"It seems that there are hypotheses that become unfalsifiable, e.g. the claim that Jupiter has four moons. Is this claim no longer scientific?" It seems to me you don't really understand Karl Popper's methodological falsificationism, or critical rationalism. "Jupiter has four moons" is falsifiable if and only if it is possible to make a falsifying observation, given that it is false, of the hypothesis, but that is only for when it is false. To use a blunt example, 53*3=159 is a true hypothesis, but it is also falsifiable, because if it was not true, when I calculate 53*3, I would not get 159, but some different number, you have made a pretty basic mistake in not understanding the notion of falsifiability. Popper does not even talk about falsifying hypotheses, but about falsifying theories, to pretend that falsifiability is applicable to hypotheses is a gross strawman. It is obvious you have not read a single page of Karl Popper. Popper would for example consider the Multiverse theory of QM as a component of QM, which makes testable predictions, instead of branding it as 'pseudoscience' for being unverifiable, because he considered the falsifiability of theories, not hypotheses. He outright states that a scientific theory has some unfalsifiable hypotheses, and other falsifiable hypotheses, which determines its "degree of testability". I am not even a "critical rationalist", I just know this because I have read him and from watching your videos I can see you don't have a clear idea of what he actually thought.
First, you're correct that this series presents what is at best an embarrassingly oversimplified outline of Popper's own views. It's also true that, at the time I made this video, I'd read very little of Popper (though I had read some, e.g. the classic excerpts from "Conjectures and Refutations" that appear in every general philosophy of science collection...). In my defense, the form of falsificationism presented here is what I was taught as an undergrad, and it's also in line with what you can find in many intro to philosophy of science textbooks (as I noted in the intro video description, this part of the series closely follows Godfrey-Smith's "Theory and Reality"). "Textbook falsificationism" is a strawman of Popper's falsificationism. If I were doing this video now, I'd do it differently. Unfortunately, it's often the case that textbook presentations of influential philosophical positions are simplifications of what the actual philosophers held. Popper seems to have been more a victim of this than others; but to take another example from the same field, the logical positivists were generally much more varied and sophisticated than "textbook logical positivism" sometimes suggests.
@Mr Kill - while Kane's excellent video might be oversimplified, Popper himself does make it easy to knock down what you call a straw-man. He even summarises himself in C&R: "One can sum up all this by saying that the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability". That's it. There are myriad ways that one could attempt to counter this statement. It seems to me that he was trying to do too much in creating a simple axiom - he was trying to send to the dustbin of history the nonsensical application of the Hegelian dialectic by marxists to areas like mathematics and biology, while at the same time giving the criteria of what separates good and bad science, and therefore what scientists should be spending their time doing. It also seems like you have no grounds for claiming what Popper would make of the many-worlds hypothesis, given it seems as unfalsifiable as it is untestable. It's also the case that what is considered on an undergraduate-level philosophy of science course is often in advance of what is considered as PhD-level philosophy in other disciplines, and Kane is very much echoing the tone and level of what is found in underdraduate Philosophy in the UK. Thanks Kane! Really helpful.
I just want to say how much I appreciate your videos! I study at the university, and listening to your stuff really contributes to my enjoyment and shaping as an academic. Thank you! 😊
You can nitpick quite a bit at Popper, and rightly so, but there's one point he makes that remains valuable: attempts to insulate a theory from clearly contrary evidence are bad practice. It's always *possible* to thus insulate it, but it's a suspect thing for a scientific theory to do. In my subfield (philosophy of language and linguistics) such insulation from evidence is absolutely rampant, and Popper's critique is very relevant! As a complete model for how science works, Popper's theory is as flawed as its various competitors, but I think the basic intuition at the heart of it is relevant to scientific practice, and shouldn't be discarded.
To say that a hypothesis is falsifiable simply means we can look for empirical evidence against it. It doesn't mean that the hypothesis is false or that we can prove it false. Viewed this way, the hypothesis that a coin is fair is most certainly falsifiable. Am I mistaken?
It seems every objection I find to falsificationism equivocates on what it means for something to be falsifiable in order to render the principle worthless. Is there some sort of neo-falsifiability that is not worthless?
Is seeing a tree scientific? If I say, "look, a tree", am I making a scientific claim? No, it's just an observation. When the four moons were observed, they left the realm of scientific theory and became observations, which have a certainty beyond any science.
"Most scientific work doesn't involve making a risky hypothesis and then trying to refute it." "Of 70 articles published in 2000 in Nature, 49 were purely exploratory." We should probably take a closer look at those 49 articles and how they were written, because it would be quite remarkable if the people who wrote them and the journal that published them really made no effort to check the content. Wikipedia says that Nature has an intense screening process because competition to get articles into the journal is fierce, and that includes peer review by other scientists. That alone would mean that 100% of Nature articles go through a falsificationist process. Even before it gets to peer review, it would be very surprising to hear that most scientists would publish an article about the structure of a protein or the existence of exoplanets without any scientific rigor at all. A scientific journal hardly seems like a good place for publishing idle speculation. It's more likely that when a scientist suspects there may be an exoplanet around a star, the scientist first looks into ways she may be mistaken about that planet and gives at least a preliminary effort to refute her own idea before she publishes. There's no shame in having your ideas falsified, but there would be shame in being falsified too easily because you never made the slightest effort before publishing. "Generally scientists do not try to falsify previous results. They accept them and try to built on them." Scientists do have to pick and choose what they will try to falsify, and trying to falsify ideas that are already well tested seems like a daunting and fruitless task, but this part of the video makes it sound like scientific ideas are _never_ well tested because scientists will blindly accept whatever they are told. That's the sort of thinking that leads things like the reproducibility problem of psychology that's all over the web, and it is scarily plausible, but we can argue that it's not really science. Science is supposed to be an investigation into the world. When an idea just pops up and is generally never tested, that's not an investigation. When someone came up with the idea that lightning was created by Zeus, that wasn't an investigation into the origin of lightning because no one had any way to check that claim. It was just a fantasy someone dreamed up, and the same goes for any claim that is never tested. If so-called scientists generally do not try to falsify previous results, then they are fantasy writers rather than scientists. "Why suspend a great theory just because of a few anomalous results?" What does "suspend" mean in this context? Scientists are in the business of falsifying ideas, so if they suspend General Relativity does that mean they stop trying to falsify it? That would have been a foolish thing to do in the wake of one successful falsification, since a successful falsification indicates that a theory is vulnerable and the perfect target for piling on additional falsifications. If no additional falsifications came in the wake of Miller's result, there are two potential explanations. For one, scientists tried and failed to find other problems with GR. For another, scientists put GR on a pedestal and blinded themselves to contrary evidence. Perhaps the actual events were a mix of both, but surely we agree that putting ideas above scrutiny and ignoring evidence is no way to investigate the world, so anyone who did that for GR wasn't doing science. It is shocking that it took scientists 50 years to falsify Miller's results. Let's hope that the delay was caused by Miller's super-sensitive equipment being so expensive and hard to acquire, not because physicists have given up on trying to falsify ideas and gone into the fantasy fiction business. "Any theory needs time to develop, but this is not the kind of ultra-critical attitude that the falsificationist seems to require." The ultra-critical attitude is exactly what a theory needs in order to develop. A theory that is never exposed to criticism will struggle to develop. Just consider any religion. All of its ideas are shielded from criticism, and religions advance at a glacial pace. Einstein's theories didn't just pop into his head fully formed as a work of instant genius. He had rough notions, but he struggled to pin it down with mathematical precision that would exactly fit the evidence, so he went through a long process of trying various formulas and checking to see how they worked, then throwing them away when he discovered they were wrong. If he hadn't been so ultra-critical of his own newborn theory then perhaps we'd never have discovered GR because he'd just have published the first formula that came into his head. "There is much debate among biologists about the proper methods of classification." Would anyone really claim that biologists are investigating the world when they are debating how we should name things? A dinosaur's biology does not depend on what we choose to call it. It's no surprise that many biologists would be interested in the philosophy of biology, but that doesn't mean that it's part of their scientific work. "Is falsificationism a descriptive claim about how science does work, or a normative claim about how science should work?" Falsificationism is a descriptive claim. It explains how the world can be investigated. There's nothing normative about that, since falsificationism offers no opinion about whether we should investigate the world or not. "If the former, it seems to have been falsified. Scientists often do not adopt the falsificationist method." The meaning of this is not clear. Are we supposed to take from that the the scientists who do not adopt the falsificationist method probably have good reason and therefore it seems they have somehow falsified falsificationism? Or is the claim that scientists do not adopt the falsificationist method being presented as direct falsification? If it's the former then that's highly unconvincing. If it's the latter then it's a non sequitur. Suppose all the world's barbers suddenly decided to mow lawns instead of cutting hair; that wouldn't falsify the idea that barbers cut hair. It would just mean that the people we're calling barbers are barbers in name only. The claim of falsificationism isn't about what so-called scientists do; falsificationism is a description of how to investigate the world. Whether people choose to do that or not is their own concern. "What grounds are there for thinking that adopting the method of falsificationism will actually help scientists? Science seems to be getting along perfectly fine as it is." This seems like a question for the beginning of a series about falsificationism. Whole books could be written about why falsificationism is the way to investigate the world, starting with illustrations of how induction is unreliable. Beyond that we could go into how it is dangerous to allow any idea to be shielded from criticism and how pushing ideas through a gauntlet of attacks filters away the vulnerable ideas while leaving only the strongest.
"Even before it gets to peer review, it would be very surprising to hear that most scientists would publish an article about the structure of a protein or the existence of exoplanets without any scientific rigor at all" Hansson's claim is not that the articles were published "without any scientific rigor", or without trying to check the content. Come on, Ansatz - that's an obvious strawman. His claim is that the scientists did not follow the falsificationist method of: (1) propose a risky, bold hypothesis and (2) attempt to refute that hypothesis.
"His claim is that the scientists did not follow the falsificationist method of: (1) propose a risky, bold hypothesis and (2) attempt to refute that hypothesis." How is that different from what I take it to mean? I presume that we all agree that proposing the structure of a disease protein or the existence of an exoplanet is a risky hypothesis, so surely that means that Hansson was claiming that they didn't attempt to refute their own hypotheses. Therefore I imagine they just crossed their fingers and hoped that their hypotheses were hard to refute, since they made no effort in that direction themselves before publishing it to the world. It would be bizarre behavior for any scientist, and it's even less likely that a respected journal would act that way, because publishing easily refuted junk looks bad. It is impossible to believe that 49 out of 70 articles were published prior any attempt at refutation. I'm all for the principle of charity, but I honestly can't see an interpretation of this claim that makes sense.
Ansatz66 no you've misunderstood what is meant by an exploratory article. It's not an article which asserts untested hypotheses, but one which discusses data and what it suggests about the question the scientist is interested in. It needn't contain any well-defined hypothesis at all, and it certainly need' entail assertion of one. Why would a scientist be embarrassed if they just say 'this is an interesting hypothesis' and someone else refutes it? Or if someone comes up with one on the basis of your exploratory analysis and falsified it? This seems baseless. If a scientist has noticed something interesting about some data (whether that something amounts to a hypothesis or not) they have good reason to want to publish their observations, and a journal may have good reason to publish it. I mean, do you really think that scientists shouldn't bother discussing phenomena or patterns they've noticed from investigating things until they have falsifiable hypotheses? It seems like such discussion in journals and other places could lead to falsifiable hypotheses being formulated. Yes such hypotheses sometimes come out of nowhere, but it doesn't mean that you can always bank on that. I mean, I just find your views here to be quite odd and indicative of an extreme idealisation of how science works.
"It's not an article which asserts untested hypotheses, but one which discusses data and what it suggests about the question the scientist is interested in." I agree. It's Sven Hansson who was saying that the claims of exploratory articles are untested, assuming this video represents him fairly. I was just saying that Hansson's claim would make for very strange behavior from a scientist. "Why would a scientist be embarrassed if they just say 'this is an interesting hypothesis' and someone else refutes it?" A scientist usually wants a reputation for competence . It's fine for an idea to be refuted, but if the refutation is too easy then it makes the scientist look like she is too incompetent to see an obvious refutation for herself. If a scientist doesn't try to refute her own ideas, then she's opening herself up to looking like she is incapable of refuting them even when they would be easy to refute for any competent scientist in her field. She only failed because she never tried, but people aren't going to know that. You probably wouldn't worry about that when you're throwing around ideas in a casual setting because people know you're just speculating, but a published article in Nature deserves more care and attention than that. "Do you really think that scientists shouldn't bother discussing phenomena or patterns they've noticed from investigating things until they have falsifiable hypotheses?" Falsifiable hypotheses are easy. It doesn't require any care or attention to make a falsifiable hypothesis, since it just needs to be a claim about something observable in the world. So yes, I really think scientists shouldn't bother discussing unfalsifiable ideas like gods and spirit worlds and so on, at least until they can figure out some way to make a falsifiable hypothesis to talk about. If it can't be falsified then it's useless for scientists.
Ansatz66 I mean, it may be that an idea is promising but it's not obvious how to test it. There is value in discussing it so that we can figure out how to test it. Falsification is not always easy, in fact a fairly brief foray into the philosophy of science will tell you otherwise (Kane gives some reasons for thinking this in the video), so there is no shame in writing about an untested idea. If someone else is able to design a test for it and it is falsified, that again is nothing to be ashamed of and may even contribute to scientific progress. And again, an exploratory article doesn't need to be talking about a hypothesis at all, it could just be about observations based on experiments, which do not amount to theories. The point is that when scientists pursue research they do not always do so with a hypothesis in mind, sometimes they just do experiments on things they're interested in, and the purely exploratory articles Hansson is talking about might just be about that. You really seem to be interpreting his argument as uncharitably as possible.
"It seems that there are hypotheses that become unfalsifiable, e.g. the claim that Jupiter has four moons. Is this claim no longer scientific?" It seems to me you don't really understand Karl Popper's methodological falsificationism, or critical rationalism. "Jupiter has four moons" is falsifiable if and only if it is possible to make a falsifying observation, given that it is false, of the hypothesis, but that is only for when it is false. To use a blunt example, 53*3=159 is a true hypothesis, but it is also falsifiable, because if it was not true, when I calculate 53*3, I would not get 159, but some different number, you have made a pretty basic mistake in not understanding the notion of falsifiability. Popper does not even talk about falsifying hypotheses, but about falsifying theories, to pretend that falsifiability is applicable to hypotheses is a gross strawman. It is obvious you have not read a single page of Karl Popper. Popper would for example consider the Multiverse theory of QM as a component of QM, which makes testable predictions, instead of branding it as 'pseudoscience' for being unverifiable, because he considered the falsifiability of theories, not hypotheses. He outright states that a scientific theory has some unfalsifiable hypotheses, and other falsifiable hypotheses, which determines its "degree of testability". I am not even a "critical rationalist", I just know this because I have read him and from watching your videos I can see you don't have a clear idea of what he actually thought.
First, you're correct that this series presents what is at best an embarrassingly oversimplified outline of Popper's own views. It's also true that, at the time I made this video, I'd read very little of Popper (though I had read some, e.g. the classic excerpts from "Conjectures and Refutations" that appear in every general philosophy of science collection...). In my defense, the form of falsificationism presented here is what I was taught as an undergrad, and it's also in line with what you can find in many intro to philosophy of science textbooks (as I noted in the intro video description, this part of the series closely follows Godfrey-Smith's "Theory and Reality").
"Textbook falsificationism" is a strawman of Popper's falsificationism. If I were doing this video now, I'd do it differently. Unfortunately, it's often the case that textbook presentations of influential philosophical positions are simplifications of what the actual philosophers held. Popper seems to have been more a victim of this than others; but to take another example from the same field, the logical positivists were generally much more varied and sophisticated than "textbook logical positivism" sometimes suggests.
@Mr Kill - while Kane's excellent video might be oversimplified, Popper himself does make it easy to knock down what you call a straw-man. He even summarises himself in C&R: "One can sum up all this by saying that the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability". That's it. There are myriad ways that one could attempt to counter this statement. It seems to me that he was trying to do too much in creating a simple axiom - he was trying to send to the dustbin of history the nonsensical application of the Hegelian dialectic by marxists to areas like mathematics and biology, while at the same time giving the criteria of what separates good and bad science, and therefore what scientists should be spending their time doing. It also seems like you have no grounds for claiming what Popper would make of the many-worlds hypothesis, given it seems as unfalsifiable as it is untestable.
It's also the case that what is considered on an undergraduate-level philosophy of science course is often in advance of what is considered as PhD-level philosophy in other disciplines, and Kane is very much echoing the tone and level of what is found in underdraduate Philosophy in the UK. Thanks Kane! Really helpful.
Kane B I am starting a PhD in philosophy this fall. Thanks for your videos. They are fantastic!
I hope you are a doctor now!
@@jeevacation haha, planning to defend next month! Thank you 🙏
@@azimuthal1392 That's awesome! I hope you get your doctorate!
I just want to say how much I appreciate your videos! I study at the university, and listening to your stuff really contributes to my enjoyment and shaping as an academic. Thank you! 😊
You can nitpick quite a bit at Popper, and rightly so, but there's one point he makes that remains valuable: attempts to insulate a theory from clearly contrary evidence are bad practice. It's always *possible* to thus insulate it, but it's a suspect thing for a scientific theory to do. In my subfield (philosophy of language and linguistics) such insulation from evidence is absolutely rampant, and Popper's critique is very relevant! As a complete model for how science works, Popper's theory is as flawed as its various competitors, but I think the basic intuition at the heart of it is relevant to scientific practice, and shouldn't be discarded.
To say that a hypothesis is falsifiable simply means we can look for empirical evidence against it. It doesn't mean that the hypothesis is false or that we can prove it false. Viewed this way, the hypothesis that a coin is fair is most certainly falsifiable. Am I mistaken?
It seems every objection I find to falsificationism equivocates on what it means for something to be falsifiable in order to render the principle worthless. Is there some sort of neo-falsifiability that is not worthless?
Brilliant, as always.
Can we get from now on the whole script ? (From an link ?)
Pastebin would be good.
20:15
"Smoking lowers the probability of cancer" may be a false proposition, but still is perfectly conceivable and implies no contradiction (Hume).
Is seeing a tree scientific? If I say, "look, a tree", am I making a scientific claim? No, it's just an observation. When the four moons were observed, they left the realm of scientific theory and became observations, which have a certainty beyond any science.
evolutionary psychology, ,,,?
"Most scientific work doesn't involve making a risky hypothesis and then trying to refute it." "Of 70 articles published in 2000 in Nature, 49 were purely exploratory."
We should probably take a closer look at those 49 articles and how they were written, because it would be quite remarkable if the people who wrote them and the journal that published them really made no effort to check the content. Wikipedia says that Nature has an intense screening process because competition to get articles into the journal is fierce, and that includes peer review by other scientists. That alone would mean that 100% of Nature articles go through a falsificationist process.
Even before it gets to peer review, it would be very surprising to hear that most scientists would publish an article about the structure of a protein or the existence of exoplanets without any scientific rigor at all. A scientific journal hardly seems like a good place for publishing idle speculation. It's more likely that when a scientist suspects there may be an exoplanet around a star, the scientist first looks into ways she may be mistaken about that planet and gives at least a preliminary effort to refute her own idea before she publishes. There's no shame in having your ideas falsified, but there would be shame in being falsified too easily because you never made the slightest effort before publishing.
"Generally scientists do not try to falsify previous results. They accept them and try to built on them."
Scientists do have to pick and choose what they will try to falsify, and trying to falsify ideas that are already well tested seems like a daunting and fruitless task, but this part of the video makes it sound like scientific ideas are _never_ well tested because scientists will blindly accept whatever they are told. That's the sort of thinking that leads things like the reproducibility problem of psychology that's all over the web, and it is scarily plausible, but we can argue that it's not really science.
Science is supposed to be an investigation into the world. When an idea just pops up and is generally never tested, that's not an investigation. When someone came up with the idea that lightning was created by Zeus, that wasn't an investigation into the origin of lightning because no one had any way to check that claim. It was just a fantasy someone dreamed up, and the same goes for any claim that is never tested. If so-called scientists generally do not try to falsify previous results, then they are fantasy writers rather than scientists.
"Why suspend a great theory just because of a few anomalous results?"
What does "suspend" mean in this context? Scientists are in the business of falsifying ideas, so if they suspend General Relativity does that mean they stop trying to falsify it? That would have been a foolish thing to do in the wake of one successful falsification, since a successful falsification indicates that a theory is vulnerable and the perfect target for piling on additional falsifications.
If no additional falsifications came in the wake of Miller's result, there are two potential explanations. For one, scientists tried and failed to find other problems with GR. For another, scientists put GR on a pedestal and blinded themselves to contrary evidence. Perhaps the actual events were a mix of both, but surely we agree that putting ideas above scrutiny and ignoring evidence is no way to investigate the world, so anyone who did that for GR wasn't doing science.
It is shocking that it took scientists 50 years to falsify Miller's results. Let's hope that the delay was caused by Miller's super-sensitive equipment being so expensive and hard to acquire, not because physicists have given up on trying to falsify ideas and gone into the fantasy fiction business.
"Any theory needs time to develop, but this is not the kind of ultra-critical attitude that the falsificationist seems to require."
The ultra-critical attitude is exactly what a theory needs in order to develop. A theory that is never exposed to criticism will struggle to develop. Just consider any religion. All of its ideas are shielded from criticism, and religions advance at a glacial pace. Einstein's theories didn't just pop into his head fully formed as a work of instant genius. He had rough notions, but he struggled to pin it down with mathematical precision that would exactly fit the evidence, so he went through a long process of trying various formulas and checking to see how they worked, then throwing them away when he discovered they were wrong. If he hadn't been so ultra-critical of his own newborn theory then perhaps we'd never have discovered GR because he'd just have published the first formula that came into his head.
"There is much debate among biologists about the proper methods of classification."
Would anyone really claim that biologists are investigating the world when they are debating how we should name things? A dinosaur's biology does not depend on what we choose to call it. It's no surprise that many biologists would be interested in the philosophy of biology, but that doesn't mean that it's part of their scientific work.
"Is falsificationism a descriptive claim about how science does work, or a normative claim about how science should work?"
Falsificationism is a descriptive claim. It explains how the world can be investigated. There's nothing normative about that, since falsificationism offers no opinion about whether we should investigate the world or not.
"If the former, it seems to have been falsified. Scientists often do not adopt the falsificationist method."
The meaning of this is not clear. Are we supposed to take from that the the scientists who do not adopt the falsificationist method probably have good reason and therefore it seems they have somehow falsified falsificationism? Or is the claim that scientists do not adopt the falsificationist method being presented as direct falsification? If it's the former then that's highly unconvincing. If it's the latter then it's a non sequitur. Suppose all the world's barbers suddenly decided to mow lawns instead of cutting hair; that wouldn't falsify the idea that barbers cut hair. It would just mean that the people we're calling barbers are barbers in name only. The claim of falsificationism isn't about what so-called scientists do; falsificationism is a description of how to investigate the world. Whether people choose to do that or not is their own concern.
"What grounds are there for thinking that adopting the method of falsificationism will actually help scientists? Science seems to be getting along perfectly fine as it is."
This seems like a question for the beginning of a series about falsificationism. Whole books could be written about why falsificationism is the way to investigate the world, starting with illustrations of how induction is unreliable. Beyond that we could go into how it is dangerous to allow any idea to be shielded from criticism and how pushing ideas through a gauntlet of attacks filters away the vulnerable ideas while leaving only the strongest.
"Even before it gets to peer review, it would be very surprising to hear that most scientists would publish an article about the structure of a protein or the existence of exoplanets without any scientific rigor at all"
Hansson's claim is not that the articles were published "without any scientific rigor", or without trying to check the content. Come on, Ansatz - that's an obvious strawman. His claim is that the scientists did not follow the falsificationist method of: (1) propose a risky, bold hypothesis and (2) attempt to refute that hypothesis.
"His claim is that the scientists did not follow the falsificationist method of: (1) propose a risky, bold hypothesis and (2) attempt to refute that hypothesis."
How is that different from what I take it to mean? I presume that we all agree that proposing the structure of a disease protein or the existence of an exoplanet is a risky hypothesis, so surely that means that Hansson was claiming that they didn't attempt to refute their own hypotheses. Therefore I imagine they just crossed their fingers and hoped that their hypotheses were hard to refute, since they made no effort in that direction themselves before publishing it to the world. It would be bizarre behavior for any scientist, and it's even less likely that a respected journal would act that way, because publishing easily refuted junk looks bad. It is impossible to believe that 49 out of 70 articles were published prior any attempt at refutation.
I'm all for the principle of charity, but I honestly can't see an interpretation of this claim that makes sense.
Ansatz66 no you've misunderstood what is meant by an exploratory article. It's not an article which asserts untested hypotheses, but one which discusses data and what it suggests about the question the scientist is interested in. It needn't contain any well-defined hypothesis at all, and it certainly need' entail assertion of one. Why would a scientist be embarrassed if they just say 'this is an interesting hypothesis' and someone else refutes it? Or if someone comes up with one on the basis of your exploratory analysis and falsified it? This seems baseless. If a scientist has noticed something interesting about some data (whether that something amounts to a hypothesis or not) they have good reason to want to publish their observations, and a journal may have good reason to publish it. I mean, do you really think that scientists shouldn't bother discussing phenomena or patterns they've noticed from investigating things until they have falsifiable hypotheses? It seems like such discussion in journals and other places could lead to falsifiable hypotheses being formulated. Yes such hypotheses sometimes come out of nowhere, but it doesn't mean that you can always bank on that. I mean, I just find your views here to be quite odd and indicative of an extreme idealisation of how science works.
"It's not an article which asserts untested hypotheses, but one which discusses data and what it suggests about the question the scientist is interested in."
I agree. It's Sven Hansson who was saying that the claims of exploratory articles are untested, assuming this video represents him fairly. I was just saying that Hansson's claim would make for very strange behavior from a scientist.
"Why would a scientist be embarrassed if they just say 'this is an interesting hypothesis' and someone else refutes it?"
A scientist usually wants a reputation for competence . It's fine for an idea to be refuted, but if the refutation is too easy then it makes the scientist look like she is too incompetent to see an obvious refutation for herself. If a scientist doesn't try to refute her own ideas, then she's opening herself up to looking like she is incapable of refuting them even when they would be easy to refute for any competent scientist in her field. She only failed because she never tried, but people aren't going to know that. You probably wouldn't worry about that when you're throwing around ideas in a casual setting because people know you're just speculating, but a published article in Nature deserves more care and attention than that.
"Do you really think that scientists shouldn't bother discussing phenomena or patterns they've noticed from investigating things until they have falsifiable hypotheses?"
Falsifiable hypotheses are easy. It doesn't require any care or attention to make a falsifiable hypothesis, since it just needs to be a claim about something observable in the world. So yes, I really think scientists shouldn't bother discussing unfalsifiable ideas like gods and spirit worlds and so on, at least until they can figure out some way to make a falsifiable hypothesis to talk about. If it can't be falsified then it's useless for scientists.
Ansatz66 I mean, it may be that an idea is promising but it's not obvious how to test it. There is value in discussing it so that we can figure out how to test it. Falsification is not always easy, in fact a fairly brief foray into the philosophy of science will tell you otherwise (Kane gives some reasons for thinking this in the video), so there is no shame in writing about an untested idea. If someone else is able to design a test for it and it is falsified, that again is nothing to be ashamed of and may even contribute to scientific progress. And again, an exploratory article doesn't need to be talking about a hypothesis at all, it could just be about observations based on experiments, which do not amount to theories. The point is that when scientists pursue research they do not always do so with a hypothesis in mind, sometimes they just do experiments on things they're interested in, and the purely exploratory articles Hansson is talking about might just be about that. You really seem to be interpreting his argument as uncharitably as possible.