 All right, our next speaker is Massimo Pagliucci. We are set, yes, we are set. Massimo Pagliucci, he's gonna be doing a demarcation. What exactly is the difference between science and pseudoscience? He is the chair for the Department of Philosophy at City University, but he is an evolutionary biologist as well as a philosopher, so both sides of that coin. His haiku, we have two speakers named M-A-S-S-I-M-O. Salve signore. Please welcome Massimo Pagliucci. Thank you, there will be no singing or dancing on my part during this talk. So it's great to be back here at TAM. I wanna talk to you about the demarcation problem and why skeptics should really care, give a damn about it. It's part of a new book that is about to come out for University of Chicago Press. Several philosophers, scientists and even skeptics have contributed. My friend Michael Shermer, for instance, has a chapter in that book. You all saw this morning Michael's presentation about science and morality, I hope so, right? And I have one or two caveats about that presentation, maybe three or four, five or six, okay. But we won't get there. If you're interested, actually there is video of a nice discussion we had recently about it. Today it's a demarcation problem, which seems like the kind of thing that we all ought to know about. You know, it's this time, of course I know when I said pseudoscience, when I see it, turns out it's gonna be a little more complicated, it's gonna be more like pornography, difficult to define. I mean, do you think you know what it is? So the first thing is, what is the problem? What exactly we're talking about? We're not talking about something as simple as what this slide is showing, which is, in this case, the difference between astrology and astronomy. It's, that one is pretty easy enough. It doesn't require that much of a cogitation. But a lot of the times it gets more complicated. There are areas at the borderlands between science and pseudoscience, there is bad science, there is quasi-okay science, and then not only that, but things change over time. Something starts out as sounding pretty much pseudoscientific and then it may turn into mainstream science or vice versa. Something starts out as a promising scientific enterprise and then it goes the other way. That's what makes the whole thing, of course, more fun, but also more frustrating. Now why doesn't matter? Why do we care? Precisely, I think there's three reasons why we should care. One, of course, is the sheer intellectual endeavor. I mean, presumably, you guys are here in part because you're intellectually curious, and so this is an interesting problem. It's an intellectual puzzle. The more practical reason is because there's a lot of money that goes into science. And so it does help the public in general and even scientists and philosophers in particular being able to figure out, well, is this actually this kind of thing worth the money or not? For instance, as you know, there is money that the National Institute of Health is putting into research on so-called alternative medicine. So is it worth the money? How much money is it worth? That sort of stuff. And then finally, because pseudoscience hurts or kills, and I don't have to convince this audience about that, that particular picture there shows a woman with a tattoo that says no AZT, so no anti-retrovital drugs. She's also pregnant, you might notice, and so that means that because she doesn't believe that HIV causes AIDS, her baby's probably gonna get AIDS. So these things matter intellectually. They matter in terms of sort of practical terms. They also matter in terms of financial terms, but also in terms of how they affect people's lives. Now, if you've heard of any philosopher, probably it's this guy, Sir Cal Popper, who is responsible for actually beginning discussion on the demarcation problem back in the early part of the 20th century, and also for inventing, for coming up with the first recent solution, recent memory to the problem, and that has to do with falsification, with the concept of falsification. Now, you've probably heard of this before because it's almost no skeptic or scientist that never heard of falsification, although they don't necessarily know exactly what that means. So one of the things that Popper pointed out is of course that the difference between science and science is that science makes progress not by confirmation of theories, but by falsifying theories, by making sure that we reject theories if they're wrong. The idea was that it's too easy to confirm stuff because, well, today we know about the confirmation bias, of course, from cognitive science, but even Popper was pretty well aware that even scientists can easily go out and collect data, information that is actually in favor of their particular special hypothesis, it's much more difficult, however, to find out something that resists falsification. So the idea is the difference was if it's falsifiable, it's scientific, if it's not falsifiable, it's anything goes. And Popper's examples of those two categories were on the one hand Marxist theories of history and psychoanalytic theories, Freudian and other versions. Those were unfalsifiable because you can always find confirming evidence of those things. Anytime you talk to a Marxist historian or to a psychoanalyst, they will say that whatever, regardless of what you tell them, they will say that, yeah, of course that fits, and they come up with a story for why that fits their theory. So every single instance of empirical evidence turns out to be in favor of the hypothesis. Well, that's not gonna work, said Popper. Real science works differently and his best example was the attempt at falsifying Einstein's theory of relativity which happened in 1919. There was a famous eclipses, total eclipse of the sun and Einstein's theory was making a very, very specific prediction about what would happen during that eclipse, particularly the bending of the light of the stars emerging from behind the disk of the sun. And that prediction was so specific that as Popper put it, the theory was sticking its neck out and it could have been shocked if the observations were not in fact in favor of the theory. It turns out that the observation was spectacular in favor of the theory. Einstein became a celebrity overnight and ever since, in fact, the theory of relativity, the general theory of relativity has survived a number of other attempts at falsification. That's why people are so confident that it is on the right track. It's not that it's the true theory, it's not that it is, we're not confident of it because it makes, it has been confirmed a lot of time but because it has, people have tried to falsify it a lot of times and it has survived. Not so Marxist theories of history or psychoanalysis. Now the problem is that this doesn't really work very well. It doesn't work in the history of science, I'll show you an example in a minute. And it doesn't work from a point of view, sort of general point of view of philosophy of science because it is in fact very difficult to honestly falsify a theory for two reasons. These two reasons are often conflated into one it's called the Duhin-Quine thesis but in fact there are two different reasons. One is due to Pierre Duham who was a physicist, who was interested in philosophy of science and he pointed out that scientists can always rescue hypotheses by assuming if the hypothesis fails to predict the data, to square with the data. If they assume that there is something else that is wrong there may be something that is not working with the equipment that has been used to make the observations or it may be that the data analysis were not correct, we just heard examples of this kind of justification, ad hoc justification. But sometimes these ad hoc justification are correct. Sometimes the equipment really doesn't work. Sometimes the data analysis really are not the best you can do. So Duhin pointed out that straightforward falsification as Popper thought does not actually work. And in fact it's worse than that. W.V.O. Quine who was one of the most influential philosophers of the middle part of the 20th century pointed out that you can make the same argument more generally, that is, the entire, what he called the entire web of knowledge, everything we know about the world, including all the assumptions that go into doing scientific research, including logic itself. All of those can be tweaked in theory in order to adjust the match between the theory and the observations, which means you have an infinite degree of number of freedom, you can do whatever you want to and eventually you square things with the observations. If that's the case, if the Duhin-Quine thesis are correct, then you cannot falsify the scientific hypothesis or any hypothesis. And then what? That makes it much more difficult to separate science from pseudoscience, good science from vet science and so on and so forth. Let me give you one example of what I'm talking about. So this is from the history of science and it's in the concerns of Newtonian mechanics. At some point astronomers noticed that there was a problem with the orbit of the planet Uranus, which is the one on the left of the screen, that it wasn't following the predictions that were being derived by using Newtonian mechanics. That is, you calculate the position on the planet and you point the telescope and the planet was slightly off. And then you do it again and the planet was slightly off and you do it again and the planet was slightly off. Check the calculations, the calculations were correct, so something had to give. Now, a naive falsificationist, a naive properium, would say, well, too bad for the Newtonian theory, mechanical theory. It's clearly wrong. It's giving me approximate predictions but the predictions are actually wrong, so let's throw it away. Of course astronomers were not even thinking about throwing away Newtonian mechanics. What they did instead was to come up with an adult hypothesis. That's well in a minute. Maybe there is another planet out there which we don't know about that is causing these anomalies and it is because of that planet that there is a mismatch between the observations and the data. That sounds like another hypothesis. You're postulating a non-invisible entity just so that your calculations come out right. Of course, this is science, not pseudoscience, but what happened is they calculated the position of the new planet. They pointed the telescopes in the right direction and sure enough, the planet was there naturally. In fact, the mathematician that actually calculated this whole thing was so sure that the result was correct that he was offered the possibility of being the first one to look into the telescope and he said, no, no, I know it's there. Talk about being confident, right? Now, this is a very nice story which simply shows that scientists don't actually use falsification in actual historical situations and so that enough is a problem for the popper's account of the location, but it gets worse because a few years later, astronomers found themselves in a similar situation as far as the planet Mercury is concerned. Mercury also was showing these anomalies and you were deploying in the calculations based on Newtonian mechanics and the calculations were giving you the wrong position as far as Mercury is concerned. Slightly wrong, but wrong nonetheless and repeatedly so. It was clearly a systematic error somewhere. So astronomers said, well, we know what's gonna happen now. There is another new planet as yet undiscovered. It is these planets that is causing the anomalies so we'll just have to calculate the orbital parameters, point the telescope and discover the new planet. Now, the new planet had to be closer to the Sun than Mercury because otherwise we would have already discovered it and they were so confident that they would find the planet that they actually named it ahead of the discovery, Vulcan, because as we know, astronomers are a bunch of trekkies. There's a problem, however, that they started looking for Vulcan and they looked for Vulcan and they couldn't find it and they redid the calculation and repointed it that as government Vulcan was nowhere to be found. In fact, as it turns out, in this case, it was Newtonian mechanics that was the problem. Turns out that Mercury is close enough to the Sun that relativistic effects, which are not accounted for by Newtonian mechanics, you need the general theory of relativity, are sufficient to cause those anomalies and therefore you need to bring in general relativity. You really throw out, in this case, Newtonian mechanics. So we have a case, an historical case, where exactly the same situation happened. In both cases, apparently, the theory was falsified by the data, but in one case, it turns out, astronomers rightly so kept the theory and modified it and that led to a spectacular discovery. In another case, on the other hand, they actually throw away the theory. So what's the difference between the two cases? There is no formal difference between the two cases. It's just that people were trying to do their best and at some point they figured, well, you know, we tried our best, we've done enough modifications of the basic idea of Newtonian mechanics, it's not working, so at this point we really do throw it away. So the theory was in fact falsified, but it took a hell of a lot more effort than Popper would have imagined. Okay, bring us back a little closer to modern time. So this was the situation up until the 50s or early 60s. And then in 1987, an influential philosopher of science, Larry London, said, well, you know what? I think this all debate on demarcation is wrong and it's not leaning anywhere. We're not gonna come up with any good answer and the reason for that is because the debate hinges on three questionable meta-philosophical assumptions. Now if you don't like philosophy, you're not gonna like meta-philosophy even less. Nonetheless, we're gonna take a look at the three meta-philosophical assumptions because I think we're gonna learn something from it. So the first meta-philosophical assumption is, you know, what are the conditions of adequacy that should be proposed for demarcation criteria? So what is it that makes it something science and something not science? What kind of conditions do we have? Is the criterion under consideration often necessary and sufficient conditions? Philosophers, like effect scientists as well, like this idea of necessary and sufficient conditions. If I can give you a short list of things that are both necessary and sufficient to separate one thing from another, then we got a good definition. And then he was asking, Lowland was asking, what actions or judgments are implied by the claim that something is, in fact, scientific or unscientific? So what are we gonna, once we've done the separation between science and pseudoscience or sudden science and bad science, then what, what are we gonna do about it? So here's the three answers to the three meta-philosophical questions. First of all, what are the conditions of adequacy? So Lowland said that the conditions of adequacy is that philosophers should agree, in fact, he said they should better agree on whatever separation between science and pseudoscience scientists come up with. Because they are the experts, they are doing the science, so we better agree with whatever they come up with. I don't think so. I mean, generally speaking, yes. I think it's indicative of a problem. If a philosopher points out to one solution, the scientist says, no, that's completely wrong, and there's no way to reconcile the two. Then we have a problem. Either one of the two is gonna be wrong. But I think that a better way of going about it is that whatever philosophers are gonna come up with to tell us how to separate science from pseudoscience, generally speaking, agrees with what the scientists say, but there may be exceptions, and those exceptions may actually be the interesting stuff. So the answer should look something like this. This is a clustered diagram that puts together things that are similar. So it clearly should be the case that both philosophers and scientists agree that, say, fundamental physics or evolutionary biology are, in fact, sciences. If they disagree on that, we got a problem. They also better agree on things like, well, astrologists definitely have pseudoscience. So those are the easy case, or intelligent design creationism is pseudoscience. Those are the easy cases, but there may be some interest in disagreement, and in fact, that may be what the interesting stuff is, in the middle. What kind of science is, for instance, evolutionary psychology, or the search for extraterrestrial intelligence? Is there a possibility that we can turn part of psychology into an actual legitimate science? Those kind of cases are much more interesting because there may be a legitimate disagreement there. So the criteria of adequacy, I think, is there should be a broad agreement between practicing scientists and philosophers of science, but up to a point. Second, what about necessary and sufficient criteria? So Lannan said that, ideally, a demarcation criterion would specify a set of individually necessary and jointly sufficient conditions for deciding whether something belongs to science or not. This is a nice, clear-cut kind of definition that comes out of basic logic and basic applications in math and simple science. So, for instance, if I give you definition of water, I say, well, water is H2O, and the story. If your chemical has a composition H2O, that is both necessary and sufficient for being water. You don't need anything else. Unfortunately, a lot of interesting concepts are much more complicated than water, or especially when it comes to the differences between science and pseudoscience. And in fact, philosophers have known this for a while. This guy here is Ludwig Wittgenstein, who, in the middle part of the 20th century, actually thought about the nature of complex concepts such as science or pseudoscience, although he rarely used that particular example. And he figured out that actually there's something intrinsic about these concepts that make it difficult to come up with simple, very clear-cut definitions. His example was the example of games. Now, try it at home, if you like, as an exerciser, tonight over beer, even better. Try to come up with a simple, clear definition of what constitutes a game. We're in a casino, so that seems appropriate. It's much more difficult than you think. Because if you start doing things like, well, games are things that have rules. Yes, but a lot of things have rules that they're not games. They're done, they're competitive. Well, actually, there are some games that are not competitive, and there are some activities that are competitive and are not games. Well, they're done for fun. Well, sex is done for fun, but it's not a game, and so on and so forth. So as it turns out, the more you think about it, the more it's actually difficult to come up with a definition of even something apparently simple as the concept of game, let alone the concept of science or pseudoscience. So what in fact it looks like is that game is a family of concepts, and there are some similarities, and there are some dissimilarities, and it's a little bit more vague, and it's a little bit more, you have to use something closer to sort of fuzzy logic to actually come up with decent definitions of what a game is. And by fuzzy logic, I don't mean woo-hoo logic. I mean technically fuzzy logic. There is a type of logic that actually deals with these kinds of situations. So applied to science as to the science, this is what the bitkensteinian games will look like. So what you have there, for instance, are a couple of axes over which you can represent approximately the, let's say, theoretical sophistication of a particular enterprise and the empirical support for that enterprise. And you can see that at one corner, at the lower left corner, you have the easy cases. Astrology, intelligent design creationism, that sort of stuff. Those have very little, if any, theoretical sophistication, very little, if any, empirical support. At the opposite extreme, you have the easy cases clearly inside science. So you have fundamental physics, you have evolution and biology, chemistry, those kinds of things, where you have a high degree of theoretical sophistication, a high degree of empirical support. But the interesting stuff is in between. Where do we put, let's say, the social sciences? Well, a lot of the social sciences, for instance, have a lot of empirical support. There's a lot of empirical data, and they're repeatable, they really look like sciences, but there's very little conceptual theoretical analysis. There's no overarching theory yet of psychology, for instance, or sociology, even less. Or some other social sciences are suffer of the opposite problem. They have a lot of theoretical structure, like economics, but that theoretical structure doesn't seem to much particularly well with the empirical world. Just look at the 2008 financial crisis, as an example. So those are the, that's the picture you get. It's a fuzzy, gradual picture where some things are clearly science, some things are clearly non-science, and some stuff is in between. You notice also the position of string theory. I don't know what it is, let's show it again. If we can, it's on the lower right corner. Lots of theoretical sophistication, zero data. Yeah, I can't control the focus, I'm sorry. Anyway, lots of theoretical sophistication, exactly zero data. Well, that's simplifying, but anyway. Now, so what I'm suggesting is that we need to move away from sort of standard approaches to using classical logic, like H2O is in fat water, to something more fuzzy logic. Now, if you moved over fuzzy logic to analyze concepts, you lose in precision, and you gain in significance. Now, here's the difference between precision and significance. The guy on the left says, and you have to read exactly what he says. That's 500 kilograms that is coming down at a certain speed and where certain acceleration. And the other guy says, get the hell out of there. The second guy is much less precise, but his advice is much more significant. So that's two. Third criterion was, well, what kind of judgments are implied? If you read that quote from Lallon, it says that, well, philosophers should, in fact, not be afraid of essentially getting into value judgments. That is, philosophers should be prescriptive, not just descriptive. But at the same time, he says, yes, but the problem is that when you start putting things into two piles and you call one science and the other one pseudoscience, now you're making the kind of value judgment that actually does have practical consequences for people. So you need to be careful. Maybe you shouldn't go there because as soon as you label something pseudoscience, you automatically, it's in a sort of derogative term and so on and so forth. It seems to me that Lallon and people like that actually are in this kind of situation. I don't know if you can read the cartoon, but this guy is working in a very sophisticated theoretical proof and then part of that proof are these two little bits that one says, one cake, the other one says, eat it too. And you can't have both. That's what Lallon is trying to do. He's saying, well, yes, it's okay to make value judgments. On the other hand, be careful because you're making value judgments. I don't have a problem making value judgments if they are based on reason and evidence. And I think that that is precisely what scientists and philosophers and skeptics ought to do if they have good reasons to do it. So to label something pseudoscience actually does do good value. It has a good thing. It is a good thing for society as long as you can back up that judgment with reason and information and data. And you know what I'm talking about. The situation is in fact pretty bad. These are some of the standard statistics that the National Science Foundation keeps churning out about how many Americans believes in ghosts and telepathy and clairvoyance and blah, blah, blah. And those numbers are abysmal. They're really bad. So we clearly do have a problem with a lot of people believing a lot of nonsense. I don't have to convince you guys of that. The situation is real. So value judgments are definitely called for here as long as they're backed up by reason. Now let me give you one example that you're probably familiar with. This happened a few years ago. Where in fact all these demarcation stuff became legally relevant in an important landmark case about intelligent design creationism. That is Judge Jones in Pennsylvania 2005. This was the famous case where for the first time intelligence at Dover, Pennsylvania where intelligent design came on trial. And some of the witnesses on both sides were scientists, though if you wanna call people that are engaged with the discoveries in scientists I think you're pushing the definition of a scientist. But nonetheless they were. And there were philosophers also that were called to help Judge Jones make up his mind about this thing. And if you read the final deliberation that Judge Jones delivered at the end of the trial it's very interesting. It's not that long, it's about 50 pages. And it is a very interesting study both in the legality of separation church and state and in the philosophy of science. Because in fact it was helped, the judge was helped by among other people two philosophers Barbara Forrest and Robert Pennock. And if you read the actual details of what Judge Jones said you will see that he says that I.T. on the point number one intelligent design violates the centuries old rules that discriminate between science and pseudoscience by or religion by invoking supernatural. That's the first point. The second point that he's making is that there are there's an argument of irreducible complexity and which is central to I.T. It's based on the same kind of illogical contrived dualism that characterizes standard creationism. And then finally the number three is well scientists have looked into these claims that actually found them wanting. The third claim is empirical. The judge looked at the scientific literature and said look actually as it turns out scientists have looked into these things the claims that you guys are making and they found them wanting. The first two on the other hand are inherently philosophical in nature. They are straight from the demarcation problem of debate. They're straight from the philosophy of science. They are about the nature of science. They are about what counts as science and what doesn't count as science. And there are legitimate disagreements among philosophers as well as about scientists. Jerry Coyne is my colleague in evolutionary biologists is here at TAM and he will be giving a talk tomorrow. And Ian and I disagree for instance on some important aspects of what counts as science and what doesn't count as science. The point that I wanna make however is that these debates are actually very relevant not just as intellectual debate, not just as meta philosophy, whatever that is. They're important because they actually have consequences for people. As a result of George Jones' interpretation of the current philosophy of science there is no intelligent design creation is being taught in Pennsylvania or anywhere else in the country because that landmark decision, 2005, that's exactly, still stands. In fact that decision has given such a fatal blow maybe not fatal but such a bad blow to intelligent design proponents that there hasn't been any legal challenge since. Every other challenge has been mounted on behalf of intelligent designers immediately the decision by Judge Jones has been brought into the discussion that was the end of the story. I don't foresee any time soon intelligent design coming back all the way up to federal courts let alone to the Supreme Court because of that decision. So that's an instance where thinking carefully about demarcation and taking seriously both what the scientists themselves are saying and what the philosophers are saying actually does make a difference for people and for our children who are now being taught better science than they were before. So this is pretty much what I wanted to say and I'm just on time. If you are more curious there is this book that is coming out it's not out there yet so you go and find it at the table it's coming out at the end of July it's available for pre-order on Amazon of course. It's by Chicago University Press it's a collection of essays by 30 something 35 philosopher, scientist and skeptics including Michael as I said and it's a series of analysis of the philosophy the sociology and the history of demarcation with implications for actual skeptic practice in practice day to day. Thanks very much for your attention. Masimo Pagucci.