. The reason for this is as follows. By definition, a hypothesis goes beyond the data relevant to it. It does this in at least one of two ways: either because the hypothesis involves a leap from some existents to all possibles; or because it includes concepts that, like those of causation, mass, intention, and national sovereignty, do not occur in the data because they are not experiential. In sum, since data do not exude hypotheses, these have to be invented. And, of course, once invented they have to confront both old and new data.... (p. 146)

To compute a value of a well-defined mathematical function is a direct problem. By contrast, figuring out a function given some of its values is an inverse problem, namely, that of curve fitting (dot joining). There are standard techniques, such as the venerable Gregory-Newton formula, for performing such interpolations. However, most of these techniques yield functions of the same dull family, namely, polynomials that can be made to pass as closely as desired to the given empirical points. The existence of such algorithms falsifies Popper's opinion that induction is a myth. Induction does occur in science, both in suggesting low-level generalizations and in evaluating the empirical support of hypotheses (Bunge 1960). Moreover, induction is sometimes "mechanizable," as in the case we just saw....

What is true is that induction does not yield high-level and amazing (counterintuitive) results. Indeed, in general the resulting data-fitting polynomials that compress and expand data differ from the true laws, which ordinarily involve far more complicated functions. For example, the elementary law of the simple (idealized) pendulum, deduced from Newton's second law of motion, is $T = (1/2π)(l/g)1/2$, an approximation that can in turn be approximated by any of infinitely many polynomials in $l/g$. In this case, which is rather typical, the problem of induction is an inverse problem with infinitely many solutions, none of which is the correct one. The quest for high-level laws from data is hopeless unless the corresponding direct (deductive) problem has been solved in very many similar cases. More on this below.

That induction is confined to low-level generalizations has been known for a long time, not so the reason for this limitation. The reason that induction cannot lead to the upper rungs of the deductive ladder is this: An inductive reasoning is semantically "horizontal," in that it leaps from particular statements to a generalization containing exactly the same specific (non-logical) concepts involved in the data. For instance, $n$ experimental dots on the $x-y$ plane can be joined by a polynomial of degree $n − 1$ in $x$, namely, $y = a0 + a1x + a2x2 + ... + anxn$. Thus, if y stands for a spatial coordinate, and x for time, the function may represent the motion of a point mass, such as the tip of a harmonic oscillator. Yet, no matter how large $n$ may be—that is, regardless of the bulk of the database—the said function is bound to be a poor and clumsy approximation to the true solution of the linear oscillator problem, namely, $y = a sin ω t$. This solution involves the oscillation frequency $ω$, a parameter missing in the data but occurring in the equation of motion, which is one logical and semantic rung above the data.

This, then, is why there can be no "vertical" inference from data to high-level laws: because the latter contain concepts absent from the former. Since experience cannot generate any high-level concepts or hypotheses, these must be invented. And invention is anything but a rule-directed process, one subject to algorithms that could be fed into a computer.... (p. 152–153)

Much, perhaps too much, has been written about the leap from data to hypothesis. And yet this kind of inference still puzzles philosophers and cognitive psychologists—as it should. There are two main myths about induction: that it is everything (e.g., sometimes Bacon and Russell; Carnap and Reichenbach always), and that it is nothing or nearly so (e.g., Leibniz, Hume, and Popper). The former myth belongs to the core of empiricism, whereas the latter [myth] is at the centre of rationalism.

Francis Bacon is usually regarded as the inductivist par excellence. This he certainly was in his New Atlantis (1624), the utopia where he depicted scientists as data gatherers and packers. But Bacon was no simplistic inductivist in his Novum Organon, published four years earlier. Here he dismissed induction proper, or generalization from particulars, as trivial. He advocated instead framing alternative hypotheses and checking them, if possible experimentally; and he introduced the concept of a crucial experiment, by analogy with a crossroad.... (p. 165)

The limitations of induction derive from the shallowness of its products, all of which concern only observable properties, with no reference to hidden mechanisms. By contrast, direct reasoning from law-statement to fact presupposes that nature is lawful rather than capricious.

Hume was obviously right in holding that the leap from "some" to "all" is logically invalid. But he was wrong in denying the objectivity of connections, in particular laws, and thus in denying the existence of natural (or nomic) necessity along with logical necessity (see Bunge 1959a)....

Many students agree that the way to account for observable facts is to frame alternative hypotheses, and then weed out the false ones, ideally by means of crucial (conclusive) experiments. This was well known to Francis Bacon (1620), William Whewell (1847), and John R. Platt (1964), among many others.

Regrettably all three authors, along with many others, called this procedure "induction," while strictly speaking induction is the formation of a single low-level generalization from the data in hand—as when a dog learns to associate leash with walk, or a baby learns to call all men Dad.... (p. 167)

Recall from section 1 that inductive generalization from a bunch of data works only for hypotheses involving exclusively observable features, such as "All mammals are hairy," but it fails miserably for high-level hypotheses, such as "All mammals descend from reptiles." How does one go about framing such high-level hypotheses? The short answer is that one invents them, though not out of the blue... (p. 169)

A cognate argument for both vulgar materialism and empiricism is that mathematicians often make use of analogy or of ordinary (incomplete) induction to find patterns, as Polya (1954) showed persuasively. True, but the result of any such plausible reasoning is a conjecture that has got to be proved (or disproved) by purely mathematical means—for example, by reductio ad absurdum or using the principle of complete induction, neither of which is suggested by ordinary experience. In short, analogy and induction by enumeration have at most a heuristic value: they prove nothing, and proving happens to be the main job and sole privilege of the mathematician. Moreover, incomplete induction and analogy may lead us to error unless we check their outcome.... (p. 208)"


 * Biogeographist (talk) 03:54, 25 June 2020 (UTC)

Again we are going nowhere


Here is a part of your quotation of Bunge To compute a value of a well-defined mathematical function is a direct problem. By contrast, figuring out a function given some of its values is an inverse problem, namely, that of curve fitting (dot joining). There are standard techniques, such as the venerable Gregory-Newton formula, for performing such interpolations. The existence of such algorithms falsifies Popper's opinion that induction is a myth. Induction does occur in science, both in suggesting low-level generalizations and in evaluating the empirical support of hypotheses (Bunge 1960). Moreover, induction is sometimes "mechanizable," as in the case we just saw..... In response to this, I am quoting myself. Bio, I am a bit tired of these discussions. The main criteria that I set for myself to decide if I continue to work on Wikipedia articles (in a given domain) is that I share a minimal understanding with other editors. Unfortunately, it seems that in accordance with your (conscious or unconscious) criteria, you feel that I am biased and don't present what you would consider a balanced view of Popper's view of science. My experience is that these criteria that you have, whatever they are, do not lead to useful discussions. We are going no where. It seems as if you will always find issues and be concerned unless the paper presents an opinion that says Popper was wrong. But, I am definitively not interested in presenting anything that is wrong. I don't see this article as being about Popper's view of science, which parts was wrong and which parts was correct, but about falsifiability and a methodology that applies it. Of course, what is interesting is what appears to be correct given the support it has in the literature and I focus on this. Your quotation of Bunge perfectly illustrates the issue. Here is another except from what I wrote that explains my concern (the emphasis at the end is added): I feel we must discuss the distinction between credit issues (who first proposed a concept: a method, a model, etc.) and issues that are directly about a concept, such as whether competing views exist. The latter is in my opinion much more important. The former, i.e., who should get the credit, is important, but it will be sad to render the article controversial because of that kind of issues. Therefore, the style that we use is important. If two authors, say Popper and Charles Sanders Peirce, have overlapping contributions, the style used should make both contributions more important, not one less important in favor of the other. Yet, the article should reflect the actual impact on the literature that each author had in the context of the content that is in question. For example, in the context of the problem of induction, the Stanford Encyclopedia of Philosophy entry on the problem of induction mentions Popper, but not Peirce. Is that mean that, in a similar context, we should not mention Peirce? Not at all. On the contrary, it will be an additional support for Popper's contribution to mention Peirce. The same is true for Pierre Duhem's contribution. So, if we avoid a controversial style, credit is not an issue. This is important, because it is very easy to find controversies in the literature that are not based on the concepts themselves. This article (Kotarbinska 1961) is an example. In these cases, I strongly propose that we use the Wikipedia policy that states that though every thing in Wikipedia must be verifiable, not every thing that is verifiable needs to be included in Wikipedia. Bunge's statement that Popper is wrong is very similar to Kotarbinska's statement. In Bunge's case, it is almost ridiculous to suggest that Popper could have thought that parameter estimation, curve fitting and things like that are not used in science. Yet, this is Bunge's point, which you so much wanted to quote.

I don't say more. It's on purpose that I referred to what I wrote previously, because it illustrates that we are going nowhere. It's always the same thing. You need to see opposite opinions. Whereas, I would be very happy to present the actual content that Bunge explained about statistic, etc. along with Popper's model, etc. in a way that makes even more clear all the concepts. I am all for the diversity at the level of the concepts and an emphasis on unity that brings out the value of the diverse concepts.

Dominic Mayers (talk) 16:16, 25 June 2020 (UTC)


 * I am pointing out literature that I think provides a broader context; of course, you are free to ignore it, and you are free to ignore my comments (although you haven't entirely ignored them so far). Mario Bunge, Deborah Mayo, Ilkka Niiniluoto, Nicholas Rescher, all of whom I mentioned above and a couple of whom I quoted, among others, are important philosophers, at least important enough on Wikipedia to have Wikipedia articles! Even if you think that some of their comments about Popper are wrong, those comments are still notable, in my view. Above you said (in April): I agree that finding out a minimal consensus among a large group of philosophers is useful. The information that I have presented is at least useful to that end. If we want to find a minimal consensus among these philosophers, we need to look at the dissensus as well, otherwise we just have unity without diversity (Popper's view without the dissenting views that would reveal the minimal consensus), not the unity in which the diversity is still clearly seen, in fact, even better seen. My previous comment pointed toward what I conjecture is something like the minimum consensus (fallibilist epistemology), but the conjecture would need to be specified in detail by laying out the views of the various philosophers and how falsifiability and falsification are positioned within that variety of views on fallibilist epistemology, to reveal the minimum consensus. In The Cambridge Companion to Popper, editors Shearmur & Stokes said that "there has been persistent concern expressed as to whether Popper's views are adequate to explain what we do seem to know and to do" (p. 10). Above I presented quotations that confirm that concern from other philosophers. You may not be interested in that dissensus, but I have good reasons for mentioning it here, including the importance of the dissensus for identifying the minimum consensus. Biogeographist (talk) 17:58, 25 June 2020 (UTC)


 * By the way, this edit was supposed to be a joke! One could make the opposite joke as well: that scientists, when they use induction, are a lot like turkeys—and Bunge would say that the induction-generated statement "we are fed every day" is not a law as the cartoon says, but instead a mere low-level empirical generalization! Biogeographist (talk) 18:52, 25 June 2020 (UTC) and 20:09, 25 June 2020 (UTC)


 * I am pointing out literature that I think provides a broader context; of course, you are free to ignore it, and you are free to ignore my comments (although you haven't entirely ignored them so far). Mario Bunge, Deborah Mayo, Ilkka Niiniluoto, Nicholas Rescher, all of whom I mentioned above and a couple of whom I quoted, among others, are important philosophers, at least important enough on Wikipedia to have Wikipedia articles! Sure, but I never complained about that.


 * Even if you think that some of their comments about Popper are wrong, those comments are still notable, in my view. Some wrong statements can be notable, but Bunge's suggestion that (I paraphrase) "Popper thinks that curve fitting is not used in science" is not. We would need a section entitled "Non sense" in which we collect this kind of statements and that would not be interesting.


 * Above you said (in April): "I agree that finding out a minimal consensus among a large group of philosophers is useful." Yes, but that depends on the context. We need to be interested about the philosophers themselves and the history of the debates, etc. Otherwise, we must focus on the concepts that are valid (as claimed by most secondary sources in the literature) and we don't care that professor X at some time thought that professor Y was wrong. Doing that would be way too complicated.


 * The information that I have presented is at least useful ... Of course, again, I insisted very much that giving a lot of credits to others, to  Duhem, Peirce and possibly Bunge (I have not read him) and others is great. That does not have to remove any credit to Popper. On the contrary, it can make Popper's contribution even clearer. The converse is also true: by bringing unity with Popper's view these other contributions can shine even more.


 * If we want to find a minimal consensus among these philosophers, we need to look at the dissensus as well, otherwise we just have unity without diversity (Popper's view without the dissenting views that would reveal the minimal consensus), not the unity in which the diversity is still clearly seen, in fact, even better seen. That's where we have a problem. If Popper's versus Peirce's view (for example) were like Einstein's versus Newton's view, i.e., if there was a fundamental contradiction between them that cannot be eliminated without fundamentally modifying the theories, then, of course, we would need to bring out that contradiction. But, if the contradiction can easily be avoided by simply taking the best interpretation of Popper's view and the best interpretation of Peirce's view, then to avoid complications and to make the article simple for the readers, we must focus on these best interpretations. The misinterpretations and criticisms that occurred in the past, because of misunderstanding, etc. are complicated. Even if we wanted to explain these misinterpretations, we might not get it right. This article should focus on the concepts and obviously I mean the valid concepts that can be verified. It's not an article about Popper, about what he said that was correct and what he said that was wrong. That would be ugly, even if it was an article about Popper.


 * My previous comment pointed toward what I conjecture is something like the minimum consensus (fallibilist epistemology), but the conjecture would need to be specified in detail by laying out the views of the various philosophers and how falsifiability and falsification are positioned within that variety of views on fallibilist epistemology, to reveal the minimum consensus. But here you assume that there are fundamental contradictions. Your premise seems to be that there are fundamentally incompatible views and the job of editors is to search for these contradictions. This is not at all the situation. Clearly Duhem's, Peirce's, Lakatos's, etc. view are very much compatible. It might be ok to say that our job is to find a consensus, but not a "minimal" consensus in the following sense of picking the most incompatible interpretations possible and bringing out statements such as (I paraphrase) "Popper thinks that curve fitting is not used in science". We are looking for a consensus in the following sense that we are looking for unity. Really, I am a bit tired, because it seems important for you that "unity" is based on all the misinterpretations and must consider statements such as (I paraphrase) "Popper thinks that curve fitting is not used in science".


 * In The Cambridge Companion to Popper, editors Shearmur & Stokes said that "there has been persistent concern expressed as to whether Popper's views are adequate to explain what we do seem to know and to do" (p. 10). Above I presented quotations that confirm that concern from other philosophers. You may not be interested in that dissensus, but I have good reasons for mentioning it here, including the importance of the dissensus for identifying the minimum consensus. Again and again, you focus on statements where readers do not have the required content to make their own judgment and in fact might even be confused. It's like you are addicted to it. If I listen to you, the article must say that Bunge said Popper is wrong. It must mention that Shearmur & Stokes are concerned about Popper's view. It must say that Schurz says that induction is possible. Regarding Schurz, again, I will be very happy that we provide enough content so that readers can see the unity with Popper's view. For example, Schurz uses a premise and it should be mentioned. As long as there is a focus on the content that allows the readers to make their own judgment and see the unity, it's great. But mentioning on the name of presenting the diversity that  Shearmur & Stokes are concerned about Popper's view makes me throw up. Your persistent feeling that the article must present that kind of dissensus is what make me totally non-interested to pursue work on this article. It's too superficial. I am not rejecting any content, any argument that is verifiable, but if we only say Mr X. expressed such and such opinions, but Mr. Y says that it's not true and the readers have nothing to make a judgment, then the article says nothing.


 * I am not going to work on an article if this style of writing is going to be supported at some point. The situation is often that Mr. Y opinion is based on a choice of definitions or a different terminology, things like that. Because of that, if we try to clarify the situation directly in the article, it  would be a mess. Issues at the superficial level of definitions, terminology should be taken care in the talk page,  not in the article.  As editors, we have the possibility to decide which verifiable content we include and we can avoid this kind of useless mess. The way to make sure that this kind of mess does not happen is to see that there is a mimimal shared understanding of the literature among the editors.


 * I think I might sum up my feeling as follows. I feel that you and other editors act as if an article on subject A is forced to consider every part of the literature on subject A. Of course, the scope of subject A must be decided among the editors and this can be difficult some times, but Wikipedia policy is not that an article on  subject A should contain every thing that is said about subject A in the literature. When editors share a minimal understanding of the literature on subject A, they can more easily agree on which part of the literature is useful in the chosen scope of the article on subject A. With a solid understanding of the literature on subject A, they can decide to focus on concepts and avoid superficial statements such as  "Shearmur & Stokes are concerned about Popper's view".


 * Dominic Mayers (talk) 20:39, 25 June 2020 (UTC)


 * Thanks. Reading through this I don't really find anything to criticize regarding what should go in the article; in general I agree with what you've said. (I disagree that the statement from Shearmur & Stokes is "superficial", because actually that statement is their summary of the literature! Of course, they flesh it out in subsequent paragraphs.) The most persuasive idea, because it is entirely congruent with Wikipedia's policies and guidelines, is that articles should be based on the secondary and tertiary literature, not on primary literature. I am familiar with the secondary literature on the Bunge–Popper differences, which I can provide if anyone wants it.
 * "Popper thinks that curve fitting is not used in science" is not an accurate summary of their difference over induction. (There are some other differences between them as well, such as on the role of confirmations and the mind–body problem.) Their difference over induction, as I understand it, is a deeper difference between different models of scientific research and perhaps different metaphysics. Look again at this quote from Bunge: "There are two main myths about induction: that it is everything (e.g., sometimes Bacon and Russell; Carnap and Reichenbach always), and that it is nothing or nearly so (e.g., Leibniz, Hume, and Popper). The former myth belongs to the core of empiricism, whereas the latter [myth] is at the centre of rationalism." I could be wrong, but this distinction sounds a lot like Popper's distinction between the bucket view of science (empiricism, in which induction is everything or nearly so) versus the searchlight view of science (rationalism, more specifically critical rationalism, in which induction is nothing or nearly so). But in contrast to Popper, Bunge does not endorse this distinction; Bunge rejects both views and promotes an alternative to those views, a kind of ratioempiricism. This is more than just a redefinition of "induction"; it is a model of scientific research that is significantly different from Popper's, but it is also not the bucket view because in Bunge's model induction-generated empirical generalizations, while sometimes used in science, can never be turned into laws through inference rules, as explained in the long quotation above. Biogeographist (talk) 21:26, 25 June 2020 (UTC)


 * I have a problem with Bunge does not endorse this distinction [between bucket vs searchlight view]. There is nothing to endorse or reject. Bucket vs searchlight vs other non-inductive views is just an arbitrary classification. It's like definitions; the definition of bucket view, etc. I suggest that the question whether or not induction is possible is the important question that has concerned us recently. In our discussion, the bucket view is simply the view that induction is possible. So, the question whether the bucket view is valid is the same old question whether or not induction is possible. How could Bunge reject that question? It makes no sense.   Of course, the issue is perhaps that the question presuppose that we have an agreement about what induction is. Indeed,  I think a lot of confusion exists because we might not be clear about what is meant by induction. It's not only true between both of us, but also in the literature. One of my points is that we must at the least clarify between both of us what is meant by induction and be able to agree on the way to interpret the literature using that definition. For example, when we see in the literature that Gauch wrote that induction is possible in some ways, we must be able to recognize that Gauch refers to something that Popper has never called induction. We must be able to do this kind of clarifications in the talk page, because the article is not the place to do that and we certainly do not want to mix different definitions in the article.  Similarly, the details of the definitions used by Bunge must be considered in the talk page and  if we really need an extra definition to present an important conclusion of Bunge, then fine but we want to avoid too many definitions and focus on conclusions. I admit that I am confused at this stage about what exactly you think Bunge wrote that is worth to be mentioned in the article. The fact he does not agree with Popper's definitions of induction is not interesting as a point in the article. Choice of definitions to be used in the article and how they relate to the too vast variety of definitions in the literature is something that we need to discuss in the talk page.  For me and I think for Popper and many philosophers, induction is the use of rules that allows to add a conclusion in our body of knowledge given premises that are taken in our current body of knowledge. The concept of body of knowledge is very general. We allow any structure to be a body of knowledge. We accept that observation statements can be added to this body of knowledge and the rules can, of course, use these statements as premises. So, this notion of inductive rules is very general.  As a parenthesis, Popper associates any inductive rule with a search for certainty, because the existence of a rule is by itself a form of certainty, irrespective of the nature of the inferred knowledge. So, the question I ask myself is whether Bunge needs a totally different, more general or more specific definition of induction and how it relates to the above notion of induction which I think is what is normally called induction. If you say that Bunge addressed a totally different question, we would have to decide if this question is important in the context of the article.


 * articles should be based on the secondary and tertiary literature, not on primary literature. I am not sure how this relates to the discussion. I hope you do not mean that you identify "secondary literature" with statements of the form "we have a concern with Popper's view".  Is this what you want to suggest?  For me, an example of secondary literature is the article of Zahar that I used to understand Popper versus Lakatos. The article of Zahar has nothing to do with statements of the form "We have a problem with Popper's view". It went in details of the content.  It actually proposed a great unity between Lakatos and Popper. I have no idea why you raised this policy in our context.


 * I am familiar with the secondary literature on the Bunge–Popper differences, which I can provide if anyone wants it. This is not by itself a point of interest in the article. Even the Popper-Lakatos controversy is not a subject of interest in the article. Popper-Bunge is even less relevant. In fact, I renamed the controversy subsections to make more explicit what were the points of interest. So, I am not sure what point you suggest should be added.


 * This is more than just a redefinition of "induction"; it is a model of scientific research that is significantly different from Popper's, but it is also not the bucket view because in Bunge's model induction-generated empirical generalizations, while sometimes used in science, can never be turned into laws through inference rules, as explained in the long quotation above. On the contrary, it seems to be exactly the use of the term "induction" for something else. I have a problem with making a big deal about this by itself. I am not saying that this different concept is not important. This is not at all what I am saying. I am definitively saying that the fact that the term "induction" is used differently by Bunge than by Popper is not by itself an important difference with Popper.  This fact is, on the contrary,  not interesting at all.  It's the concept, irrespective of the label "induction", that matters. My bet is that the concept that Bunge presents is, on the contrary, entirely consistent with Popper's view. So, I have no idea why you want to emphasis differences, etc. when Bunge's work can most likely fit very well with Popper's view. I can see that you refer to secondary litterature on the subject, but if it is a confusion about definitions and terminology, this is exactly where or shared understanding of the literature becomes important. It would be perfectly legitimate to decide that we would not include in the article statements that are based on confusion at the level of definitions and terminology, even if they are verifiable. Clearly, this is all very dependent upon  our understanding of the literature. If primary school students were to write a Wikipedia article on General Relativity they would not pick the same material to present than if experts in physics wrote the same article. If experts can discuss and agree that this author simply made a confusing statement, it might be verifiable, etc. they will still not use it and they would not have rejected a viewpoint in doing so. They would simply have removed an absence of viewpoint, something that is only confusion. In fact, assuming that the original source was notable, such in the case of Mario Bunge, not only they would have removed the confusion, but the experts by their correct understanding, would present the key point made by the original source in a much better way, by  picking the correct secondary sources. So, nothing will be rejected, except confusion. At some points, with who you work to edit a Wikipedia article does matter. There is nothing to complain about here.


 * Dominic Mayers (talk) 23:31, 25 June 2020 (UTC)
 * Dominic Mayers, I got something to complain about: this talk page is longer than A la recherche du temps perdu, and while you made some 1500 article edits, the article itself is still in a rather crappy state. I'm wondering if Wikipedia is really for you. Drmies (talk) 00:16, 26 June 2020 (UTC)
 * Why are you complaining about me when every time there is a discussion, I wrote usually about in the same proportion than the other editor, maybe more than the other editor, but not by a large factor, twice as much maybe, may be less. Outside of any discussion, I also write material in the talk page before I move it in the article. Maybe it's a style that you don't like, but is it really against Wikipedia policy? What hurts me a lot though is when you say the article is in a crappy state. I am really hurt by that. I felt that I contributed to make a great and useful article with only verifiable content. On the contrary, on many respects, I see it as one of the best article in philosophy of science in Wikipedia.  But, I think I cannot contribute anymore, because there is a lack of common understanding and on top of it, now there is complete lack of appreciation of my work. Dominic Mayers (talk) 00:38, 26 June 2020 (UTC)
 * You have 1538 article edits here; the next editor is at 145. OK, maybe the article isn't crappy, but it shouldn't be at C-status. Moreover, a cursory glance reveals that a lot of the content is argumentative and not encyclopedic in tone, paragraphs aren't verified, citations don't have page numbers, there are way too many quotes in there (possibly falling foul of FAIR USE), and I can go on. In addition, you have a whopping 1348 edits on the talk page here. In other words, the edit to quality ratio isn't very high, and this talk page is unreadable. Drmies (talk) 00:51, 26 June 2020 (UTC)
 * You are right that long discussions in the talk page make it hard to read. It can also frighten away potential editors. I am concerned by that, but it's mainly the long discussion that can frighten away people. Is there a way to quickly archive sections, because a lot is obsolete and can be archived. I have a lot more edits than others because I focused on this article. It's my style of working to focus on a single article. Wikipedia guidelines say that there is nothing wrong about that. So, why are you making a big deal about the 1538 vs 145?  You are damn right that it should not be at C-status, but it was in C-status when I started on it. I think it should be reevaluated, but only after the additional work that is required about statistic is done.  What do you mean by argumentative? If you mean that arguments from the sources are given, then it's useful. Readers should appreciate it. You cannot mean that it is a problem.  I have no idea what is your complain at this level.  About verifiability, you must be joking. Every thing is totally verified in the article. If one or two paragraphs have no source, it's because they recapitulate what was already written or maybe a source applies to two paragraphs. It happens some times that sources apply to more than one paragraph.  If there is a few cases where the page is not given, it's the exception and I will check. It could be a reference that was added before I came. Usually, I provide the page. There are not so many quotes, unless you mean notes at the bottom of the article, but this should only be appreciated. It does not affect much the reading of the article.   Now, I really want to know what brings you here?  Did you receive a note or something? Or somehow you got interested in the subject and you noticed the long talk page and you felt some disdain about my style of working and you had to complain and even suggest that I go away. Dominic Mayers (talk) 01:30, 26 June 2020 (UTC)


 * I am going to remove this useless discussion about me. I don't deserve the critics. If you really believe there is a problem, go ahead and make an official complain. I will wait that you respond. After that, I am going to take a break from Wikipedia. I had decided that before. Dominic Mayers (talk) 01:35, 26 June 2020 (UTC)
 * Don't remove this. I am not making an official complaint, I am raising a point about article quality, and I guess about forum posting--even your response here is far from concise. By "argumentative" I mean that the article reads like an argumentative essay, not like a compendium of published material. That you don't understand that objection says a lot. And why I noticed it? I saw this huge edit go by on Recent changes and thought I'd have a look. Drmies (talk) 01:56, 26 June 2020 (UTC)
 * Removing this again and again is just going to get you blocked, and walking away won't make the article any better. If you could sit down and listen, or if you could reorient yourself toward what an encyclopedia expects, you might end up helping the article. Drmies (talk) 02:08, 26 June 2020 (UTC)
 * Such a negative attitude toward someone who provided a lot of useful content in the article. I do not blame myself a second to have expected that you would have removed this text. I cannot believe that it is User:Biogeographist that put this stuff back. I regret very much to have offered my time to improve this article. Obviously it was a big improvement.  So, why blaming me for its supposedly "crappy state" when I only improved it? Yes, long discussions in the talk page are not ideal, but I have not discussed alone. Yes, after some thoughts, I understand what is meant by "argumentative". There are a few sentences there and there and a few paragraphs that create an argumentative style. I know that it is on purpose that I made it that way, because, in my mind, it made the article more convincing.  But, it does not mean that the overall content is not verifiable.  The trick is that one must let the sources themselves do the argumentation. Some sentences there and there made the article do directly some of the argumentation.  This issue could have been addressed without modifying much the content of the article.  The remainder of the criticism was rude and unwarranted.   Dominic Mayers (talk) 02:20, 26 June 2020 (UTC) (copied from before)  — Preceding unsigned comment added by 173.206.72.194 (talk)
 * Yes, Dominic Mayers, it could have been addressed, but it wasn't. How about addressing it now? And how about removing all that quoted material from the citations? See WP:NFCCP. Drmies (talk) 17:29, 26 June 2020 (UTC)
 * I have no problem with the text being removed with the permission of, but one should not remove another editor's comments (and the comments to which they were responding) without the other editor's permission. Thanks for your work on the article. Biogeographist (talk) 17:22, 26 June 2020 (UTC)
 * Having a non-argumentative article is not the problem. The problem is that Popper's solution to the problem of induction is so controversial that, if I understood correctly, some encyclopedia can not find "neutral" experts to cover the subject. This problem can be explained with an analogy. Having different languages such as French, English, etc. is not a problem, because we understand that they represent the same reality, it just a question of translation. But imagine that those speaking English consider that English itself is important and that we represent a different reality if we use French. That's basically the problematic with regard to the different definitions of induction and associated models. Some authors claim that their new model is like a new concept, that is, what is normally only a choice of definitions, a way to express laws, becomes a major contribution by itself. In that view, a natural application of Wikipedia policy is to present this supposed major contribution.


 * It looks OK. Doesn't it? I cannot argue against that. It would be like POV pushing. Yeah, but let's go back to the experts that understand well the model of Popper. What I understand is that none of them would tolerate this interpretation of the policy. It would just make no sense for them to do so. It's like asking them to use bad practices. This is the way I feel. Let me present the corresponding perspective with regard to the Wikipedia policy. It's not POV pushing or original research, because we don't select different concepts. We simply express them using one set of definitions, which is good practices also on Wikipedia. I am not defending this position anymore, even though I still believe that it is perfectly reasonable. It's just that I accept the reality that some will say no no, it's way more than just a definition issue and they will add, for example, that a book was written about this new model and therefore the article is not complete, etc.


 * I am sorry not to be concise, but it's difficult, because it is subtle. In particular, I need to emphasize again that the point is not that using Popper's model we will not present the book. On the contrary, we will be able to present the book perfectly, perhaps even better, but the superficial controversy related to the use of different models and some related claims will disappear. This is still not OK, apparently. That's why this section was entitled "Again we are going nowhere".


 * Please be kind with me. My position is very natural. I have been nice here. I think the criticism directly about me was unfair. You should take it out. It serves no purpose anyway, because the current dynamic will apply to any other subject that interests me and therefore I am quitting Wikipedia. Criticism that apply directly to the article should not be addressed to me, not even indirectly.   Dominic Mayers (talk) 15:12, 26 June 2020 (UTC) (copied from before)


 * And I just saw your last edits of the lead and it is now even further away from matching the article than before.  The article is perfectly verifiable regarding the fact that falsifiability means that the law is contradicted by statements. The existence of multiple parts in the falsifier is also verifiable: there is even a small section that explains why with sources. Sincerely, your edits made me laugh.  So funny. Thank you. You made my day! You illustrated very well why I was not interested anymore in working on this article. I mean, I knew that with the current level of shared understanding, this was likely to happen. It is sad, but it is only an article in Wikipedia.  Dominic Mayers (talk) 19:08, 26 June 2020 (UTC) (copied from before)

Original research in the comparison of "negative pragmatism" with falsifiability.
Here is the sentence that is original research

The fact that William Ernest Hocking presented the notion of "negative pragmatism" is not contested, though I suspect that it existed way before Hocking's time. The issue is the specific connection with falsifiability that is suggested by the position of this sentence in the article. Who made this connection? We would need to see secondary sources that made this connection. The naive connection that is suggested is so opposed to Popper's main arguments that more that one secondary sources should be required. The naive connection that can be made is based on the idea that Popper would have considered the asymmetry between truth and false (as well as falsifiability) at the pragmatic level. This totally miss the point that Popper considered that the asymmetry between truth and false was significant at the logical level only and that, at the pragmatic level, both falsity and truth are subject to the falsification problems, a fact that is well discussed and sourced in the article. The subject of falsifiability starts with the problems of falsification and the related distinction between the logical and pragmatic level. A connection that ignores this distinction is a misrepresentation of falsifiability. Because it suggests this naive connection, this new paragraph is simply weird. It is not supported by sources (in terms of the connection with falsifiability) and it also breaks the rule that the lead should be a summary of the article.

The idea that negative pragmatism existed way before Popper and most likely also way before Hocking, is not unrelated to falsifiability. For example, this idea could fit after the following sentence in the article:

But then it becomes a connection with a pragmatical aspect of Popper's solution and it should be phrased accordingly, making clear that this aspect of Popper's solution is not, by itself, a new contribution made by Popper. Even then, secondary sources would be required to bring any connection with "negative pragmatism". Popper insisted a lot on the importance of this asymmetry, but that's does not mean at all that he claimed that it was by itself his new contribution. Popper's real contribution cannot be seen without considering the important distinction between the logical and pragmatical aspects and how Popper used that distinction to address all the problems of falsifications.

There is a way more serious issue with this sentence. It simply also misrepresents "negative pragmatism" as defined by Hocking. Here is the way negative pragmatism is described in the literature. "Perhaps Hocking's most important contribution to philosophy is "negative pragmatism," which means that what "works" pragmatically might or might not be true, but what does not work must be false."

- https://www.liquisearch.com/william_ernest_hocking/negative_pragmatism and from Hocking himself: "In the preface of a book published some years ago under the title, The Meaning of God in Human Experience, I suggested that pragmatism as a method of reasoning is half true: negative pragmatism, holding that "what does not work is not true," has a validity which cannot be claimed for the positive maxim that "what works is true.""

- Hocking, https://www.questia.com/library/94836417/science-and-the-idea-of-god In this more faithful description of negative pragmatism, the clear distinction between the pragmatic level, where we speak of "not work" instead of "disproved" and the logical level, where falsifiability exists, becomes more obvious. There is a strong bias in the sentence.

173.206.155.30 (talk) 17:52, 7 August 2020 (UTC)

Typo / incorrect meaning
"The Popperian criterion excludes from the domain of science not unfalsifiable statements but only whole theories that contain no falsifiable statements;"

The last part doesn't make sense. I believe this is correct instead:

"The Popperian criterion excludes from the domain of science not unfalsifiable statements but only whole theories that contain non-falsifiable statements;"

Because the criterion excludes a "whole theory" if it contains any non-falsifiable statements.

Currently it says that a "whole" theory is excluded only if it doesn't contain any falsifiable statements. This is not correct; if a theory contains both falsifiable and non-falsifiable statements, it will be excluded.

NOTE: Whoever wrote this is mistaken. A theory is falsifiable if, and only if, it entails at least one falsifiable statement. It's irrelevant if it also entails unfalsifiable statements, e.g. all theories entail all tautologies, and tautologies exemplify unfalsifiable statements.

I am signing this so that it gets archived. The bot has the limitation that it does not archive section without signature. and I changed the date to this edit Dominic Mayers (talk) 18:52, 4 September 2013 (UTC)

Lakatos Vs Popper
I am aware that we cannot make our own thesis in a Wikipedia article. Nevertheless, knowing what notable philosophers have written (as primary source) is important and can guide us in finding proper secondary sources that presented a view point on these writings. With this perspective in mind, here is a quote from Lakatos, which seems to misinterpret the very definition of falsifiability given by Popper.

""A theory is 'scientific' if one is prepared to specify in advance a crucial experiment (or observation) which can falsify it, and it is pseudoscientific if one refuses to specify such a 'potential falsifier'. But if so, we do not demarcate scientific theories from pseudoscientific ones, but rather scientific method from non-scientific method. ...

"Is, then, Popper's falsifiability criterion the solution to the problem of demarcating science from pseudoscience? No. For Popper's criterion ignores the remarkable tenacity of scientific theories. Scientists have thick skins. They do not abandon a theory merely because facts contradict it.""

- Imre Lakatos

We will see that this quote from Lakatos is very close to what Popper wrote about the scientific method. Popper has obviously written a lot of things to motivate the definition of falsifiability, things that were not his definition of falsifiability: we must distinguish between the motivation for a definition and the definition itself. It's the first time that I hear the thesis that Popper's falsifiability does not apply to a theory (or a statement), but to the overall scientific method. We don't even say that in our article. Here is a few quotes from Popper that show how Popper viewed his definition of falsifiability in the larger context of a scientific method. In this first quote, Popper stresses, as he has done in many occasions, that falsification is not possible and he even admits that it creates an issue regarding the applicability of a falsification criteria.

""It might be said that even if the asymmetry [between universal and existential statements] is admitted, it is still impossible, for various reasons, that any theoretical system should ever be conclusively falsified. For it is always possible to find some way of evading falsification, for example by introducing ad hoc an auxiliary hypothesis, or by changing ad hoc a definition. It is even possible without logical inconsistency to adopt the position of simply refusing to acknowledge any falsifying experience whatsoever. Admittedly, scientists do not usually proceed in this way, but logically such procedure is possible; and this fact, it might be claimed, makes the logical value of my proposed criterion of demarcation dubious, to say the least.""

- Karl Popper

Popper continues by responding to his imaginary critic that we might conceive (in principle) of a scientific method that would work, if only scientists would apply it.

""I must admit the justice of this criticism; but I need not therefore withdraw my proposal to adopt falsifiability as a criterion of demarcation. For I am going to propose (in sections 20 f.) that the empirical method shall be characterized as a method that excludes precisely those ways of evading falsification which, as my imaginary critic rightly insists, are logically possible. According to my proposal, what characterizes the empirical method is its manner of exposing to falsification, in every conceivable way, the system to be tested. Its aim is not to save the lives of untenable systems but, on the contrary, to select the one which is by comparison the fittest, by exposing them all to the fiercest struggle for survival.""

- Karl Popper

To my knowledge, Popper was not very successful in describing what is the historical scientific method. The fact that his proposed scientific method has not been applied and would perhaps never be applied does not imply that his definition of falsifiability has no merit. In any case, it is important not to assume that this quote from Popper his a definition of falsifiability. It is only Popper's view on what the scientific method should be, ideally. He is only trying to motivate his definition of falsifiability, which comes later in the book and perhaps appears in simpler terms before. After about 45 pages of analysis, here is the definition of falsifiability given by Popper.

""A theory is to be called ‘empirical’ or ‘falsifiable’ if it divides the class of all possible basic statements unambiguously into the following two non-empty subclasses. First, the class of all those basic statements with which it is inconsistent (or which it rules out, or prohibits): we call this the class of the potential falsifiers of the theory; and secondly, the class of those basic statements which it does not contradict (or which it ‘permits’). We can put this more briefly by saying: a theory is falsifiable if the class of its potential falsifiers is not empty.""

- Karl Popper

Here is one of many other quotes from Popper which explains that we should not confuse his technical notion of falsifiability as a proposed criteria of demarcation and a less technical notion of falsifiability, which only mean that it can be proven false. Popper stresses that falsifiability in this latter sense is impossible.

""We must distinguish two meanings of the expressions falsifiable and falsifiability: "1) Falsifiable as a logical-technical term, in the sense of the demarcation criterion of falsifiability. This purely logical concept — falsifiable in principle, one might say — rests on a logical relation between the theory in question and the class of basic statements (or the potential falsifiers described by them). "2) Falsifiable in the sense that the theory in question can definitively or conclusively or demonstrably be falsified ("demonstrably falsifiable"). "I have always stressed that even a theory which is obviously falsifiable in the first sense is never falsifiable in this second sense. (For this reason I have used the expression falsifiable as a rule only in the first, technical sense. In the second sense, I have as a rule spoken not of falsifiability but rather of falsification and of its problems)""

- Karl Popper

To my knowledge, critics of Popper's falsifiability fall in three categories. Those who critic the consistency of the definition itself, those who critic the usefulness of the definition within any scientific method, those who critic the specific scientific methods advanced by Popper to motivate the definition. Lakatos, in the above quote, did not critic the definition itself, but only a specific view avanced by Popper on the scientific method. Indeed, Popper failed at few occasions to apply his falsifiability criteria within a method to choose between scientific theories (see  for what Popper said about this), but we should not confuse the process of choosing between different scientific theories and the process of demarcating between scientific theories and other forms of knowledge. The falsifiability criteria works well for the latter. Nevertheless, it's a complex situation, because it's natural to view that Popper's definition failed, if we cannot find a way to apply it to pick a theory among many. On the other hand, as explained in Talk:Falsifiability, even Lakatos reused the definition of falsifiability in his more holistic approach.

Reference
Signing this so that it gets archived. The bot has the limitation that it does not archive section without signature. Moreover, it is actually me that wrote this section. Also, I change the date of the signature to the date of this edit. Dominic Mayers (talk) 19:04, 10 June 2018 (UTC)

New subsection after the definition of falsifiability or in the controversy section
I propose to add this new subsection. It's not really a controversy. It simply a confusion about the definition and purpose of falsifiability. But, it is easier to explain after we have presented the Popper-Lakatos controversy. So, I am not sure where it should appear.

Apologia for falsifiability: the bucket and the searchlight
shows that falsifiability is useful to demarcate between scientific and non-scientific theories, but why falsifiability is the criterion for scientific theories? The apparently obvious answer is that we want the possibility to actually and conclusively falsify them. Formally, as explained in, this answer is wrong, because, rigorously speaking, it is not possible to falsify a theory. Moreover, as explained in, falsification plays no decisive role in the choice of a new theory. Therefore, even if it was possible to rigorously falsify a theory, the purpose of falsification and indirectly of falsifiability would remain unclear. For Popper, this apparently obvious answer is wrong and suffers from the problems of falsification because it refers to an inadequate theory of science, the bucket theory of science. His answer (see ) relies on a different theory, the searchlight theory of science.

The bucket theory of science
Some tentative explanations for the growth of scientific knowledge are based on what Popper calls the bucket theory of science. In this theory, observation statements accumulates in a bucket through observations and various procedures are used to make sure that the observation statements are valid. Next, in order to obtain new laws, inference rules are used to process all the knowledge that is available in the bucket. In this picture, Hume argued that we cannot obtain new universal laws (except what can be obtained through deductive rules). Hume's argument is based on reasonable premises: non-deductive rules are in need of justification, circular arguments are not valid, etc. If we accept Hume's premises, all attempts to explain the growth of knowledge in terms of the bucket theory of science are doomed to fail.

Popper's solution to this problem is simply to reject the bucket theory of science. His main argument is basically that he accepts Hume's result. Moreover, to help people get rid of the limitations associated with the bucket theory, Popper remarked that the processes in the bucket are physical processes and the laws that govern these processes are biological. There is no reason why these processes can be described by the objective knowledge already in the bucket together with fixed inference rules. Instead, we can consider that biological predispositions and expectations are responsible for new conjectures, but there is no rules to explain these conjectures.

The searchlight theory of science
Popper proposed to replace the bucket theory of science with what he called the searchlight theory of science. In that theory, Popper wrote, there is no reason why any methodology should work. It is easy, Popper said, to imagine universes where no methodology can work or even only exist. If you want to believe that the methodology will work, it must be postulated as an axiom. In Popper's case, the axiom is that the methodology of conjectures and refutations is going to work. The conjectures are the searchlight, because they lead to observational results. But this axiom will not help any objective rule in the justification of scientific knowledge. There is no point in attempting any justification in the searchlight theory. For a popperian, the absence of these objective rules is expected. It is not a failure.

Popper's scientific methodology that accompanies falsifiability contains rules such as "He who decides one day that scientific statements do not call for any further test, and that they can be regarded as finally verified, retires from the game." In general, the rules of Popper's methodology influence which theories will be chosen or rejected, but these rules do that only through decisions taken by the scientists. As explained in, every rule to determine or choose theories must rely on the good judgement of the scientists.

Why falsifiability is the criterion for scientific theories?
Back to the original question: why falsifiability is the criterion? It is not that falsification directly leads to the rejection of a theory. That would be a rule of the bucket theory. It is not that we must always look for theories that are more falsifiable. That would also be a rule of the bucket theory. Popper's main methodological rule is that scientists must try to guess and corroborate (or equivalently falsify) bold and useful conjectures and take any falsification as a problem that can be used to start a critical discussion. In other words, the usefulness of falsifiability is that falsifiable conjectures say more, because they prohibit more and, in the case of their falsification, they lead to useful problems, which steer the creative process of science. For Popper, this is exactly what we should expect from a scientific theory. The original answer provided, which is the possibility to actually falsify the theory, missed the whole point.

<!--

Resulting confusion
For Popper, this impossibility to explain or justify the creative process of science in terms of rules, does not mean that it is an irrational process. The purpose of science remains a rational discrimination or classification of theories. Scientists and engineers only want to use theories that, for the time being, they consider appropriate for the problem situations that they face. This leads popperians such as David Miller to make a distinction between (rational) classification of truths and (rational) justification of truths. The purpose of science, Miller says, is to classify truths, not to certify them. Popper wrote that, from a rational point of view, we cannot rely on a theory for practical action, but we can prefer it for practical action.

Perhaps without surprise, this has created some confusion. As pointed out by Popper, until the year 1970, he was described as a positivist who was trying to rescue the idea that all authentic knowledge allows verification, which is the opposite of a methodology that always relies on good judgement. Popper referred to this as the Popper legend.

Nowadays, falsifiability is still often misinterpreted as the basis for methodological rules, which do not rely on good judgment and can not work. In particular, as pointed out by Popper, an entire literature has mistaken a contradiction in the falsifiability criterion for an actual rejection. For example, Lakatos wrote that in Popper's methodology a theory that is falsified is also rejected. Lakatos's argument relies on the following quote from Popper: "one must specify in advance under which condition a theory is falsified". But this is not a rule in Popper's methodology. It is only a part of the definition of falsifiability, which, as Popper clearly explained, has nothing to do with actual falsifications or the methodology.

A similar confusion occurs when one uses the degree of falsifiability as a criterion to reject a theory over another. A law is more falsifiable than another, if it prohibits more basic statements, i.e., if it is more precise or universal. Popper wrote "As regards auxiliary hypotheses we propose to lay down the rule that only those are acceptable whose introduction does not diminish the degree of falsifiability or testability", but immediately after qualifies the rule to say that it is not an hard rule. This excerpt is sometimes taken out of context to conclude that Popper's rule requires that we never move to a law that is not more falsifiable, in some cases, even when the current law has been falsified. This is a confusion, because it is inconsistent with a scientific method that relies on the judgment of the scientists. The accepted falsifying basic statement could have been accepted by mistake. Therefore, it indeed often makes sense to hold on to the theory that is more falsifiable, even after it was falsified, but it cannot be an hard rule. -->

Signed so that the bot can archive it. Changed the date to match with the actual date this section was written. Dominic Mayers (talk) 21:24, 2 June 2020 (UTC)

A basic statement in industrial melanism.
The purpose of every example in the section is to illustrate the process of demarcation. This process requires a basic statement. The basic statement is given and only then the law that is falsified by the basic statement is given. In the case of industrial melanism, the basis statement is "In this industrial area, the relative fitness of the white-bodied peppered moth is high." Note that as an observation statement, it must refer to a specific industrial area. A possible improvement would be to use a known industrial area instead of "this industrial area" to make it more clear that it is an observation statement.

If the reader found the basic statement complicated, it is perhaps because he expected an explanation of the law. Or perhaps, the reader completely failed to understand that a basic statement is usually false, because it must contradict a true law. Of course, if one expects some explanation of a law and is given a simple basic statement (which contradicts the law), it can be confusing, as if something was missing or even wrong. But then the fault is not with the simple basic statement.

The fault is likely to be with the lead and how falsifiability is badly introduced in the article. Many people expect the common view on falsifiability when they read the subject title "falsifiability". Unfortunately, how falsifiability is introduced in the article confirm them in their misunderstanding. The common view on falsifiability is simply that the law can be falsified by (true) evidence. It is not a wrong view, because indeed is is believed that all laws, even the law of general relativity is going to be falsified by some future concrete evidence, just as the law of classical mechanics has been falsified. It is not a wrong criterion, but it is not at all this notion of falsifiability that is the subject of this article. Wikipedia is not a dictionary and an article does not have to cover all meanings of a term.

The meaning of falsifiability in this article, when applied to a law, is "the law can be used to make predictions". But falsifiability applies also to other statements, which are not laws. The link between the general definition of falsifiability and its more specific meaning in the case of laws is not obvious. In fact, some authors claimed that "can make predictions" was a better criterion than falsifiability, without realizing that Popper had already explained that it was the same definition when applied to laws. Some people might expect that the examples only illustrate that the laws can make predictions and therefore avoid the use of basic statements. But this would be incomplete and missing an important point, because the notion of basic statement is necessary to give the basic and general concept of falsifiability. There is no way the concept of falsifiability can be understood without it and this entire section does a good job in illustrating the concept of basic statement. It is so sad that some people try to make the article fit with their wrong or limited understanding, which please them, instead of trying to improve the article so that the correct understanding is conveyed. 173.206.25.82 (talk) 17:37, 24 September 2020 (UTC)
 * 173, a simple question - How would you falsify the statement that "this moth is more fit for this environment?" The answer is, you can't, because you don't always know what makes an animal "fit", and it changes from situation to situation. On the other hand "in a soot-polluted areas, you will find more dark colored moths than light colored moths" is a basic falsifiable statement. It doesn't rely on someone's subjective opinion on what animals are fit and what animals are not. You simply measure the pollution and count the moths, and if you don't find dark-colored moths in polluted areas, you know it's false. And, of course, that statement is far easier to understand for everyone who reads are articles, since it does not depend on abstract concepts and doesn't require a background in evolutionary biology. Remember, Wikipedia is not a scientific text, our audience is the non-scientist public. Oiyarbepsy (talk) 18:11, 24 September 2020 (UTC)


 * You wrote You simply measure the pollution and count the moths, and if you don't find dark-colored moths in polluted areas, you know it's false. Simply report this observation and you have your basic statement.  A report could be "In this polluted area, there was many white-colored moths and no dark-colored moth."  We have the exact same notion. The only distinction is the emphasis on the basic statement. This emphasis is justified because it is central in the general concept of falsifiability.
 * To give some context, note that the basic statement itself is falsifiable. It is contradicted by the opposite basic statement "In this (same) polluted area, there was many black-colored moths and no white-colored moth." Yet, none of these basic statements can be used to make (new) predictions. They are not universal laws. The "basic" in basic statement refers to this fact. More precisely, it refers to the fact that they are singular existential statements. The "singular" means that it is about a specific observation in a specific area at a specific time. You also wrote On the other hand 'in a soot-polluted areas, you will find more dark colored moths than light colored moths' is a basic falsifiable statement. This statement is indeed falsifiable just as the basic statements are, but being a law, i.e., a pure universal statement (not specific to any area or time),  it is not what we call a basic statement.


 * The fact that a falsifier must be a basic statement (i.e., a singular existential statement) is essential. Otherwise, the pure existential statement "there were some polluted areas somewhere at some time, in which the relative fitness of the white-bodied peppered moth was high" would be falsifiable, because it is contradicted by the falsifiable (but non-basic) statement "in [every] soot-polluted areas, you will find more dark colored moths than light colored moths". The above pure existential statement, just like any other existential statement, is not falsifiable, because whatever way you try to contradict it, a believer can reply that you have not looked every where or that you looked at the wrong time. It is a fundamental part of falsifiability that non-singular or pure existential statements are not falsifiable.  Anyway, the point is that basic statements, because of their role as falsifiers to show falsifiability, cannot be pure universal statements as in your example. They must be singular existential statements. Popper wrote many sections in LScD to explain this. It is an essential part of falsifiability.


 * I sympathize with those who feel, as you seem to feel also, that it is more natural to explain the criterion directly in terms of predictions: a law is falsifiable if it makes predictions that can be false just as in your example. It's more natural, because it is implicit in the concept of prediction that both the initial condition and the prediction can be measured with some available technologies and also because the prediction is one possibility among many. If the prediction was the only possibility, it would not really be a prediction. There is a lot that is hidden, but yet natural and implicit in the notion of a falsifiable prediction. The concept of basic statement is necessary to make these hidden notions explicit, just like I did with your example with the help of your own argument.  It is important that we make these notions explicit, because otherwise we get a confusion with the common notion of falsifiability, which I described above. Popper complained so much about this confusion in the literature.  The only way to avoid this so common confusion is the concept of basic statements. Besides, the concept of basic statements is also required to obtain the general definition that does not refer to predictions and is thus not restricted to laws.


 * 173.206.25.82 (talk) 19:01, 24 September 2020 (UTC)

Sources with a synthesis of the common and technical definitions
There has never been a consensus about the definition of falsifiability to use in the lead. Many would prefer the common definition, which refers to actual observations in empirical tests, while adding a qualifier such as "in principle", "eventually", etc. or use expressions such "has the capacity", "has the potential", etc. to account for the fact that these observations might not be possible. This common definition conflicts with the technical definition, which insists that the falsifier is only a basic statement, also called a test statement or an observation statement. The distinction is important, because in the technical definition, there is no direct reference to observations. It is sufficient that a contradicting basic statement exists in the same language that is used to state the law. So, technical falsifiability requires, first, that the language to state the law is agreed upon and, second, that there exists a falsifier of the law in this language. In this way, falsifiability is defined in a purely logical manner, at the level of statements and their logical relationship only, no reference at all to actual observations to evaluate the truth of potential falsifiers.

It is a very serious issue in the literature on falsifiability, because the common definition has been interpreted in a completely different manner than the technical definition by many philosophers, including illustrious philosophers such as Imre Lakatos and John W. N. Watkins. The difference is that they consider that the falsification that exists in principle, i.e., the potential falsifier,  must not only be possible in the formal language of the theory, but must also be provably a falsifier in the theory that underlies the observations. For example, to my knowledge, Lakatos never stated in his work the technical definition of falsifiability. He certainly did not in "The methodology of scientific research programmes." In this book, he went instead into an analysis of different levels of falsificationism, namely, dogmatic falsificationism, naive falsificationism and sophisticated falsificationism, raising different issues with the common view of falsifiability at each level. Similarly, Watkins is very explicit that Popper's basic statements as potential falsifiers are "absurdly overstrong", but yet cannot actually falsify a theory. He wrote (emphasis mine): "Popper undertook the impossible task of upholding the falsifiability of just this theoretical core. Apparently forgetting that he had once said 'Duhem is right when he says that we can test only huge and complex theoretical systems', Popper set out to devise potential falsifiers just for Newton's fundamental assumptions. [Popper's examples] are absurdly overstrong considered as [potential falsifiers] of a properly fleshed out version of [Newton Mechanics] but, of course, not strong enough to be [potential falsifiers] of this irrefutable core. They mostly involve planets moving in highly erratic ways. But Newton's laws of motion plus his law of gravitation say nothing about the physical makeup of the planets; in particular they do not rule out the possibility that the planets are enormous rocketlike devices (see O'Hear, 1980, p. 102) that can accelerate themselves in all sorts of ways. Another example has apples that have fallen from a tree rising up and dancing round the tree. But Newton's laws of motion and law of gravitation say nothing against the possibility of such a spectacle being produced with the help of invisible elastic threads."

- Watkins

But Popper totally agrees with the fact that, if we start to consider how the actual observations that are described by basic statements provably or demonstrably falsify the theory, Newton's theory can never be actually falsified in practice by any observation. As pointed out by Thornton, Popper's logic is utterly simple: If an apple does not fall under gravitation, then it cannot be that all objects fall under gravitation. A potential falsifier and its associated contradiction in Popper's falsifiability remain entirely at the logical level. The problems of proving an actual falsification are irrelevant in falsifiability. Thornton's precise statement is "Popper has always drawn a clear distinction between the logic of falsifiability and its applied methodology. The logic of his theory is utterly simple: if a single ferrous metal is unaffected by a magnetic field it cannot be the case that all ferrous metals are affected by magnetic fields."

- Thornton

Not only Thornton, but also Chalmers and others understood that falsifiability is indeed on purpose defined in terms of what Watkins qualifies of "absurdly overstrong" potential falsifiers, at a purely logical level. For example, to show the falsifiability of the law "Heavy objects such as a brick when released near the surface of the earth fall straight downwards if not impeded", Chalmers gave the following "absurdly overstrong" potential falsifier: "The brick fell upwards when released."

- Chalmers Note how this potential falsifier logically contradicts the law in a very simple manner. This cannot be denied. The fact that perhaps an invisible elastic thread was attached to the brick, as Watkins would point out, does not change the fact at the formal level, simply looking at the logical relation between the two sentences, the law is contradicted. So, clearly when Watkins says that these "absurdly strong" statements are "not strong enough to be [potential falsifiers] of this irrefutable core [of Newton's gravitational law]", he refers to problems in an actual falsification. It might be worth to quote Kuhn on this respect (emphasis mine): "To be scien­tific a theory need be  falsifiable only by an observation statement not  by actual  observation. The relation between statements,  unlike that between a statement and  an observation,  could  be the conclusive  disproof familiar from logic and mathematics."

- Thomas Kuhn It is even a part of popular culture. For example, it is well known that to show the falsifiability of the theory of evolution, J.B.S. Haldane gave the "absurdly overstrong" potential falsifier: "fossil rabbits in the Precambrian". Again, there could be plenty of ad hoc explanations for the actual discovery of a fossil rabbit in the precambrian, but it still remains that it is a valid potential falsifier that was clearly accepted to show the falsifiability of the theory of evolution, because the simple logical contradiction with the law is there.

The current article does a very good job at dealing with this confusion. The point of view of Lakatos is fully presented, especially the problems of falsification as seen in the different levels of falsificationism. No controversy is hidden, except one: the controversy about the different definitions, but that is because it is not sufficiently notable. There exist no synthesis that discusses explicitly the two definitions as being opposed. Instead, what we have are excellent sources that very clearly and unambiguously present one of the two definitions, but none present a synthesis that is explicit about the confusion between these two definitions. I am OK with this. Delving with confusion around definitions is not interesting and it is not useful for the readers. Popper, of course, complained a lot about this confusion. For example, he wrote about 10 pages in his 1982 Introduction to Realism and the Aim of Science discussing in details this confusion and insisting that falsifiability is purlely at the logical level. However, this is a primary source.

On the contrary, instead of discussing this confusion between two distinct definitions, some excellent authors have proposed a synthesis of the two formulations as if they correspond to a same definition, while making clear that the unique definition is the popperian definition given in terms of basic statements, no direct reference to observations. For example, Thornton wrote: "Accordingly, [Popper] held that, from a logical perspective, a system of theories is scientific only if it is refutable or falsifiable: "I shall not require of a scientific system that it shall be capable of being singled out, once and for all, in a positive sense; but I shall require that its logical form shall be such that can be singled out, by means of empirical tests, in a negative sense: it must be possible for an empirical scientific system to be refuted by experience."

- Popper This demarcation criterion was most clearly defined by Popper in terms of the relation that holds between a scientific theory and ‘basic statements’, where the latter are to be understood as singular existential observation-reports of the form ‘There is an X at Y’. On this definition, where a theory is scientific, it must exhaustively divide the class of basic statements into two non-empty subclasses."

- Thornton

A synthesis that unifies two apparently opposite formulations is even more clear in Nickles' view: "Popper proposed falsifiability (empirical refutability) as the criterion of demarcation. For him a statement is scientific if and only if it is falsifiable in principle, that is, if it can fail an empirical test. This is equivalent to saying that there must be some possible observation statement (true or false) that logically contradicts the claim in question."

- Nickles

Note that the second formulation is only in terms of observation statements. Note also that despite a mention of "an empirical test" in the first proposed formulation, it is clear in Nickles' view that Popper's falsifiability is disconnected from the "body of belief-worthy claims about the world, let alone demonstrably true claims". He wrote: "Reflecting on the steady weakening of proposed criteria of demarcation, Laudan (1983) concludes that demarcation is no longer an important philosophical problem. Popper’s falsifiability criterion, he says, weakens the criterion almost beyond recognition. No longer does the criterion of demarcation mark out a body of belief-worthy claims about the world, let alone demonstrably true claims."

- Nickles

— Dominic Mayers (talk) 19:29, 29 October 2020 (UTC)

The following statement from WP:OR is important in the situation that is described above: Rewriting source material in your own words, while substantially retaining the meaning of the references, is not considered to be original research. Doing that is necessary here because the technical definition looks weird at first and the common definition is misleading. The important is to retain the meaning of the references. The technical definition briefly stated is "The statement is falsifiable if it is contradicted by an observation statement." The problem is that "an observation statement" seems a weird way of saying "a possible observation". The issue is that the reader has no clue that in the concept of falsifiability, the contradiction must only exist in some language, which abstracts away many experimental issues. When one uses "a possible observation", it can be interpreted and has been interpreted by Watkins, Lakatos and many others to mean that the contradiction must also take into account all possible experimental issues such as invisible elastic threads. The only way to remove the weirdness while retaining the meaning of the references is to make sure that this purely logical context is understood. So, what is needed is one or two natural sentences that define falsifiability by making reference in a very simple manner to the concepts of statements, logical relations between statements, etc., all of this separated from actual observations. I mean that the reader should not have to guess it from "observation statement" only. It should be much more explicit than that. It is likely to be the wrong guess anyway. This is the challenge that we face. — Dominic Mayers (talk) 17:21, 1 November 2020 (UTC)

Search for a formulation in our own words of technical falsifiability in the lead
The goal is to convey the technical notion of falsifiability in a way that is natural and simple and yet cannot be misinterpreted as the common notion. The logic behind this exercise is that if the weirdness disappears only because the statement can be interpreted to mean the common notion, then the real issue has not been addressed. Once we have a good formulation for the technical notion, a second definition of falsifiability that refers or might refer to the common notion should appear unnecessary, even confusing. The common notion corresponds to falsficationism in the article, as it is the case in Lakatos' book. It is not ignored at all. Only the conflict of definitions must be avoided because it is confusing for the readers and not notable in the literature. The key point is that this statement will have to be in our own words. Please avoid some kind of Witch-hunt against false OR because you see a new formulation. There is no ambiguity that the meaning is verifiable. Here is a first idea, but I am sure that we can do better: "In the philosophy of science, a theory is falsifiable if it respects two conditions. First, it must be written in a language with a conventional empirical interpretation. Second, it must be contradicted by some basic statements of that language, statements that can be interpreted as true or false observations. The basic statements that contradict the theory are the potential falsifiers. Falsifiability is independent of whether the contradicting statements are true or false."

-- Dominic Mayers (talk) 18:51, 2 November 2020 (UTC)


 * You place a lot of emphasis on language, but scientists would probably place more emphasis on disconfirming experimental observations. AnonMoos (talk) 01:54, 20 March 2021 (UTC)
 * There is nothing in the text that says that evidence or "disconfirming experimental observations" are not important in science. On the contrary, falsifiability is about concrete evidence through the required empirical language. Perhaps, this is not emphasized enough. It is important to understand that even so called absurd potential falsifiers, such as the "The brick fell upwards when released" from Chalmers (see above), refer to a language that is empirical. Therefore, you are right that science is about concrete evidence. However, clearly the potential falsifiers must be allowed to contradict fundamental laws, that is, they must be allowed to be totally absurd, because that is their role. Their role is to contradicts even the most fundamental laws. This is the key point. It has to be very clear in the lead. Dominic Mayers (talk) 02:47, 20 March 2021 (UTC)
 * Here are the new introductory statements, which refer more clearly to scientific evidence: The term "logically possible" is introduced to describe in "our own words", as suggested in Wikipedia rules, the concept of basic statement. The term "basic statement" was a bit too abstract for the lead. The "logically possible" has the same meaning as "potential" in "potential falsifier". Popper said of potential falsifiers that they are events that are logically possible. Dominic Mayers (talk) 10:25, 5 May 2021 (UTC)

"Could get" instead of "is"
Concerning this diff, the "is" is perfectly fine and simpler than "could get". It corresponds to falsifiability, the logical concept defined by Popper, the subject of the article. The use of "could get" is NOT easier to understand or more correct. It only presents a different notion than Popper's falsifiability which some prefer and thus perceive as simpler or more correct. As a logical criterion, falsifiability is not something vague that requires a "could get". It's very simple: if there is an observation statement such as "the brick fell upward when released", then the statement "all objects follow Newton's gravitational law on earth" is falsifiable. One might feel that "could get" is better, because we don't know if that observation statement obtains or not, i.e., if the brick really felt upward or not. But, the most important aspect of falsifiability is that it has nothing to do with whether or not the observation statement obtains. The distinction between the logical side and the methodological side of science, only the latter dealing with actual experiments, is at the core of Popper's philosophy. Dominic Mayers (talk) 07:59, 17 October 2021 (UTC)


 * a) Only possible observations occur. b)is contradicted by an actual confirmed observation means that the theory is toast already. I guess there is confusion about to two meanings of the phrase "possible observation": An observation that's not impossible, or an observation that has not been seen yet. In the article, the first meaning appears to be used. So in that context, "is contradicted by possible observation" could be read as "is contradicted by a an observation which was not impossible.", but this is not what was meant, because that then theory is already false.
 * If a theory is contradicted by an actual observation (whether "possible" or not), it is false already, not just falsifiable. The "is" takes on the meaning of "has been" upon parsing. One could also clarify that the possible observation is not necessarily assumed to actually have occurred (in which case it is not an actual existing observation yet) Jmv2009 (talk) 17:23, 21 October 2021 (UTC)
 * I agree that falsifiable does not mean falsified by an actual observation. The current text in the lead does not say that, but the fact that you comment here suggests that it could be made clearer. Before we discuss how it could be made clearer, let us make sure here that we agree on what is to be made clear. It is important that falsifiability is a logical concept that has nothing to do with an actual event that has already occurred or will occur in the future. If you know about logic, you know that, for a fixed theory, a statement that is false remains false irrespective of events in the real world: it depends only on the logical theory and the statement. In the same way, a theory that is falsifiable remains falsifiable irrespective of what happens in the real world: it depends only on the language, which must have an empirical interpretation. Even the empirical interpretation is independent of specific events: it's the same empirical interpretation no matter which events are actually observed. So, once the language and its empirical interpretation is fixed, we do not have to look at all at what actually happens in the real world to determine if a theory is falsifiable. The definition is purely logical: if the theory is contradicted by a logically possible observation (a purely logical concept), it is falsifiable. One way to make it clearer would be to say that a theory is falsifiable if it is contradicted by an observation statement, also called a basic statement. However, some people do not see that this observation statement must have an empirical interpretation. It's not any observation statement, but an observation statement that has an empirical meaning like "the brick felt upward when released". The mistake that people do when they try to understand falsifiablity is failing to make the clear distinction between the purely logical criterion and the   methodology that deals with actual observations. Thornton in the SEP article explains that Popper insisted a lot on this distinction.
 * So to sum up, the challenge is that two very important ingredients or concepts are part of falsifiability:
 * the language and thus the falsifiers must have an empirical interpretation in the real world and
 * a theory is falsifiable or not irrespective of what happens in the real world. It is falsifiable if there is a falsifier for the theory.
 * The fact that these two ingredients superficially appear in contradiction is perhaps the challenge. It's important not to explain one of the two ingredients in a way that creates confusion about the other ingredient. Dominic Mayers (talk) 18:16, 21 October 2021 (UTC)
 * It's just that the first sentence in the lead should not parse as: "This theory is contradicted by an observation (...)" results in the theory is falsifiable. Because the consequence of "This theory is contradicted by an observation" is that the theory is toast. At least cursory, it would not occur to a reader that the alluded observation is not an actual observation, but just an hypothetical one. (Try to parse the sentence without any pre-existing knowledge of the words "possible" and "falsifiable") Jmv2009 (talk) 20:29, 21 October 2021 (UTC)
 * I understand perfectly. In fact, I did not know anything about falsifiability when I first started to edit the article and I had the same experience as you have. The first sentence said something like "the theory must be contradicted with a possible observation". The idea was that the term "possible" was supposed to make a big difference. But, the term "possible" was not defined. In a very natural interpretation of "possible", the theory is "toast" as you say. I mean, clearly, if the theory can possibly be in contradiction with an actual observation, then it is toast. It does not matter that the observation did not happen yet and might even never happen. Similarly, the term "potential" does not help much. It's the same thing. This is why I am saying that I understand the issue, but adding "a potential" is not a solution, because, in a very natural interpretation of "a potential", the theory is still toast. You might feel that, on the contrary, adding "a potential" is sufficient to prevent the "toasting" of the theory, but then it's not clear at all why. This is not falsifiability. Falsifiability is a very well defined concept. It does not rely on vague concepts. The reason why a falsifiable theory is not "toast" is very clear: the falsifier is only an observation statement, which has nothing to do with what has happened in the past, happen now or will happen in the future. Alan Chalmers's example makes this clear: "The brick felt upward when released". Clearly, this does not say that Newton's theory is false. It only says that it is falsifiable because there is a potential falsifier. And there is nothing complicated in the notion of an observation statement that contradicts a theory. It might be too simple and people who do not know expect something else, but we cannot give the incorrect, more vague and more complicated definition, because people do not expect the simple notion. It's true that falsifiability also requires that the observation statement has an empirical interpretation, but this is something else. It's called the material requirement aspect of falsifiability.  Dominic Mayers (talk) 22:21, 21 October 2021 (UTC)

In any case, I modified the text to answer your concern. Now, it reads "This theory is contradicted by an observation statement (...)" It cannot be parsed as "This theory is contradicted by an observation (...)", because "observation statement" is not the same thing at all as an "observation". However, it might be useful to make the point even clearer by adding a complete sentence. Dominic Mayers (talk) 22:39, 21 October 2021 (UTC)

Adding "a potential" before observation
Concerning this diff, it is an attempt to resolve a contradiction that is only there because of a misunderstanding. I don't think that simply adding the word "potential" like that resolves the issue, because the meaning of "potential" here is not clear. It's not defined. It could mean anything. Again, falsifiability is a precise logical concept. It should not be needed to add a word like that. If there is something not clear for, then sure we should try to make the text clearer, but adding this single world is not the solution. It might be necessary to add a complete sentence that goes to the heart of the misunderstanding to resolve it. Well, for now, I replaced "observation" by "observation statement". The notion of statement is clear and its use here corresponds exactly to the definition of falsifiability. Dominic Mayers (talk) 18:52, 21 October 2021 (UTC)

Falsifiability as a requirement for predictive power.
Concerning this diff, predictive power is an interesting aspect of falsifiability, but it is only valid when applied to universal statements and falsifiability applies to singular existential statements as well, in particular, to basic statements. This more general scope is part of the concept of falsifiability. This other interesting aspect, which is opposed to this general scope, is only mentioned briefly in the small section "Initial condition and prediction in falsifiers of laws". The usefulness and role of falsifiability is already mentioned in the last paragraph of the lead. I moved the sentence there, where it fits. The usefulness of falsifiability is the capacity to test and this applies even to simple observation statements with no predictive power (because location and time are fixed). It's more general. For example, we can test a statement about a particular star at a given time (necessarily many years in the past). But, because testing is typically (incorrectly) associated with prediction, I mentioned it anyway in the lead. Dominic Mayers (talk) 08:23, 17 October 2021 (UTC)

Methodology versus the logical criterion: avoiding original research or lack of neutrality.
These recent modifications illustrate well the need to understand the sources regarding falsifiability as a logical criterion that is separated from methodology. In many high quality sources such as the SEP article about Popper, but more importantly in Popper's books directly, it is very clear that Popper's falsifiability, the subject of this article, is separated (but of course interacts with) the methodological aspect of science. However, even though the importance of this distinction is verifiable in high quality sources, it is also not well understood in many other high quality sources. It's a confusion that reflects the current state of knowledge in the literature. Note that it is not claimed here that this confusion is well explained or even well discussed in the literature, because, if it was the case, then there would be no confusion any more. Therefore, to avoid original research (and be verifiable), because of a lack of sources, not by lack of pertinence, the subject of the article is not this confusion. Instead, the current article tries to present the definition of falsifiability to its best as expressed in many high quality sources that understand this distinction, which is intrinsic in the definition. Yet, it also tries to present the other view, especially the one of Lakatos, but not in a way that will prevent the readers to understand the correct definition. Doing otherwise would be a POV pushing that favors the point of view that ignores the distinction, a lack of neutrality to avoid. Dominic Mayers (talk) 15:01, 17 October 2021 (UTC)

Dense language
I had a lot of trouble even getting through opening paragraphs due to it being really densely packed and using some strange jargon. A few key phrases in particular did not read well to me: "Obtains" as it is used in this article is not used how I expect it to be used, as I expect "obtain" to be a transitive verb. To hear that something "obtains" but not know what it is obtaining confused me greatly, and I read sentences over and over to no avail. "The contradictory observation statement must correspond to a state of affairs, be a potential falsifier, but it can be imaginary." is another sentence where the grammar caught me off guard, the "must do x, be a y, but z" struck me as weirdly conversational. I'd expect a structure like "must do x, or be y, but z", "must do x and be y, but z", or "must do x (i.e. be y), but z". Overall the opening paragraphs are dense with examples that I'd expect later in the article, maybe in the second section, and they are rather confusing to someone without a background of knowledge on the topic and terms. I don't know how to quantifiably explain it but it doesn't sit well with me, so I hope someone else will take a look to confirm that I am not alone (or that I am). TheJonyMyster (talk) 12:31, 27 October 2021 (UTC)


 * Except for blaming the use of useful examples, thank you for these fine critics. The language issues can be easily corrected and they will be. However, you have other concerns that have nothing to do with the language. These other issues are delicate, because the concept of falsifiability is often misunderstood, often confused with Lakatos' falsificationism and its problems. It's inevitable that readers will regularly come here to complain that it's not easy to follow, when it's only that they expect something simpler that is not falsifiability, but a misunderstanding, away from the main concept. So, there must be a careful discussion regarding conceptual issues, such as the notion of imaginary state of affairs, before we modify the article accordingly. Just saying that it is dense, etc. is not sufficient. How do you understand falsifiability (from the sources)? In which way the lead says something else?  Again, the language issues will be corrected. My concern here are the conceptual issues only. Dominic Mayers (talk) 16:15, 27 October 2021 (UTC)
 * I have modified the article to correct the mentioned language issues. However, again, there were also unspecified conceptual concerns in the above comment. Again, it is to be expected that some readers will have issues. When it is a mathematical or theoretical physics article, the readers understand that there are subtleties that are fundamental and cannot be avoided, but when it is a philosophical article, some expect that there is no subtleties of this kind. That is an incorrect expectation. So, a discussion is needed to see that the request is for a better explanation of these fundamental subtleties (so that a general audience can understand), not to remove them. Dominic Mayers (talk) 20:48, 27 October 2021 (UTC)
 * Finally, I moved the challenging examples in the body of the article. The examples made the key points mentioned in the lead more explicit. They are very good examples taken from excellent sources, so it appears natural to include them in the lead.  However, many readers expect a different notion and can get confused by these examples.  The rational to move them is that the remaining text still mention the key points and, even though it is due to an incorrect expectation from some readers, it is better to avoid confrontation and give a chance to these readers to more progressively get the main points. Dominic Mayers (talk) 16:16, 29 October 2021 (UTC)

The problem with new and incomplete theories which aren't falsifiable at birth but might evolve to be later
New and incomplete theories aren't necessarily well postulated from the beginning, neither do they evolve in the exact order of causal hierarchy (the ingredients within theories don't evolve in the manner we teach them but usually in a chaotic, non-didactic manner). Thus people who overestimate falsifiability, they might end (or go against) a theoretical evolution that could result into a falsifiable theory.

Falsifiability is very important, but not a reason to stop evolving new theories. — Preceding unsigned comment added by 2A02:587:4F17:6DB4:BCD2:B83A:A527:E0E9 (talk) 15:28, 14 November 2021 (UTC)


 * You refer to the paradox that falsifiability is presented as a criterion of demarcation between science and non-science and at the same time science is described, even by Popper himself, in terms of research programs that do not need to only contain falsifiable laws. In fact, Popper is the first to have insisted that many things in these research programs are useful and not falsifiable. Popper referred to these research programs as metaphysical, but Lakatos did not. However, Popper never used the term "metaphysical" in a negative manner. Even Quantum Mechanics as a general framework is not falsifiable, because it's not specific enough, and thus qualifies as metaphysical. To understand the situation, imagine that we had no specific models of Quantum Mechanics that are falsifiable. In that fictive situation, the framework of Quantum Mechanics would much less deserve to be called scientific. Imagine further, that, for some reason, the kind of observations referred abstractly in Quantum Mechanics could hardly have an empirical interpretation even in a foreseen future, thus making falsifiability out of reach. It would then deserve even less to be called scientific.  The point is that the reference criterion is falsifiability. It is only in that sense that falsifiability is a criterion for the demarcation of science and non-science. Dominic Mayers (talk) 16:34, 14 November 2021 (UTC)

Checking how the 170K distribute over sections
Dominic Mayers (talk) 11:59, 22 November 2021 (UTC)

Broken redirects
in this edit you changed a heading that was a redirect target, and you added anchors, but you didn't change the redirects, resulting in broken redirects. I changed those redirects so that they are not broken. I noticed there are a couple of other broken redirects: and. Anyone care to propose a proper redirect target for those? (I apologize for not participating further in the RfC above; I did not receive a notification when Dominic mentioned me, and I don't have time to read the discussion right now.) Biogeographist (talk) 04:55, 1 December 2021 (UTC)
 * I spent some times trying to figure out a good redirect or to decide whether it would be better to let the search engine do the work through its index, but could not reach any conclusion. Thank you for looking at this. It's not easy. If I was forced to vote, my vote would be a redirect to Falsifiability. Dominic Mayers (talk) 11:05, 1 December 2021 (UTC)
 * I was sure I had dealt with those - thanks for fixing that! For the other two pages, redirecting to Falsifiability sounds good to me. Sunrise (talk) 03:30, 3 December 2021 (UTC)

Still garbage 10 years later.
I'm sorry to see that the same obsessive compulsive, and incorrect, author still controls this article with verbose pendantic, dismissive verbiage and out of context jargon. What has changed is the air of superiority and self-righteousness with an air of dismissiveness to the steady stream of objections, —has increased. I think the technical term for that is browbeating. Or is it gaslighting?

In short this author has swapped out the meanings falsifiable "(or refutable)" in his mind with the meanings of falsified or refuted. Falsifiable does not mean falsified.

10 years ago (?) this author was wrongly suggesting that the discovery of the black swans is what made the statement "all swans are white," falsifiable. No, it was falsifiable before the black swans were discovered because it was possible that black or red swans might be discovered. However "ghosts exist" cannot be falsified, ever, -just try it. That, in a nutshell yields the definition of falsifiable. No gobbly gook nor mental contortionism required, it's a pretty simple concept.

Simple, yet this concept is the bedrock of science itself. "Falsifiable" is sometimes used loosely as a synonym for testable.

I guess our favorite author here sees himself as surfing on the edge of human understanding itself.

This obvious error has caused a number of logical contradictions which in the last 10 or 15 (??) years, that I see have been patched and patched again so that now there appears to be more patch than original explanation. One obvious example of this is the caption under the illustration of the two black swans, which for years indicated the true, clean and clear definition of falsifiable,— in contradiction to this author's version of reality. But now even there I see the introduction of this silly, impenetrable gibberish. Or is it gobblygook?

What I hate to see is the steady stream of objections over the years being overcome by this author's undying energy -some might call it obsession,- until like me they walk away shaking their heads thinking it's only a damned Wikipedia article. I guess I was hoping a hero would eventually come along.

But what really bothers me the most is 10 years worth of people who have been miseducated by this piece of crap. I think Wikipedia needs to spend some money here and find somebody with some credentials who cannot be gaslighted nor browbeaten by an unending stream of haughty gibberish and confused spaghetti logic.

Or who knows, one might be able to look in a real encyclopedia to check things out? Again, falsifiability is not a complex concept. One would never know that by attempting to read this mess. I think the easiest fix might be a complete rewriting from scratch.

Sorry for my crude Android phone editing. --Doug Bashford. 2607:FB91:798E:19D5:8CB7:C3FF:FE80:910D (talk) 18:51, 2 December 2021 (UTC)
 * Don't expect philosophy of science to be very simple conceptually. It's not that complicated either, but the logical nature of falsifiability is important. That's not me that says that. It's the sources. It's not for nothing that Chalmers uses the example "the brick fell upwards when released". It's not for nothing that Thornton says "Popper has always drawn a clear distinction between the logic of falsifiability and its applied methodology. The logic of his theory is utterly simple: ...". It's not for nothing that Popper himself wrote 4 pages in the 1982 Introduction of Realism and the Aim of Science only to explain the distinction between the logical and methodological aspects of science and conclude "An entire literature rests on the failure to observe this distinction." But I can see that it's challenging for you and others and it might be necessary to adapt the approach. If you don't understand why the logical nature of falsifiability can be important, to make yourself useful, you need to work with those who understand the sources instead of blaming them.  Also, Wikipedia is not a democracy. It's not the opinion of the majority that matters. It's the in dept discussion of the sources that matters. I say "in depth", because picking some sentences in sources to support your view and then present this to argue that it's verifiable while ignoring the global context would not be conducive to a useful discussion.  Dominic Mayers (talk) 20:02, 2 December 2021 (UTC)
 * I have to laugh when you say that it is me with the failure to understand, while you claim that your bloated imaginary world is a better explanation for Wikipedia's targeted Everyman than Encyclopedia Britannica's clear explanation.  Just like the flat-earthers with their detailed, well-constructed imaginary world, they know more than all the real experts do anyway, right?  And all of the evidence supports them, they really believe that. What do those silly scientists know, they're all on-the-take anyway right? But you're not overtly challenging Science are you? You claim it's your buddy. You claim your respect it.   I laugh.  I'm sorry that you found the philosophy of science to be so confusing that you were befuddled and apparently tried to work that out here in this article.  But it seems you still haven't been able to figure it out so why don't you go home now.   And you misrepresenting my words pretty pretty well shows your integrity here doesn't it?  I never said science was very simple, I said it's not the incomprehensible spaghetti knot that you have tied it into in this article.  Again I'm sorry that you find this concept so complicated.  But your attempt to untie it here in public is doing nobody any good.  It is your personal knot, nobody else sees it.  Those explanations you cited were not complications, nor germaine to our conversation.  You might take a hint from them.  — Preceding unsigned comment added by 2607:FB91:798E:19D5:8CB7:C3FF:FE80:910D (talk) 21:01, 2 December 2021 (UTC)


 * Let's get to some specifics here. I can't believe somebody would put their name on the first two sentences of this article, and claim that was a clear explanation, or even decipherable. It looks more like somebody struggling in their own mind with something that I cannot comprehend.  It doesn't even have enough solid substance to criticize. Since I already understand the definition, with a great deal of charity I can guess what was intended, but could somebody seeking meaning? This does not meet Wikipedia standards.
 * Regarding the caption under the black swans photo: the author seems to be confusing the word "possible" with the diminutive "imaginary." Doesn't that imply things like "imaginary friend," fairy tales and such?  (It's like the author is being forced to finally admit the truth.)
 * Another author has already complained about the improper grammar and usage of the word "obtain" in the Talk topic Language Too Dense. The reply was undecipherable. It's usage destroys the entire caption. What would the typical Wikipedia user think or feel?
 * This article like all imaginary worlds can never, ever be falsified. It should be deleted on the grounds that it is diseducational, bewildering, confusing, it is not explanatory, it does not address Everyman, and it contradicts the founding principles upon which Wikipedia was founded.
 * It hurts me to think of the number of budding scientists who came here and walked away frustrated and confused and possibly discouraged or worse.

--Doug Bashford — Preceding unsigned comment added by 2607:FB91:798E:19D5:8CB7:C3FF:FE80:910D (talk) 22:09, 2 December 2021 (UTC)


 * I don't think this is going anywhere. Whatever I say, you simply continue to express that you are confuse and you seem to be, but yet I don't see any effort from you to understand, i.e.. to show some trust that I use the sources adequately. You show no sign of willingness to collaborate. Instead, you seem convinced that it's not worth it. Anyway, I will answer your claim that Britannica is better. I am not rejecting Encyclopedia Britannica on Falsifiability. I am sure it's fine. I will try to read it even though it's a paid encyclopedia. We should not reject anything. However, points of view must be properly attributed. Popper's falsifiability, according to most sources and Popper himself is a logical criterion, so this is the definition adopted in the article.  Lakatos presented falsificationism and somehow saw it as the essence of Popper's falsifiability and ignored the logical aspect. This has also a lot of weight in the literature - we can see this angle on falsifiability a lot, but this is there in the article as a different view. There is nothing hidden. I will be glad to consider Britannica as another source. However, it's not going to change the logical nature of falsifiability, which is so well sourced.  Dominic Mayers (talk) 22:29, 2 December 2021 (UTC)
 * I just read the so called article on Britannica. It's a small one paragraph entry on the subject written by Matt Stefon. I see that it has a first sentence that does not refer at all to the logical nature of falsifiability. I think some compromise is needed to avoid challenging too much people. I was planning to propose such a compromise anyway, even before the IP's criticism. I wish, the IP would nevertheless realize that many excellent sources, way more qualified  than Matt Felton, emphasizes the logical nature of falsifiability.  Dominic Mayers (talk) 23:00, 2 December 2021 (UTC)

For reference, the first sentence in Britannica is Dominic Mayers (talk) 23:41, 2 December 2021 (UTC) I just added the compromise I had already in mind in the RfC section. Dominic Mayers (talk) 08:30, 3 December 2021 (UTC)


 * I am so happy now that we have had a "compromise" bestowed upon us!  But I'm sad because I'm pretty sure that "cooperation," is not in the cards.  Instead I see a fight and a hassle for every inch. Somebody thinks that as God, they own this article.
 * Yes, Encyclopedia Britannica explained the concept of falsification in 150 words, but the same author gave a more complex topic, "Phenomenology of Religion," over 800 words.  How many words does this article have by now?   I think this illustrates Design Vs.  Rube Goldberg; -it also reminds me of the monstrous Winchester Mystery House with its doors and stairs that go nowhere, built by an obsessed woman.
 * After visiting some of the talk sections above I was somewhat surprised to see that you were indeed capable of writing clear, clean, lucid sentences, sometimes several in a row. In my book, clear, clean writing is good writing, is good communication.  Good communication!  I love that stuff.


 * This raised a question in my mind, Well then, why does he come to the article and write such consistent garbage?

I think it's because you are giving a number of things higher value and priority than the communication of large concepts from your mind into the mind of Everyman. If you can't do that here, you are a Failure.


 * If that is NOT your highest priority, please go away.
 * I was going to suggest that you go read the clear, clean "Communication" article.  But not being a fan of lazy-links, I revisited it first.  It now has almost the same disease as this article!   The author is so proud of himself!  He is so clever!  ("Communication" is now about pain-avoidance!)

By the way, I never explicitly mentioned that I am only talking about the lead sections here, I hope that's obvious.

It does seem to be that the majority of the criticism of your writing over the years has been criticizing your communication. In other words when you come here to the article itself you become a shitty writer, seemingly because you put your hobby-horses and pet projects up on a gilded pedestal, seemingly helpless to control yourself. Or put another way; you're writing is often completely unorganized and without direction and it seems out of context, it is haphazard and therefore indecipherable. There is no theme or goal. Like you have a pocket full of facts and you have to get rid of every one of them. ( In the lead! ) This is compounded by the stilted nature of your writing there which is quite painful to read. Damned near impossible, in fact. Nobody likes Eddie Haskell.

Added to that pain is your indiscriminate usage of jargon and blue-link rabbit holes rather than bothering your pretty little head with the hard work and effort to write some good meaningful explanations. Some might see laziness there. Or perhaps just a failure to understand the effort required in good writing. Like maybe good writing should come naturally to smart people like you, or something. —But a total lack of empathy or blinding self-centeredness could do it too.

Rather than attempting to get the big picture across, you amplify little tiny details that would normally be a dust bunny under the bed and spotlight it right up there with the rest of everything else that has real importance. And that dust bunny, your hobbyhorse, your windmill, —you will defend and insist on amplifying until your dying day. Consistently, predictably, and condescendingly from ex cathedra. Fixed, immobile, and certain. Does that sound like a scientific mind?

For example, just from your writing above in this thread, one of your hobby-horses appears to be "logic." Apparently (I'm wild-guessing) because somewhere you read an engaging debate between logic and methodology or some such. Somehow it got real big on your radar. Guess what guy? Everyman never saw that and could give a rat's ass about it. It has zip to do with his understanding the general concept of falsification. You seem utterly incapable of recognizing this, —or you just don't care. If it's big for you, you have to address it. That's what I mean by your personal knots that nobody else sees. It certainly does not deserve sentences in the lead, and certainly not in your above replies to me. (WTF??)  It looks like a compulsion from here, what else could explain it?

There seems to be another related problem here. You can't seem to empathize with Everyman's position. Why did he come here? Have you ever asked yourself that? Have you ever asked yourself what is the purpose of this article? What is its function? Or do you just see it as a place for you to play and gain self-gratification and stimulation, like mind candy? Because from here it looks like there's a whole bunch of too much self going on.

OK, Here is a main scenario why Everyman might come here. Like everyone else she knows that lie equals false equals bad. Then she starts to notice that all of a sudden falsify equals good. WTF? Then it happens on Wikipedia so she jumps into the blue rabbit hole and lands here. She has one simple goal, one simple objective. She wants to understand the bewildering sentence that has the word falsify in it. Why is falsify good? How can the falsification of Darwin's theory be good in the mind of this respected scientist she might be asking, not quite grasping this new terminology. Or, "Ghosts can't be falsified!?" WTF? she is asking. Yet "falsification" can be explained in three sentences, clarified with two more, and then two more for the purposes of heuristic over-learning. One paragraph or so. Sort of like Encyclopedia Britannica?

(By the way, one does not have to pay to read that article, it's too short.)

It's not like an experienced philosopher is going to come to an encyclopedia looking for advanced explanations regarding arcane philosophical minutiae, (especially not in the lead section).

If my comments seem baseless or too abstract I'm happy to follow up with clarifications or concrete examples and perhaps suggestions.

Again I apologize for my crude phone editing, I have no wiki tools, hints, nor preview etc. --Doug Bashford — Preceding unsigned comment added by 2607:FB91:7903:4D6D:D4EE:ABFF:FE1B:609B (talk) 02:32, 5 December 2021 (UTC)
 * This is a long rant against me and the article with little helpful content. I already knew that the logical nature of falsifiability is challenging. Now, I know that it troubles a lot this IP that signs under the name Doug Bashford. I am already looking for ways to help these people. After a  useless long rant of unhelpful comments about me, the IP explains his personal experience, which he wrongly attributes to every man. He says that he was told (before coming to Wikipedia) that "falsify equals good" and adds "WTF?" He wanted to see a sentence with "falsify" in it to see why "falsify is good". He then claims that the whole point can be explained in a short paragraph or so. I am happy to see his perspective. Obviously, he came to the article with a wrong expectation.  Falsifiability is not specifically for falsification: it is as much for a meaningful corroboration than for a falsification. Falsifiability supports both, because of its logical nature. The key point is that the value of falsifiability is not in the possibility of an actual falsification. If this was its value, then it would be a  failure, because rigorous falsification is impossible.  But falsifiability achieves perfectly its purpose, not more, not less. The IP understood correctly that its purpose is the possibility of tests (somewhere within these long rants, he wrote that). Yes, falsifiability is for testability, but with an understanding  that the outcomes of these tests is never certain. It's the sum of many tests and a lot of common sense that makes science useful. Falsifiability achieves very simply its goal, because of its logical nature, which is free from the problems of falsification.  This being said, I would agree that the article should be written with the goal of helping even those who come first with a wrong understanding and false expectations. I would say that the simple goal of testability should be mentioned earlier. This might help a lot. Dominic Mayers (talk) 03:57, 10 December 2021 (UTC)

arbitrary break
Dominic Mayers -- I'm sure you're well-intentioned in your own way, but you conspicuously lack the ability to explain things clearly to those who aren't professional philosophers, and trying to discuss things with you can be extremely frustrating (as when you first claimed that the Omphalos Hypothesis was falsifiable, but then later changed your mind, without admitting that you had changed your mind). The concept of "Falsifiability" has no practical value whatsoever unless it can be understood by scientists who are NOT professional philosophers, and you don't seem to be able to very usefully advance that goal... AnonMoos (talk) 23:19, 13 December 2021 (UTC)
 * Regarding the example of the Omphalos Hypothesis, I agree that it could be simplified. The key point is that this ad hoc hypothesis is not falsifiable, because there is no way to measure the claimed "actual" time of creation that is proposed by this hypothesis. My point when I discussed with you was that, if an had hoc hypothesis does not change at all the predictions of the original theory, then the theory remains falsifiable. Discussing that could have been useful in a context where we discuss weaknesses of falsifiability: it cannot distinguish two different theories with the exact same predictions, but it is confusing to do that in the context of the Omphalos Hypothesis, because it ignores that the ad hoc hypothesis also proposes  a different actual time of creation, which cannot be measured, though it can be known by reading religious texts. Moreover, the new theory is also not falsifiable, because it says that every thing occurs as we observe: by definition, it cannot be contradicted. However, these subtleties that confused me are not very fundamental in comparison with the logical nature of falsifiability. (The fact that I was confused at a time by an inherently difficult subject, only makes me an ordinary person, not a bad communicator.) It would not be fair to suggest that the logical nature of falsifiability is another subtlety of this kind. I feel like you bringing this at this time to build a case against me. Please do not make this a case against me as an indirect way to address the issue of the logical nature of falsifiability.  This is a logical fallacy: attacking the person instead of discussing the actual content.   Dominic Mayers (talk) 03:06, 14 December 2021 (UTC)


 * It's been 10 days since my last note in large part because I find this thread to be unpleasant. What makes it easier for me is that Mr Mayers at every step of the way verifies every claim I have made, but actually I was hoping he would prove me wrong and show that he was capable of writing with clarity. Just a careful re-reading of this thread alone emphasizes Mr Mayers' inability to communicate clearly and coherently.  Yet I do notice that the article has improved, but it still does not come close to meeting Wikipedia standards for all the reasons I have listed. I had at one time started to make some suggestions before I realized the obvious, —that would be pissing into the wind.  I'll just address a few things.
 * Mr. Mayers has insinuated that I and another writer have used the ad hominem fallacy against him. Not so much. Blunt talk is not an ad hom attack. Saying that you're fired because your writing is garbage is not an ad hom attack even if it may hurt your feelings. It attacks your writing.  An attack against the person only becomes a fallacy if that attack is used to divert attention away from the main topic.  Therefore saying you're fired because you're an obstinate, pugilistic, shitty writer is not an ad hom fallacy because mentioning your attitude and capability enunciates and focuses the nature of the problem. I have been toying with those ideas all along and they have finally firmed up. I think you need to be banned from Wikipedia.  I've spent a great deal of time explaining why.  The good news for you is I'm not a joiner and I'm not going to join Wikipedia just so I can get you removed. It's going to take a member to make that request and start the process moving.
 * Some people can come to Wikipedia and learn a great deal as they write an article, and everything gets better. But you come here and fuck it up for everybody as you learn, pretending to be Mr Smartypants as you spew hunches and pet entanglements as fact. And I don't think you can be rehabilitated, I don't think that's in your nature. I'm no psychiatrist but.... You are right out here in the open for everybody to see. If anybody missed it, just reread this thread carefully, it's right out in the open.
 * The world is full of fine professionals that cannot explain how to tie their shoes. And most of them are here trying to explain something and get published in Wikipedia where there is no expert evaluation.
 * And now you want to start a new Wikipedia article on a topic that is not in any real dictionary nor encyclopedia in the world!? I couldn't make this stuff up. --Doug Bashford
 * This time I could not find anything that could help improve the article, but I keep reading what you write, because I trust that you might have some good points regarding the subject and the content of the article. Dominic Mayers (talk) 13:30, 16 December 2021 (UTC)

The main problems with the content of this article appears to be the hiccups in your mind that only you are aware of. But I'll give you two broad tips, you'll have to figure it out and do the work yourself since you will fight anybody else.

The first is an aphorism that every good writer knows, which I will paraphrase: "I am so sorry I have written you such a terribly long letter, I would have written you a much shorter letter but I just didn't have the time nor the energy."

The second tip is actually a long essay which I think you should read and then reread again. You will probably need to modify it a little bit to fit your needs. In particular please pay attention to the paragraph that begins with this sentence: "Many Aspies show the "little professor" aspect described by Hans Asperger in regards to these speech issues through not only often going overly in-depth for the average audience on a given subject, but also in having developed the habit of pre-emptively explaining what they are trying to express (sometimes many times over in the same commentary on the same idea)...." https://tvtropes.org/pmwiki/pmwiki.php/UsefulNotes/AspergerSyndrome

(I discovered that article just today by googling something like: stupid but articulate.)

I doubt either of those tips will be of sufficient aid by itself but combined, and with sufficient effort, I could imagine a possibility.

I also have a question which could be useful if thrown into that Cauldron. Over the years you have insinuated: "I am not wrong, everybody else is wrong." My question is, is that because you are superior, or is it because everybody else is inferior? I think you would do well to ponder that question. --Doug Bashford.


 * I still did not find anything here about the subject, sources on the subject or any content that can be helpful to improve the article. Regarding your question, I don't find it interesting. I am not even sure what "superior" means in your mind. My question to you is what is the content that you would like to see in the article? What's your view on the subject that you feel is so important and not conveyed in the article, so important that you write these long rants. Perhaps that your criticism is that the article is complete garbage, so much, that you cannot even pick a part in particular that can be improved. This position is, of course, completely useless. Also, I am not interested in reading suggestions that are not about the subject.  Please, advance something about the subject and the article using sources. It would help people to see what's the real issue here. Perhaps you already said it: you would like to see a short article like the one in Britannica. I disagree and I would consider this as an attempt to suppress an important subject, which is falsifiability (the logical criterion).   Dominic Mayers (talk) 20:25, 16 December 2021 (UTC)
 * Communication is a two-way street. It involves 1. absorbing concepts from other people's mind into your own mind, and 2. it involves conveying concepts from your mind into the mind of another. There are two things going on there, and it seems that you can't do either. Watching a football game with your buddies requires absolutely no language other than grunts, hand waving, and the concepts: good, bad, beer and pizza.  And much of the rest of the day isn't a whole lot more complicated than that, because people already know what's on your mind for the most part. On the other hand here, effective and clear communication will require all of your powers.

You want feedback? Why would I give you feedback when every time I do, you say it's shit? Why would I repeat myself? The only thing that has changed in the article is now there is no longer a terrible "sea of blue." Everything else I said is just as true now as the day I wrote it.

You yourself have been kind enough to highlight the areas that you were incapable of understanding, those areas were communication failed you. Those areas you highlighted with words of scorn and disbelief. In places (Personal attack removed) such as in the area where I was talking about Everyman and her needs. All you could comprehend from me was a jumble when it comes to putting yourself into another's head and imagining another perspective. It appears you put absolutely zero effort and to trying to grasp what I was trying to convey. In a way was a little bit amusing, every place I said there was a problem you would come back and highlight it by doing the exact thing I was talking about. So if you want a place to work on, those are your clues. And again, what I already said in the last 2 weeks —reread it.

(Personal attack removed)


 * That's what you need to work on.
 * Sometimes you act like there's a certain secret thing I want you to say. Nope. Since you like repetition I'll say it one more time. I want a decent article. And I doubt you can do it. Puking up a tangled mess of words using proper grammar is not writing. And you seem blind to everything but your own little playpen.
 * For a more concrete suggestion, put the definition in the first one or two sentences. And clean up the photo caption. Like read the Wikipedia Manuel of Style. Study it.  And for God's sake stop talking like Eddie Haskell!  You do understand that he was a laughing stock don't you?  And a poor communicator? --Doug Bashford
 * This time, I decided not to read you. There might be some good points, but I will not read a long rant that focuses on me instead of on the content. Dominic Mayers (talk) 14:36, 20 December 2021 (UTC)
 * Well, I read the last paragraph, because it was just above, but nothing else. You lack of respect for people. Suggesting that people that suffer from autism cannot contribute to Wikipédia  is very rude.  By respect for them, I would not argue that I am not autistic, as if autistic people should be ashamed of that.  You would do much better to stick to the content. Dominic Mayers (talk) 15:55, 20 December 2021 (UTC)

RfC: Adding a challenging, counterintuitive but instructive and well sourced example in the lead
Should the lead of Falsifiability include the example "The brick fell upwards when released", which is opposed to a wide-spread expectation? Added point: The discussion naturally considered a similar challenge in the first sentence of the lead: it refers to observation statements instead of evidence that actually exist. In both cases, it's the purely logical aspect of falsifiability that is challenging. Dominic Mayers (talk) 19:20, 26 November 2021 (UTC)

Here is the current proposal for the lead in view of received comments:

Dominic Mayers (talk) 03:08, 13 December 2021 (UTC)

I closed the RfC with the informal conclusion that the example is acceptable in the lead, but the lead seems challenging. I will change the lead and do another RfC. Dominic Mayers (talk) 05:49, 14 December 2021 (UTC)

Discussions
When people visit a Wikipedia article on mathematics or physics, they expect that some challenging subtleties cannot be avoided. A common misunderstanding, even seen in popular accounts on the subject, is that there is no intrinsic challenging subtleties of this kind in falsifiability. In this naive view, falsifiability is simply that the theory can be (potentially) proven false by an observation and other subtleties are fancy details. But there exists a challenging subtlety in falsifiability that is not a fancy detail. The first paragraph convey this key concept pretty well. It explains unambiguously, as much as it is possible without examples, the fact that falsifiability is a logical criterion. The question is whether we should add a concrete example to illustrate this concept and at the same time clearly bring out this intrinsic challenging subtlety. Here is an idea of what it could be: This might be challenging for some readers, because it is opposed to the incorrect but wide-spread expectation that falsifiability cannot be about imaginary events of this kind. Yet, it only illustrates the basic and central concept of falsifiability with an example taken from a well known introductory book on the subject. Is there a better way to use this example in the lead? Should we use it at all? Dominic Mayers (talk) 23:27, 29 October 2021 (UTC)


 * Yes, after the sentence about "an imaginary state of affairs", why not add something like: For example, "The brick fell upwards when released" falsifies Newton's law of gravitation. Biogeographist (talk) 18:32, 21 November 2021 (UTC)
 * I do that now. This is not a polemic RfC, so I don't think there is an issue that I do not wait until it is closed. The article can be modified again in view of additional comments. Dominic Mayers (talk) 18:37, 21 November 2021 (UTC)
 * Most RfCs are not formally closed, especially when consensus is clear.VR talk 16:24, 23 November 2021 (UTC)
 * I know, but the point of an RfC is not to close it after the first comment. So, I would let it go for a while before closing it. Dominic Mayers (talk) 16:45, 23 November 2021 (UTC)

I am going to close the RfC with the conclusion that the example is acceptable in the lead, because there were supports and no objections. This will allow me to change the lead and possibly do another RfC specifically about the lead. Dominic Mayers (talk) 03:53, 13 December 2021 (UTC)
 * In principle this is fine, but the example paragraph is more detailed and intricate than necessary for the lead. In particular, it's extremely important to be understandable to the broadest possible cross-section of readers, all the more so in the opening paragraphs which are often the only thing that people read; this is something that the article seems to struggle with and is also reflected in the proposed text. The challenge is to maximize the amount of information that readers retain when they are merely opening the page and glancing over the top of the article for a few seconds. Furthermore, on this article, many readers are in high school (or younger) and will not be able to follow the description without implausible dedication. I've attempted to address some of this in the current lead (as the issue is not limited to the text in question), but a complete fix would require a considerably greater amount of time and effort. Sunrise (talk) 10:25, 25 November 2021 (UTC)
 * This illustrates the very point that is the purpose of the RfC. A text that is inaccurate, less informative, even ORish can be acceptable to many readers when it does not conflict with what they expect. They might retain more information from such a text than from a text that represents correctly what sources say globally. This is the challenge that the current article (before any edits done after the RfC started) struggle with, by trying to be this text that represents correctly what sources say globally, not any text that some readers will be more comfortable with. I created the RfC because I accept that some compromise might be needed. However, I don't see that the new version that created is this compromise. On a technical side, I don't think it was correct for me or for Sunrise to modify the article in the middle of an RfC. I will revert the article to its original version at the time the RfC started. This is not in itself a judgment on these modifications and the corresponding proposed new version. Dominic Mayers (talk) 13:54, 25 November 2021 (UTC)
 * Full response below, but I will note that suggesting that this has to do with making the article inaccurate or less informative is a significant misunderstanding of my point. Sunrise (talk) 04:37, 26 November 2021 (UTC)
 * More details below and later, but I wish to say now that what I wrote does not imply that I believe Sunrise has bad intentions or is incompetent. As I wrote in my initial comment to the RfC, some people sincerely do not expect intrinsic subtleties in falsifiability that fundamentally requires concepts extracted from mathematical logic. They think that, if we make use of that kind of concepts, then we are making things unnecessarily too complicated, when in fact it's intrinsic to the essential nature of falsifiability. More importantly, among these people, there are excellent mathematicians that know very well mathematical logic. So, it's not because of ignorance of mathematical logic or mathematical incompetence or, in general, lack of academic intelligence that people ignore the fundamental role that these logical concepts play in falsifiability. It's because they have different expectations, those of Lakatos's falsificationism, which have little to do with the purpose of falsifiability as purely logical criterion. So, they look at falsifiability in terms of these other objectives and this lead to a conflation with Lakatos' falsificationism, in which the separation between logic and methodology is not so fundamental any more. Even excellent mathematicians did that, including Lakatos. The position that one takes here (including a possible conflation with Lakatos' falsificationism) depends a lot more on one's fundamental philosophical position than on one's academic intelligence.  It's important that this position and the corresponding conflation is not expressed in Wikipedia's voice. We must keep in mind that the subject of this article is Popper's falsifiability or simply falsifiability (in philosophy) and that we do not want to conflate two views in Wikipedia's voice and the subject of the article is not (Lakatos') falsificationism and its problems. Dominic Mayers (talk) 10:30, 26 November 2021 (UTC)
 * The exceptional level of jargon and complex vocabulary makes the lead unreadable to someone not already versed in the concept, which is a letdown to our readers, as an average 11-year-old should be able to understand the basic concept. To describe it in a way that an 11-year-old could understand would not involve presenting false or misleading information, but just avoidance of jargon and better writing. For instance, A statement is falsifiable if there is a piece of evidence that would show the statement is incorrect. For instance, if Alice says that "all swans are white" then Bob can prove her wrong by showing her a black swan—so, Alice's statement is falsifiable. ... Gravity is a sensible example to use as something that is falsifiable, along with one like the swans, as a clear demonstration that "falsifiable" does not imply "false" (or "true"). But the problems with the lead will be solved neither by the inclusion nor the exclusion of this example. — Bilorv ( talk ) 00:35, 13 December 2021 (UTC)
 * , I understand that I am told repeatedly that the lead is challenging and I intend to make it less challenging, but I need help from people that understand falsifiability. Please show that you understand the logical nature of falsifiability and why it is fundamental. First, understand that falsifiability is not simply the possibility to be proven false by a piece of evidence. Not only it is not that, but the main disastrous confusion on the subject is that people conflate falsifiability with this other concept. This other concept is falsifiability + the extra requirement that a falsification can occur using a piece of evidence, but it is fundamental that this extra requirement is not there at all. It might be chocking that the lead uses a language that insists on that, but it's important, because otherwise it is not falsifiability.  Falsifiability is entirely at the logical level. It only exists in a very limited world of statements and their relationship with no contact with the real world, except through an empirical interpretation of the language. It's the only way, the statement "the brick fell upward when released" can show gravity to be falsifiable. In the real world, "the brick fell upward when released" proves nothing. People will find tuns of explanations to reject this falsification. For example, they will ask how do we know that there is no invisible strings? This is a valid issue in the real world, but not an issue at all for falsifiability, because the only thing that matters for falsifiability is that formally  (i.e., purely at the level of statements) the law is contradicted by an observation statement. The notion of invisible strings cannot even be expressed in the limited formal world of statements. This is not me that insist on that. It's Popper himself that kept repeating that. Also, the importance of this is explained in secondary sources.  Dominic Mayers (talk) 02:12, 13 December 2021 (UTC)
 * I want to emphasis that I am not arguing against change. I am happy to accept that the lead should be less challenging. I am only saying that it's not helpful if people propose sentences that conflate falsifiability with the ability to be possibly falsified by an experiment as in Lakatos' falsificationism. The subject of this article is not Lakatos' falsificationism. Dominic Mayers (talk) 02:24, 13 December 2021 (UTC)
 * I just added a proposal for a new lead. However, it might still be challenging, because people come here with false expectations. They expect statements such as A statement is falsifiable if there is a piece of evidence that would show the statement is incorrect and no subtleties regarding imaginary state of affairs, etc. With this kind of expectations, it's difficult to please them without totally failing to introduce falsifiability. Dominic Mayers (talk) 03:18, 13 December 2021 (UTC)
 * I don't know what you want me to demonstrate understanding of, since I quite clearly said in my statement that falsifiability is independent of truth value. I don't often invoke my degree on Wikipedia, but I do think it suffices here to say that I got a first-class mark on my Logic course at Oxford. We can have a conversation about the formal negation of a statement and how $$\neg (\forall x. (x\ \text{is a swan}\rightarrow x\ \text{is white}) \equiv \exists x.(x\ \text{is a swan} \land \neg (x\ \text{is white})$$ if you want. The problem, however, is that I am not your target audience. The common language definition I gave is perfectly valid (pun intended), and more accessible to a general audience. — Bilorv ( talk ) 17:15, 13 December 2021 (UTC)
 * Thanks to mention your degree and your understanding of logic, but even experts in logic can fail to appreciate the importance of the logical nature of falsifiability. The best example is Lakatos. He was one of the greatest logicians of his time and yet, as clearly illustrated by his writing and Popper response to them, he did not appreciate this distinction. Popper wrote that "an entire literature rests on the failure [by excellent mathematicians and philosophers, especially Lakatos] to observe this distinction". The term "failure to observe" here does not mean at all "mathematical incapacity to understand". In Popper's philosophy, rationality exists in the logical side, but it is also embedded in a complex methodological process, which we know is rational, because, of course, science is rational and it uses a lot of rational tools, statistical tests, etc. but cannot be turned into a logical system of inference. (This is what Popper means when he says that the methodological side is hardly rigorous—he does not mean that it's not rational or that it does not use mathematics.) Lakatos and many others have tried to make it rigorous into a formal system, because it's hard to accept that this rationality in science cannot be (or at the least has not been) formalized. In this manner, the lovers and experts of logic who would want science to be explained by it are the most likely to not appreciate the distinction, because the distinction separates these two sides: the logical side and the methodological side. This distinction is very very fundamental in Popper's philosophy and it is fundamental to understand that falsifiability is on the logical side, without any actual interpretation in the world of theory-ladden practical observations—only the rules of interpretation are fixed. (Here, for this last point, your knowledge of logic and the role of interpretation in logic to give a semantic to the formal calculus can be useful.)   Dominic Mayers (talk) 17:56, 13 December 2021 (UTC)

Sunrise's proposal
proposed this new version. There are so many modifications that I cannot comment on them all. It would be preferable that Sunrise breaks down his modifications in few important parts that can be discussed here. I am sure there is many good points to extract from these modifications, but it seems that the main premises behind them are ORIsh in the sense that they conflate criticism of Lakatos's falsificationism (by Lakatos himself and others) with criticism of Popper's falsifiability. Of course, for Lakatos and many others who follow his view, there is no conflation here, but this view cannot be presented in Wikipedia's voice. For example, the sentence "A contradictory observation does not necessarily mean that the original theory or hypothesis is false, because of the possibility of experimental error." clearly conflates falsifiability with Lakatos's falsificationism and its problems and completely fails to illustrate the key concept of falsifiability. , since you already commented, your view on Sunrise's proposal here will be appreciated. Dominic Mayers (talk) 14:14, 25 November 2021 (UTC)


 * I won't comment to all the changes, but I think "In the philosophy of science, a theory or hypothesis is falsifiable (or refutable) if it can be contradicted by evidence." is far preferable as an opening sentence. It seems far more clear for a general audience. I will also say that the sentence you take issue with seems to be reliably sourced to the SEP article on Karl Popper and the quoted text in the citation seems nothing to do with Lakatos, so I'm not sure how this can be construed as OR that conflates the views of Popper with those of Lakatos. In fact, this seems to be about the distinction between the logic of scientific discovery and methodology, which is already noted in the current lead: "as pointed out by Duhem and others, definitive experimental falsifications are impossible. However, Popper insisted that falsifiability is a logical (thus, not methodological) criterion and, therefore, does not have these problems and is sufficient to make these mathematical tools applicable within a critical scientific discussion."
 * As to the RfC, I think that having an example could make things clearer, but it should be made clear by the wording that the sentence is a potential falsifier. So either "is potentially falsified" or "would be falsified" might make things less confusing for a general reader than "is falsified". Alduin2000 (talk) 16:20, 25 November 2021 (UTC)
 * , focusing on the first sentence can make us lose the big picture. However, considering a sentence can be useful to make the discussion concrete while trying to see the big picture. Let me consider the question of whether "is potentially falsified" or "would be falsified" might make things less confusing for a general reader than "is falsified". This question is central to the big picture and the distinction between logic and methodology that is so much emphasized by Popper and explained in many reliable sources including SEP. If it's understood that falsifiability is a logical criterion, then "is falsified" is very natural, not confusing at all. More importantly, in that case, using "is potentially falsified" or "would be falsified" is actually confusing, because "would be" or "potentially" makes no sense in that case. In logic, it's very simple: it's a contradiction or it's not—no room for "would be" or "potentially". Sure, if it is not understood that falsifiability is a logical criterion, then the naive view is superficially more pleasing, only superficially, because it leads to a problematic conflation of Popper's falsifiability with Lakatos' falsificationism—it creates this conflation very clearly. Therefore, the goal must be to make it easy to understand for a large audience that falsifiability is a logical criterion.  If I started the RfC, it means that I am open to discussions about how this can be achieved. However, I don't think that globally the modifications proposed by Sunrise achieve that purpose and here I don't focus on the first sentence. For example, the following sentence, which you support,  was removed by Sunrise. This is only an example. He also moved Lakatos criticism of his own falsificationism in the Controversy section, arguing that it was a criticism of Popper's falsifiability. It's totally clear that the premises behind Sunrise's modifications reflect a conflation of Lakatos's falsificationism (that he and followers attribute to Popper) with Popper's falsifiability, which is a logical criterion that achieves a completely different purpose than Lakatos' falsificationism.  Dominic Mayers (talk) 17:46, 25 November 2021 (UTC)
 * I just want to respond to make sure I'm a bit clearer about my previous comment. I'm not necessarily opposed to using even the wording "is falsified" so long as it is made clear that no such observation has to actually be made. I suppose that the current lead already does that ("The contradictory observation statement, also called a potential falsifier, can correspond to an imaginary state of affairs: the corresponding (past, present or future) falsification may not be the case.") but I think the wording could maybe be made clearer. The only worry I have is wording and not to the substance of your proposal; and therefore I only mean to provide some suggested changes to the wording. For example, "he contradictory observation statement, also called a potential falsifier, can correspond to an imaginary state of affairs: the corresponding (past, present or future) falsification need not actually occur for the theory to be falsifiable" I think is a little better whilst retaining the same meaning. I have no strong opinions on all of Sunrise's changes, but I do think their proposed opening sentence is clearer. I will stay neutral on their other changes without having checked on them properly; hopefully Sunrise or Biogeographist can speak to them. But I'm not sure that that is connected to the RfC entirely, which I support in principle. Alduin2000 (talk) 12:16, 26 November 2021 (UTC)


 * The changes I made are not primarily related to the RfC question; I've redone the one change that concerns content outside the lead, in a way that should address your comment about it (see the edit summary for details), but have left the rest alone. I should really ask you to specify which ones you actually object to, but the principles on which my changes are based are described in my comment and my edit summaries. Please don't take this the wrong way, but my infrequent watching of this talk page over the last few years suggests that you are very difficult to have a productive discussion with, in particular due to a tendency to meet objections with argumentum ad nauseam. As such, I do not wish to comment here at length.
 * With regards to conflation, I think you are assuming that the logical framework is the only valid way to describe falsifiability, which to me is a neutrality issue, because its actual use within science is centered on whether particular hypotheses are either falsified or not falsified at any point in time. Another way to think of this is that there is little relevant difference in information content between Falsifiability is the logical possibility that... and Falsification is what happens when... but the second is expressed in concrete terms and is more likely for a reader to be able to understand. This concrete framing, which you seem to think is simply incorrect (though I disagree), is also much more practically relevant to readers.
 * For instance, this idea is the source of A contradictory observation does not necessarily mean that the original theory or hypothesis is false, because of the possibility of experimental error, which is intended as a straightforward paraphrase. It contains essentially the same relevant information as the sentence This existent contradiction with an observation statement does not imply at all that the theory is methodologically, i.e., in practice, false, with the phrase about experimental error being derived from the note. Furthermore, the original version is phrased in an unnecessarily complex manner, and thus requires an unnecessarily high level of English-language comprehension that we cannot assume from our readers, a large proportion of whom do not have English as their first language. My rephrasing is itself more complex than it ideally should be, but at minimum it is better than the original.
 * Since the Duhem statement came up in the followup comments, I will address that briefly as well. I don't object to expressing the idea in general, but this version is uncited, and furthermore uses WP:W2W language like "pointed out" and "insisted" which are often indicators of bias. As the central idea was still mentioned in the lead, if not under that name (the statement about experimental error), I decided it was best to simply remove it pending a better description. An additional issue is whether we should imply the Duhem-Quine argument(s) to be factual or "correct" in the first place, when the actual main article about the topic does not. Sunrise (talk) 04:37, 26 November 2021 (UTC)
 * , I will respond in more detail later, but you should not take your personal view that your modifications are not primarily related to the RfC question to justify doing major modifications during the RfC. On the contrary, they are deeply related to the RfC question. I am not judging the modifications. They can be fundamentally related to the RfC (as they are) without necessarily being fundamentally bad. So, in a neutral manner, I suggest that to respect the RfC and avoid making this exception to a good rule, you revert your last modifications. But, we can discuss them. Dominic Mayers (talk) 10:43, 26 November 2021 (UTC)
 * , you wrote I have no strong opinions on all of Sunrise's changes, but I do think their proposed opening sentence is clearer. This contradicts The only worry I have is wording and not to the substance of your proposal. The substance of my proposal is that we should not use a wording that is confusing for those who see correctly the logical nature of falsifiability. The use of "potentially" or "would be" can only support a confusion. As a logical criterion, falsifiability is straightforward and simple: it's contradicted (out of time) by an observation statement. (The falsifier must refer to an event in time, but the contradiction itself is out of time and simply is there, not potentially or whatever.)   Yes, the falsifier corresponds to something that might happen or not and I agree that this second part should be made clear, possibly in the same sentence, but not at the expense of the logical nature of falsifiability, because it results in a conflation with Lakatos's falsificationism that has created such a confusion in the literature. We should not encourage that. We can present both views, but one is not the subject of this article and the only purpose for including it is the context that it provides, a background so that we explain falsifiability in a context. Presenting it as an opposite view in an implicit manner as if falsifiability had the same objective is misleading. That's the way Lakatos and followers would see it, but we cannot adopt this view to decide how to organize the article. It would be like including this view in Wikipedia's voice, implicitly through the organization of the article. I don't think it is a coincidence that those who favour "would be" and "potentially" are also those who would like to present Lakatos' view as a criticism of falsifiability as if it had the same objective as (Lakatos') falsificationism and failed. Falsifiability has almost nothing to do with falsificationism, except as an historical context. It has not at all the objective of falsificationism.  Dominic Mayers (talk) 13:03, 26 November 2021 (UTC)
 * I do not believe there is any contradiction; the wording of the RfC simply asks whether "The brick fell upwards when released" should be included in the lead as an example, and that is what I was referring to when I said "your proposal". I take issue to you saying I don't think it is a coincidence that those who favour "would be" and "potentially" are also those who would like to present Lakatos' view as a criticism of falsifiability as if it had the same objective as (Lakatos') falsificationism and failed. I made no comment as to how Lakatos' views should be presented in the article and I am the only one who can be construed as "supporting" the wordings you mention as nobody other than you and I have commented on them. Please do not make insinuations of bad faith. I would also appreciate it if you engaged a little less argumentatively; the whole point of Wikipedia as a project is that we as volunteers should engage cooperatively. I probably will not engage further here as I'm not sure there is much more I can contribute to the conversation. I support Sunrise's opening sentence for the lead particularly, and I support the RfC as worded (i.e. the addition of an example to the lead), but I think that clarity for non-specialist audiences should be taken into account when further developing the exact wording. I remain neutral as to Sunrise's other changes, but I will note that despite accusations of OR and conflation, no concrete examples have been given, apart from a sentence that appears to be well supported by a reliable secondary source. I suggest that perhaps discussion of Sunrise's changes to the article be moved to its own section so it doesn't muddy up the RfC discussion. Of course if other editors think it is relevant to the RfC then it should be kept here; personally I think otherwise. Thanks. Alduin2000 (talk) 14:39, 26 November 2021 (UTC) (Edited to strikethrough parts of my comment after communication on my talk page. 18:09, 26 November 2021 (UTC))
 * Most of it are accusations of not assuming good faith. I am not responding to that, because it's not the place. In this talk page, we must keep the focus on the article. There are important issues to discuss and understand. Let's focus on that here. Dominic Mayers (talk) 15:51, 26 November 2021 (UTC)
 * , wrote I support Sunrise's opening sentence for the lead particularly. Generally speaking, a RfC is not a vote, so discussion is what matters. You provided no rational for this support, except your impression that it is clearer, but did not respond to my point that it actually contradicts the logical nature of falsifiability. Dominic Mayers (talk) 18:55, 26 November 2021 (UTC)

Link with the RfC
People coming for the RfC might not immediately see how the RfC is related with the use of "potentially" or "would be" or with the role of falsificationism in the article. The whole discussion here turns around whether we should hide or minimize the challenging purely logical aspect of falsifiability, which is what the RfC is about. Without assuming any intention, "potentially" and "would be" correspond to hiding the logical aspect of falsifiability. Without assuming any intention, the proposed modifications attach to falsifiability a purpose that is directly in terms of actual falsifications (as in falsificationism) and thus ignore the logical nature of falsifiability or minimize its importance. Without assuming any intention, in Sunrise's proposal the example that illustrates the logical nature of the criterion is kept, but in a disconnected paragraph that seems to play a secondary role. In his proposal, instead, what appears first is a sentence where the logical nature of falsifiability is rejected. It is rejected by the use of "potentially" or "would be" or whatever else is said to be "easier to understand", but actually hides the challenging purely logical aspect of falsifiability. Dominic Mayers (talk) 14:54, 26 November 2021 (UTC)

Proposals for a first sentence
This proposal uses a suggestion made by for a sentence later in the lead: it makes clear that the falsifier needs not actually occur. In contrast with a statement that would use "would be" or "potentially" or "in principle" or "can be" to qualify "contradicted", this statement does not mislead people away from a correct understanding. It is perhaps challenging for people who expect something more usual, say in terms of concrete evidence. However, I don't think it is wise to mislead people in the first sentence and then try to correct this confusion after. For concreteness, here is the opening proposed by Sunrise: I ask the following questions: Does "evidence" is the proper term to use to refer to an imaginary state of affairs that might happen to actually occur (though it's irrelevant in the definition), but typically does not? (My answer is clearly NO.) Is the second sentence sufficient to remove the confusion and make clear that falsifiability is entirely a logical criterion, i.e., with no need at all for the falsifier to actually occur? (My answer is definitively NO) Most likely, the second sentence will be interpreted in view of the usual meaning of evidence. The whole thing is entirely misleading. Even if the answer to this last question was Yes, it's not, but even if it was Yes, what would be the rational for the additional complexity? Dominic Mayers (talk) 17:44, 26 November 2021 (UTC) A compromise would be to have something more neutral as a first sentence that do not attempt to immediately define falsifiability. Something like Dominic Mayers (talk) 23:41, 2 December 2021 (UTC)