Talk:Scientific method

Thought experiments

"the truth consists of hard-to-vary assertions about reality is the most important fact about the physical world" --David Deutsch#Invariants is already cited in the article.

If a thought experiment were to become the basis of a science, it would need to be invariant. That's a high barrier, compared to a real experiment. -- Ancheta Wis   (talk | contribs) 15:57, 14 October 2023 (UTC)

Thought experiments are important, especially in physics. I don't say they all what's required to pass as a theory, but they are the beginning of a theory. E.g. Newton knew the speed required to bring an artificial satellite in orbit. tgeorgescu (talk) 16:15, 14 October 2023 (UTC)

Tricky thing is that just as there's no science as to what counts as science (see the Demarcation problem), there's no science as to what counts as the scientific method — which is really just so much more of an attitude of rationality and openness to evidence that we just "know it when we see it". Otherwise the articles on science and its "method" would be easy to hammer out — things would either be correct or incorrect, not debatable. The consensus for school children does not mention mind experiments explicitly as valid. But the consensus among science philosophers and historians is obvious: Einstein isn't the only one who figured out quite a bit in his head long before anything was proven on paper (let alone in the physical world). For example I think it wasn't until the 1960s or 70s that we could finally do our first experiments to validate relativistic time by flying some atomic clocks around for a few hours and then comparing them. Clearly, the work was already considered real science long before the experiment was conducted. Destrylevigriffith (talk) 16:44, 14 October 2023 (UTC)

The scientific method is for establishing new knowledge (to the extent that predictions are possible in the new, unknown domain). In a nutshell, that's the demarcation problem. If the prediction/explanation has to change to match observation, the explanation fails (is shown to be in error), as stated in the article. Conjecture/ hypothesis is welcome, of course, but has to be tagged as such. When a scientist changes his explanation to match the observation, this is tampering. The article explains this point (e.g. researchers are to sign and date their contribution as part of their protocol; have separate teams for explanation versus experimentation, etc.). We don't yet know how to do this for an experiment conducted entirely in one's mind. There is room for physical, communicable experiments, not involving neologisms. Digital twinning is far from thought experiment, as yet.

One thousand years ago, Alhacen (Ibn al-Haytham) made your point, that scientists are not free from error, and proved several points (in optics and in space science) to be in error, using experiment (it's in the article). So an open question in science has to be recognized as such; the science at that point remains a model, only a concept. --Ancheta Wis   (talk | contribs) 18:49, 14 October 2023 (UTC)

My point: thought experiments are a tool in the scientific method. They are of course not the only tool. Evidence is needed to test their assumptions. tgeorgescu (talk) 22:05, 14 October 2023 (UTC)

Stemwedel quote

for editor convenience, I have retrospectively prefixed here the quotation whose addition is being challenged under WP:BRD:

The big difference Popper identifies between science and pseudo-science is a difference in attitude. While a pseudo-science is set up to look for evidence that supports its claims, Popper says, science is set up to challenge its claims and look for evidence that might prove it false. In other words, pseudo-science seeks confirmations and science seeks falsifications.

— Janet D. Stemwedel, Scientific American

Discussion

Binksternet left a message on my talk page that said: Please do not add commentary, your own point of view, or your own personal analysis to Wikipedia articles, as you did to Scientific method. Doing so violates Wikipedia's neutral point of view policy and breaches the formal tone expected in an encyclopedia. Binksternet's response is puzzling, as it refers to this edit in which I did not add anything to the article—there is no "commentary" or "personal analysis" to be found!

As for my justification of the removal of the Stemwedel quote from the beginning of the "Overview" section: I said in the edit summary: science may also seek confirmations—see, e.g., the footnotes in Critical rationalism#Variations, such as "Producing evidence" (Bunge 1983). I will quote the footnote on Mario Bunge to which I referred:

• Bunge, Mario (1983). "Producing evidence". Epistemology & methodology II: understanding the world. Treatise on basic philosophy. Vol. 6. Dordrecht; Boston: D. Reidel. pp. 59–113 (70). doi:10.1007/978-94-015-6921-7_2. ISBN 902771634X. OCLC 9759870. Critical rationalism (e.g. Popper, 1959) agrees that experience is a test of theories (its only concern) but claims that only negative evidence counts (against), for positive evidence is too easy to come by. True, unsuccessful attempts to refute a theory (or discredit a proposal or an artifact) are more valuable than mere empirical confirmation. However, (a) the most general theories are not refutable, although they are indirectly confirmable by turning them into specific theories upon adjoining them specific hypotheses (Bunge, 1973b); (b) true (or approximately true) predictions are not that cheap, as shown by the predictive barrenness of pseudoscience; (c) positive evidence for the truth of an idea or the efficiency of a proposal, procedure, or artifact, does count: thus the US Food and Drug Administration will rightly demand positive evidence for the efficiency [efficacy] of a drug before permitting its marketing.

Earlier Bunge had argued:

• Bunge, Mario (1973). "Testability today". Method, model, and matter. Synthese library. Vol. 44. Dordrecht; Boston: Reidel. pp. 27–43. doi:10.1007/978-94-010-2519-5_2. ISBN 9027702527. OCLC 613670. Quite apart from their historical merits, is either of the two criteria [of science—confirmability versus falsifiability] actually satisfied by today's science? Have they withstood four decades of momentous advances in pure and applied science? This is the problem of the present investigation. The outcome of it will be negative: neither confirmability nor refutability is either necessary or sufficient for every single component of science. Nor will any other single trait do: science is too complex an object to be characterizable by a single property.

These quotes from Bunge refute the claim in the Stemwedel quote that "pseudo-science seeks confirmations and science seeks falsifications", which is a far too simplistic characterization of science, as the quotes from Bunge indicate. As I said in another edit summary reverting the addition of the same quote to another article, the Stemwedel quote seems to conflate falsifiability as a demarcation criterion with falsifications—a hypothesis can be falsifiable and yet still be confirmed/corroborated, and it's fine for scientists to seek confirmations/corroborations as long as they use sufficiently severe tests.

Even critical rationalist philosopher John W. N. Watkins emphasized that science involves confirmations/corroborations:

• Watkins, John W. N. (December 1995). "Book review of: Critical rationalism: a restatement and defence, by David Miller". British Journal for the Philosophy of Science. 46 (4): 610–616. doi:10.1093/bjps/46.4.610. JSTOR 687902. (2) A main component of Popper's methodology was his theory of corroboration (see the concluding chapter and appendix *ix of his [1959]); corroborations are what ultimately govern the rational acceptance of theories. This disappears without trace in Miller's 'restatement'. In his index there are six entries against 'corroboration', five of which refer to places where an author is being quoted or reported. The sixth comes in the course of an examination of my [1984]. I had tried to give a fresh answer to the question, 'Why do corroborations matter?' Miller writes: 'The answer is that corroboration doesn't matter' (p. 120). (3) Popper had a horror of anything like rationality-scepticism. He insisted that a theory's being currently the best corroborated, while not justifying the theory, does justify a preference for it over its rivals. He may not have kept these two kinds of justification as separate as he should have done, but his philosophy allows there to be sufficient reasons for accepting one theory as better than its rivals at the present time.

Or if those quotes aren't sufficient, take Graham Oddie's summary of why falsification without confirmation/corroboration is insufficient:

• Oddie, Graham (2001). "Truth, verification, verisimilitude, and evidence: philosophical aspects". In Smelser, Neil J.; Baltes, Paul B. (eds.). International encyclopedia of the social & behavioral sciences. Vol. 23. Amsterdam; New York: Elsevier. pp. 15932–15937. doi:10.1016/B0-08-043076-7/01014-7. ISBN 9780080430768. OCLC 47869490. Pseudoscientific theories (Popper's examples were psychoanalysis and astrology) are replete with 'confirming instances.' Everything that happens appears to confirm them only because they rule nothing out. Popper argued that genuine theories must forever remain conjectures. But science is still rational, because we can submit falsifiable theories to severe tests. If they fail, we weed them out. If they survive? Then we submit them to more tests, until they too fail, as they almost assuredly will. There are three serious problems with falsificationism. First, it does not account for the apparent epistemic value of confirmations. The hardline falsificationist must maintain that the appearance is an illusion. Second, it cannot explain why it is rational to act on the unrefuted theories. Confidence born of experimental success reeks of inductivism. Third, pessimism about the enterprise of science seems obligatory. Although truth is the goal of inquiry, the best we can manage is to pronounce a refuted theory false.

In summary, the Stemwedel quote is not a good summary of scientific method or of demarcation of science from pseudoscience. That is why I removed the quote. Biogeographist (talk) 23:25, 2 February 2024 (UTC)

"Simplistic" material has its purpose. Popper made his simplistic pronouncement for a purpose. Why can we not keep the Popper quote followed by defining context from Oddie and Watkins? We can excite the black-or-white reader and the shades-of-grey reader at the same time. Binksternet (talk) 23:34, 2 February 2024 (UTC)

It is a good quote taken from a well known scientific journal. However, it gives the impression that there is no debate about what distinguishes science from non science. Clearly, our personal perspective is irrelevant, but since Biogeographist gave his view, I will add mine: Popper found a good criterion when taken in its proper context, but clearly one cannot reduce science to falsifiability and I don't think Popper could have reasonably claimed that out of a proper context. This is in fact a key point. Every thing we say, always depends on a context. Nothing is universally true. This is something known since the ancient Greece: every truth depends on a previous truth. This also applies to the falsifiability criterion. Dominic Mayers (talk) 00:46, 3 February 2024 (UTC)

The reason that I added that quote is because the article as it stands is very dense and almost legalistic. The quotation is clear, concise and accessible to non-specialists. It is entirely opposite. Biogeographist's response merely confirms my view of the need for such a clear statement because it is written in the same legalistic tone as the article: it is not wrong but it is not helpful. --𝕁𝕄𝔽 (talk) 01:04, 3 February 2024 (UTC)

I agree 100% with Dominic Mayers about the importance of context and about the high value of falsifiability within a proper context, and I'm certainly not opposed to JMF's goal of making the article more readable, but I don't think the Stemwedel quotation helps, for the reasons already stated. Perhaps you can take inspiration from Stemwedel's article and from this discussion to craft a new edit that meets your goals while avoiding these objections. Biogeographist (talk) 02:08, 3 February 2024 (UTC)

By the way, it is mistaken that Stemwedel's article is from a "scientific journal" as Dominic said; it is really a blog post. The blogs at https://blogs.scientificamerican.com are mostly not published in the magazine. I don't see anything wrong with the blog post as far as it goes, but as Stemwedel said, she's just explaining Popper's demarcation criterion, not claiming that it's the only criterion or that there is no debate about demarcation. In fact, she wrote: "before we get into how real science (and real non-science) might depart from Sir Karl's image of things, I think it's important to look more closely at the distinction he's trying to draw." But she never comes back to "how real science (and real non-science) might depart from Sir Karl's image of things". An interesting article that was published ten years after Stemwedel's blog post and that you may want to read for more recent views on the demarcation problem is: Hirvonen, Ilmari; Karisto, Janne (June 2022). "Demarcation without dogmas". Theoria. 88 (3): 701–720. doi:10.1111/theo.12395. Biogeographist (talk) 03:38, 3 February 2024 (UTC)

This last paper only confirms that there are still debates about what distinguishes science from non science. Also, any review paper in philosophy is still only representing a point of view, not the truth about the current state of the art in philosophy, and must be treated as such in Wikipedia. Dominic Mayers (talk) 13:19, 3 February 2024 (UTC)

That supports my view that Biogeographist is being excessively purist about this. Here, and especially in the pseudo-science article, we need a clear, concise and accessible statement of what makes real science different from nonsense. If a reader follows a link in a general article, they should not be confronted with a philosophical treatise. The advantage of using the Stemwedel quote is that it provides just such a statement but because it is a quote, we don't have to qualify it extensively as we would if it were in Wikivoice. In all honesty, I can't help feeling that Biogeographist's concerns, while valid, are about edge cases. When the topic is something clear-cut like crackpot cures for CV19, then Popper's exposition is very broadly representative of mainstream scientific consensus going back to Bacon. WP:Think of the reader who is likely to be someone who has little or no scientific education. We don't have to impress professional scientists who won't be reading this article in any search for knowledge. --𝕁𝕄𝔽 (talk) 13:53, 3 February 2024 (UTC)

I just want to insist that Popper always spoke in a context. He often criticized people, because they took his statements out of context. In particular, even the falsifiability criterion have to be understood within science as a problem solving activity. When Popper says that falsifiability is sufficient and necessary, it is within that context. In other words, Popper has never claimed that science does not have the positive goal of solving problems. The idea that Popper would have claimed that the goal of science is not to solve problems, but only to make falsifiable statements, is ridiculous. Yet, there are some to criticize Popper as if he had made that claim. I say that because the quote could lead to such a misunderstanding when it concludes "pseudo-science seeks confirmations and science seeks falsifications" out of context. Dominic Mayers (talk) 14:49, 3 February 2024 (UTC)

I agree with everything Dominic Mayers has said so far.

In the introduction of the article by Hirvonen & Karisto that I mentioned above, there is a helpful distinction between two types of demarcation, general and specific: "General demarcation is about providing blunt instruments for laypersons." But specific demarcation is also necessary, they said, because: "Explaining in detail why some instance of pseudoscience is pseudoscience may demand in-depth knowledge about the field of inquiry that it deals with. This could require one to answer such questions as: What kind of methods qualified scientists use in the field? Why and how do these practices differ from the pseudoscientific ones? What scientific claims do pseudoscientific theses contradict?"

I would say that JMF's example above of "crackpot cures for CV19" is a case where science communicators would need to apply what Hirvonen & Karisto call specific demarcation. The general confirmation/falsification heuristic wouldn't be specific enough; to refute CV19 pseudoscience you would need to discuss some of the in-depth background knowledge about the field that a particular CV19 pseudotheory ignores.

So, I expect there is a way to insert Stemwedel's point into the article while emphasizing that it is a general demarcation heuristic that will not be sufficient for all contexts. Biogeographist (talk) 15:57, 3 February 2024 (UTC)

Just to add that my contribution here is not at all an endorsement of the article in its current state. I had a quick look at the article and this quote was the least of its issues. It's full of naive personal opinions that reflect what the average scientists with no formation on philosophy would think. There is a difference between presenting well thought philosophical content in a way that the average people can understand and presenting what average people think. Having to do the former should not be an excuse to do the latter. Dominic Mayers (talk) 18:16, 4 February 2024 (UTC)

References

State and scope of the article March 2024

I iterated this post several times now as I worked through the article (and hopefully managed to make useful changes).

The only question I have remaining is on the general scope of the article. Several sections (Scientific inquiry, Philosophy, and Relationship with mathematics) seem they are very much their own thing and, to me, well-outside the scope of the article.

Are they here simply because of the high article traffic and to give people jumping off points to further reading? And have they been justified previously?

JackTheSecond (talk) 22:42, 8 March 2024 (UTC)

Thank you for your updates. The sections you mention used to have extensive coverage in this article. They relied on the thinking of Charles Sanders Peirce, based on the thousands of posts to his papers. User:The Tetrast was called to lead a Peirce website, and stepped away from this article a dozen years ago. User:Jon Awbrey (see Template talk:Infobox philosopher/Archive 1) and User:Prof. Carl Hewitt also contributed to the scope of this article, and to the encyclopedia.

In the last decade, the rise of proof assistants is directly traceable to Peirce, Wittgenstein, Emil Post, Alonzo Church, etc. These influences (as documented in the arXiv.org and Internet Archive) show that mathematical method and scientific method have intertwined since the 1870s. --Ancheta Wis   (talk | contribs) 09:28, 10 March 2024 (UTC)

Well, those are literal lifetimes of history... I think the most actual contribution beyond a copy edit I could give the article would be to introduce an {Empty section} on 'Modern methodology and the scientific method', as the article seems to be quite light on things past the 1960s.

Another suggestion I have is to write a 'DNA Experiment' header for the 'Confirmation' header, that gets into how DNA research moved on since. JackTheSecond (talk) 11:00, 10 March 2024 (UTC)

Direction of the 'History' section.

This 2018 diff shows that the vision for the 'History' section of the article had been to give a brief overview with a link to the article 'History of scientific method'. It has now grown and there should probably be a conversation about what to do with it and the 'Theory' section.

• Should the section keep growing? In my opinion the main article should have a complete (even if brief) History section and that means expansion at the moment. (Which is why I introduced sectioning, to encourage and organise; even if I don't know if the focus is entirely correct.)
• Is the heavy use of notes justified? And is it implemented in the best way... is perhaps the better question.
• The 'Theory' section I introduced should probably be moved or even merged elsewhere. I just didn't know what to do with one particular paragraph formerly in the 'History' section and wrote an introduction to a fitting section. ~

JackTheSecond (talk) 19:42, 10 March 2024 (UTC)

I added an anchor to characterization; credit for this term is due to user:Banno back in 2004. --Ancheta Wis   (talk | contribs) 11:24, 11 March 2024 (UTC)

Lead sentence

The further I get into the topic of the article, the more I come around to the idea that the first sentence of the lead section should not read 'empirical' and even more so not 'empirical method'. My arguments are:

• The empirical method is its own, much more limited, concept.
• Empirical implies empiricism, which is what the scientific method is built on, but far, far from being descriptive of it.
• Most openly available sources do not actually use 'empirical'. Source 2 and Britannica both do not; The Stanford Encyclopedia of Philosophy's entry on "Scientific Method" does not. (Source 3 does).
• And, above all, that Thomas Kuhn was right when he wrote about the scientific method being a 'social framework', as written about in the article already here (second paragraph).

The previous (quite high-quality) discussion that added the term was started by User:Markworthen and is archived here: 2018.

What I would change the opening line to (without the brackets):

1. The scientific method is the sociological framework that has characterized the sciences since the 17th century. (Thomas Kuhn)

2. The scientific method is a framework for procedure that has characterized the sciences since the scientific revolution of the 17th century.(Thomas Kuhn + source 2)

3. The scientific method is the sociology of science. It is not a sequence of steps, or elements – even if this article, as many educators do, will section itself such in order to give structure to explanation. (the flippant version)

4. The scientific method is a method of procedure that characterized the sciences since the scientific revolution of the 17th century. (source 2, verbatim)

I like 1 most, because the way Kuhn's definition works allows for other interpretations of it as well—no matter if one's views on it are informed by Dewey, Feyerabend or anyone in between. JackTheSecond (talk) 23:58, 19 March 2024 (UTC)

You have a point. I think the scientific method is generally a method of induction and is metaphysical. I am thinking of Francis Bacon and going back to medievals like Roger Bacon. Ramos1990 (talk) 00:39, 20 March 2024 (UTC)

I don't see any way to avoid the empirical aspect of what this article is talking about: the article is not talking about purely formal a priori knowledge. There is an empirical aspect.

I don't think it's true that Empirical implies empiricism (where empiricism refers to a particular epistemological view opposed to, e.g., rationalism) as JackTheSecond said above. Empirical can refer to the experiential or experimental aspect of a procedure, as opposed to, e.g., guessing or looking up data in an actuarial table. And, again, that empirical aspect is central to what this article is talking about.

Saying that the scientific method is sociology or sociological seems badly misleading in at least one respect: physicists, for example, are not doing sociology when they use scientific method. You seem to mean that it is social, but that property does not differentiate it from most other human activities.

For all the stated reasons, I don't see any one of the proposed sentences as an improvement over the status quo (which is not to say that that the current lead sentence could not be improved). Biogeographist (talk) 15:42, 20 March 2024 (UTC)

Good point - empirical method is imprecise and not consistent with the article as a whole. How about replacing it with scholarly method?

Scholarly method is not controversial (I don't think) and is consistent with the article as a whole. Mark D Worthen PsyD (talk) [he/him] 17:37, 20 March 2024 (UTC)

A couple of suggestions to play with:

• The scientific method is a scholarly method for acquiring knowledge that has characterized scientific inquiry since the 17th century, albeit with much debate regarding how scientific discovery actually occurs.
• The scientific method, a systematic approach to acquiring knowledge, has been a hallmark of scientific inquiry since the 17th century. However, there is considerable debate about the actual process of scientific discovery.

Mark D Worthen PsyD (talk) [he/him] 17:53, 20 March 2024 (UTC)

References