Episode 64: What is a “Refutation”?

• Links to this episode: Spotify / Apple Podcasts

Unofficial, reviewed transcript. It may still contain mistakes; check the podcast when wording matters.

Speakers are labeled by first name where identified. Unidentified-speaker labels may refer to different people. Bracketed question marks indicate uncertain wording.

Transcript

[00:00:10]  Bruce: Welcome to The Theory of Anything podcast. We got Peter here, we got Cameo here somewhere, but I think she's away from the mic at the moment, so she might show up later in the conversation. We had an interview prepared today, and then we had to delay, so we thought we would just kind of have a chat. Peter actually said he was listening to our previous podcast episodes on Popper Without Refutation, and had some questions about that, and that will probably lead to other topics, so maybe we can start with that. So, Peter, go ahead and ask any questions you'd like.

[00:00:40]  Unidentified speaker: Okay.

[00:00:41]  Peter: Well, like you say, I was delving into the archives. I'd actually heard that episode before, but having spoken to you directly about some of the, what seems to me, a pretty unique take on Popper, but also seems to be at least how you explain it to be quite well-supported by what Popper actually said. I thought it might be helpful to go over that in maybe a little more succinct way, make kind of dial down what exactly is the difference between what you're saying, what most Popperians are saying, and try to explore that a little bit. First of all, perhaps we could start at the word refutation.

[00:01:31]  Unidentified speaker: What is it?

[00:01:32]  Peter: What is it to a regular Popperian? What does it mean to David Deutsch? And what does it mean to a Bruce Nielson?

[00:01:41]  Unidentified speaker: Okay, good question.

[00:01:44]  Bruce: So I think the word refutation is problematic because it comes with a lot of psychological baggage. And I've kind of argued that Popper being English as a second language, it may not have had the same psychological baggage to him as it would to a native English speaker. And so I think he chose a word, went with it. I think he had good reasons for why he chose the word. I think that when you're talking about logically speaking, it's like the right word. But the moment you try to move it into a scientific context, it takes on a different meaning to most native English speakers. And I think that the net result is is that you end up confused and that it would be perfectly natural that people would try to read Popper as saying something that he didn't actually say. And to make matters worse, they're not entirely wrong. Popper had this concept of falsification or refutation, which I'll explain in just a second.

[00:02:43]  Bruce: And then he also said, and you know, you can try to generalize this to the concept of criticism. A lot of times what people are trying to do is they're trying to take the word refutation and they're trying to generalize it, because it's just a natural thing to do with the English language.

[00:02:57]  Peter: Okay, so Bruce, here's what I'm seeing in the dictionary. The action of proving a statement or theory to be wrong or false.

[00:03:05]  Unidentified speaker: Yes.

[00:03:06]  Peter: Is that the same, that basic definition is...

[00:03:11]  Unidentified speaker: That is.

[00:03:12]  Bruce: People are assuming that's what Popper meant, and I'm arguing he didn't.

[00:03:17]  Unidentified speaker: He didn't.

[00:03:17]  Unidentified speaker: Okay.

[00:03:18]  Bruce: Okay, so what did Popper actually mean? Well, in full context, and here's the thing is Popper wasn't completely consistent. I've actually documented where he sometimes uses the word exactly like it's in the English Dictionary, but then later he will explain, well, yeah, actually, this is what I mean. And in full context, he does a pretty good job. But if you're just reading one of Popper's books, particularly if you're not reading The Logic of Scientific Discovery, where he probably does the best job of explaining himself, if you're reading like Conjecture and Refutation, let's say, there's a really good chance you just misunderstand Popper. I know I did. At least I did, right? And so, and I've met other people who I think clearly misunderstand him also. So when Popper uses the term refutation, what he really means is a counter example by experiment.

[00:04:09]  Bruce: Okay, so now this is maybe a little surprising, especially if you came to Popper through Deutsch, because Deutsch does not use the word that way, okay? When you talk about Popper saying, science is about conjecture and refutation, by the way, he always means science, empirical science, right? He's not talking about some generalized view. He occasionally dips into, here's how you might generalize this, but that just wasn't his area of interest, okay? He always means, I'm talking about scientific theories, they're empirical, they have universal laws. These universal laws are completely impossible to verify or confirm because they have infinite consequences. So because you can't go do an experiment that confirms all of the infinity of possible outcomes that you would need to to confirm this theory, instead, you're held to the idea of looking for counter examples, okay?

[00:05:05]  Bruce: And he called counter examples refutations. So a refutation is very specifically, I have an empirical theory, it's a scientifically empirical theory. It is something that has universal laws. It's impossible to actually ever confirm the entire theory because that would require infinite amount of testing. So instead, we're gonna make progress by trying to take conjecture different universal laws, and we're gonna look for counter examples to them, which he calls refutations, okay?

[00:05:38]  Peter: Okay, so the theory overall could be fine.

[00:05:41]  Bruce: The theory overall is fine.

[00:05:43]  Unidentified speaker: Popper's theory is fine.

[00:05:45]  Peter: There's still refutations to the theory in question.

[00:05:50]  Bruce: Okay, so let's say, let's use the example of... I may be getting the planets wrong. I think in my original talk or podcast, I used the example of Jupiter, and maybe I'm getting which planet was wrong. But I think it was that we knew about Jupiter and its orbit was off, and so it meant that there was another planet we didn't know about, which I think was Uranus. That would be a counter-example to Newton's laws of motion. Does that refute Newton's laws of motion? Of course not, okay?

[00:06:30]  Unidentified speaker: That would be silly.

[00:06:31]  Bruce: And so calling it a refutation, while it makes a certain amount of sense, it's just the wrong word. It's really just a counter-example. In this case, what it really meant was that there was a planet we didn't know about. So it didn't refute Newton's laws of motion in any meaningful sense at all. Now, it was a counter-example to Newton's laws of motion, though, like without a doubt, it was. Everybody knew it was a problem that needed to be solved. So what is it that Popper, so when Popper talks of refutation, what is he really talking about? If you really wanna think of it as a refutation in the dictionary sense, it is, but only if you think of it as a combination of the theory plus the background knowledge. So if you think of it as, look, it's a refutation not to Newton's theory, but to Newton's theory plus the assumption that there were x many number of planets, then yes, we can think of it as a refutation in the dictionary sense, but nobody thinks of it that way.

[00:07:34]  Bruce: When we talk about trying to refute a theory, we're talking about trying to refute the theory, right? We're not thinking in terms of the theory plus the background knowledge. So if what you think is that Popper's epistemology allows you to refute a theory by an observation, you're wrong. That is literally impossible. There is no single observation that could ever exist that would actually refute a scientific theory, ever. Now, Deutsch points this out in his paper, The Logic of Experimental Tests. And it was really, in part, Deutsch's paper and reading Popper and seeing the way people tried to use Popper that started to make me realize, oh my gosh, we're getting something wrong. Okay, we're misinterpreting something. So Deutsch is correct that... So Deutsch tries to use the word refutation to mean actually proving the theory wrong, which is a perfectly natural way to try to read the word refutation.

[00:08:32]  Bruce: It just doesn't happen to be what Popper meant by it.

[00:08:34]  Peter: Truthfully, it's kind of what I always thought it meant.

[00:08:38]  Unidentified speaker: That's good.

[00:08:38]  Unidentified speaker: Okay.

[00:08:39]  Bruce: So Deutsch says, well, you can't actually refute a theory with an observation. What you really need is you need a second competing theory. Now, if we're talking about scientific theories, which is what we're normally talking about with Popper, then Deutsch is correct. Now, David Miller criticized me bringing this up, and he said, no, Deutsch doesn't only means [spoken wording unclear] for certain types of scientific theories, but like you could refute the statement, all swans are white with a single observation of a black swan. Now, I need to come back to that because like he's not really wrong with his statement, but it's actually a misleading statement in many ways. Deborah Mayo says, whenever you see Popperians trying to use examples of how you can refute a theory with a single observation, they always use examples that will only ever come up in philosophical circles.

[00:09:39]  Bruce: They'll never come up in actual scientific studies.

[00:09:44]  Unidentified speaker: And she's right.

[00:09:45]  Bruce: And there's actually a good reason for why that is. And let me come back to that. The key point I wanna make here though, is that at least when it comes to actual scientific theories, which are complex, they're not little tiny simple things like, all swans are white. Which by the way, isn't even a good scientific theory anyhow, because it doesn't attempt to explain anything, right? The only thing it can be said to explain, is it can explain why every swan you happen to have seen up to this point is white. That might make it an explanation. Maybe that even makes it a scientific theory in a sense. It surely isn't the type of theory that a real scientist is going to care about. And honestly, we all understand evolutionary theory at this point. We know that even if all swans did happen to be white, a mutation might change that in the future. So we still know it's not the type of universal law that a scientist is ever going to care about, right?

[00:10:40]  Bruce: So they're not great examples. Once you realize this, you realize that Deutsch uses the word refutation to mean specifically they try to refute the theory, whereas Popper actually meant it as refute the theory plus the background knowledge. And I give tons of quotes to prove that he actually did mean that. Like he says this, right? You have to have the right quote. And that's part of the problem is that he says it, but then like it's in passing and then he never mentions it again. And then he even uses it more like the dictionary. He often acts like you could refute this theory, okay, with an observation. So he's a little inconsistent, okay? But if you actually look at everything he says, you can absolutely find proof that when Popper talks about refutations, he actually means you're refuting the entire theoretical system, not just the one individual theory.

[00:11:30]  Bruce: Okay, so I was talking to Danny Frederick, who's a now passed away, but respected philosopher of critical rationalism, who by the way, believes some things that I think are just not right. And I criticized him on other grounds. The main one being that he thinks that the goal of scientific inquiry shouldn't be to find truth, which I think is just false. I don't even think critical rationalism can survive such a critique. But that's like a totally different story. In a discussion with him, him and I were arguing, and I started to realize that we were talking past each other. I never did successfully convince him we were talking past each other, but I became convinced that we were talking past each other. And I realized I was using the word refutation to mean actually refuting the theory, whereas he was using the word refutation to mean refuting the theory plus the background knowledge.

[00:12:29]  Bruce: Now, interestingly, Frederick has a different term for what Deutsch calls a refutation, where you try to refute the theory itself. He calls that rejection of the theory. So when Frederick uses the term refutation, he means you have a counter example to the theory, but what you actually refuted was the theory plus the background knowledge. Whereas once you then later have a second theory that explains the success of the first theory, then you reject the older theory. So once you have general relativity from Einstein, you can... So we used the example of Jupiter. Let's use a different example now. The example would be perihelion of Mercury. So that's the same thing as the Jupiter example. You've got this motion of Mercury that doesn't match Newton's laws. And so the same thing happened. They initially thought we must have missed a planet. So they calculated where this planet, which they named Vulcan, was supposed to be.

[00:13:28]  Bruce: That's where the term Vulcan comes from. And they then went out and looked for it, and they could not find it, which was the opposite of what happened in the other case. Now, in this case, now you have this counterexample, but the number of planets you counted still looks to be right. You've tested the possibility that the number of planets was wrong, and you failed to find the other planet.

[00:13:55]  Unidentified speaker: Okay?

[00:13:57]  Bruce: So in this case, we now understand that now that we have Einstein's theory, and we have general relativity, that that was actually the problem. The problem was that Newton's laws of motion were actually wrong. And that's why Mercury doesn't follow the Newton's laws of motion. But you can't really know that until you have this other theory that explains what's wrong with Newton's laws of motion. And up until you have Einstein's theory, you just don't know what... You don't have no way of knowing if the problem was with Newton's theory, or if the problem is just something else that you're missing. And so because of that, you can never really refute a theory, or as Frederick would put it, you can never really reject a theory until you actually have a second theory. That was what I realized. Okay, Deutsch is calling refutation what Frederick is calling rejection.

[00:14:53]  Bruce: And what Frederick's calling refutation is what Deutsch calls a problem to solve. Okay, they're using different terms, but they're saying exactly, conceptually, the same thing. Okay, now when Frederick explained this, and I realized we were talking past each other, I tried to convince him we were talking past each other, and he just, he doubled down. He was unwilling to accept that possibility. As typically happens with conversation on the internet, the conversation just sort of died, and there was never a resolution. But I became really curious, and I thought, is it possible Frederick is actually right that Popper is using terms the way he does? Frederick, by the way, admitted the term rejection didn't come from Popper, that that was one he made up himself. So does Popper possibly use the word refutation the way Frederick does, as meaning the theory plus the background knowledge, refuting the combination?

[00:15:44]  Bruce: So now that I knew what to look for, I started looking up in Popper, and sure enough, Popper absolutely clarifies that. And that was when I realized, okay, I had been misunderstanding Popper this entire time. I thought every time he used the word refutation, he actually meant refuting the theory itself.

[00:16:03]  Unidentified speaker: And in fact...

[00:16:05]  Peter: Just to clarify, when the Deutsch version of refutation, where he's talking about the theory, does that more align with a mainstream Popperian take?

[00:16:17]  Bruce: Okay. That more aligns with the mainstream use of the word refutation.

[00:16:25]  Unidentified speaker: Okay.

[00:16:25]  Bruce: Separate from what Popper means by it.

[00:16:29]  Peter: Okay, but what you're saying is that you're making the case of what Popper actually meant, but then say like maybe a David Miller... Does he agree more with Deutsch? Or is it a third interpretation?

[00:16:43]  Bruce: Here is what David Miller said when I actually presented this. He said, well, Deutsch is talking about scientific theories, and that might be true for theories of chemistry or physics or something like that, but it's not true for all theories. So Miller, who really is probably the greatest living Popperian, is trying to argue a sort of middle ground. He's trying to say, no, actually, there are some theories you can refute with a single observation. And the stereotypical example is All Swans Are White. Now, I'm not saying Miller's wrong, because from a certain point of view, the statement, All Swans Are White, clearly can be refuted by the statement, I have a black swan here. But you just have to stop and think about it for a moment. And you almost immediately see that this middle ground doesn't really exist like you think it does, okay? Let's say that you and I thought that there were no white swans. [Wording as spoken.]

[00:17:40]  Bruce: And then we go to Australia and we see a black swan walk by. Does the observation, what observation actually exists?

[00:17:47]  Unidentified speaker: Does the observation exist?

[00:17:49]  Bruce: I see a black swan there. Did that happen just because we saw a black swan walk by? No, of course not, because neither of us are experts in swans. We would see a black bird walk by that looked swan-like, and then we would have no idea if that was actually a black swan or not. We would want, we might think it is. We might say, wow, that looked like a black swan.

[00:18:09]  Unidentified speaker: Someone could have spray painted this swan black.

[00:18:12]  Unidentified speaker: Right, exactly, okay.

[00:18:14]  Bruce: So we would probably wanna go like capture the swan. We'd probably wanna see if it's painted. We'd probably then go, what if its DNA is like a totally different bird? And it just happens to look like a swan because of convergent evolution. So we'd probably wanna go find an expert in birds or something, and we'd probably wanna say, maybe an expert in genes. It would probably wanna say, look, is this actually a swan? After we went through all those tests, yes, at that point, the observation, here is a black swan now exists, but you did not get there through a single observation. You got there through a number of different tests of trying to test out competing possibilities. That's actually the answer to what David Miller was saying, is that, yes, from a certain point of view, here is a black swan refutes the statement, all swans are white, but that observation doesn't exist like that.

[00:19:09]  Bruce: An observation is you and I just sort of see a black bird, and then we have to start testing our alternate theories. Was it a painted bird, was it a painted swan, was it? We have to go test that. This is actually the answer to the whole question, is that observations are always theory impregnated. Popper's famous for saying that. People haven't thought through though, what that means in the case of using an observation to refute something. Is an observation statement assumes you've done certain kinds of tests to get to that observation statement.

[00:19:43]  Peter: So the thought experiment is misleading in a way. The whole black swan, white swan thing is, it's not really how people make theories about the world.

[00:19:54]  Bruce: No, it's not. But it also explains why it's completely unproblematic. We could be skeptics. We could say, look, an observation can't refute a theory, which is what people say about Popper. That's like the big criticism, I quote tons of people criticizing Popper, showing that you cannot with an observation refute a theory.

[00:20:17]  Unidentified speaker: And they're right.

[00:20:18]  Bruce: But it does not mean a thing because you have to understand Popper not as literally meaning an observation refutes a theory, but as meaning an observation creates a problem that allows you to start conjecturing how to solve that problem. And you can go out and you can test the different parts of the theoretical system. You can test, was my guess about the number of planets correct? You can test, you know, you can test anything, right? You start to do tests. You start to severely test the different parts of the theoretical system. And at some point, you just cannot find the other planet that you expected to see there. And you start to wonder, is it possible that Newton's laws are wrong? And do I have to actually challenge Newton's laws? I never have to actually refute Newton's theory, per se. I simply have to get to the point where I can't solve the problem any other way and then happen to come up with someone who happens to come up with a good conjecture, general relativity in this case, that actually does solve the problem.

[00:21:19]  Bruce: That will eventually now become the premier theory. We never truly refute theories. Like Newton's theory never really went away, right? We even still use it as at least an approximation of other theories. But even if we didn't use it as an approximation of the theories, what really happens is the theory just sort of dies out because it stops being productive compared to its competitor. And that's what Popper's epistemology really gets you. It objectively creates problems through observation that then can be solved through testing, and through conjecture and testing, okay? Once you realize that, you realize that the skeptic's view is wrong, but also the standard way of explaining Popper's epistemology is misleading. It's not even that complicated, right? I just explained it to you in a few minutes. It's not hard. And really, if you just started using the word counter-example instead of refutation, you would almost assuredly get people understanding you better, right?

[00:22:18]  Bruce: And even for the cases where you literally can refute a theory with a single observation, like all swans are white, even then thinking of it as, well, actually refuted the theory plus the background knowledge is really useful because then it encompasses the idea, we should check to see if this is a dirty swan or a painted swan or something like that, right? That's something you should absolutely test, okay? So it's always a good idea to think of it as a Popperian refutation is a refutation of the theory plus the background knowledge. And even in cases where you can legitimately argue that isn't the case, it was still productive to think of it that way at some point in time, right? So I'm going to just go on a limb here and I'm gonna say, number one, a Popperian refutation is always, if you're thinking of it as a single observation, is always a refutation of the theory plus the background knowledge.

[00:23:12]  Bruce: It is never just of the theory. If it is of the theory, the reason why is because you've done a series of tests prior to that point so that the observation statement now contains those severe tests as well. You've actually tested alternative theories, just not the theory proper. You tested, did my instruments work? You tested, is this a painted swan? You tested, did my number of planets actually, did I count them correctly?

[00:23:38]  Unidentified speaker: Did I miss one?

[00:23:39]  Bruce: Okay, you see what I'm saying is that it's never just the observation. It's always the series of tests you've done. So if you just rethink of it as a Popperian refutation is a refutation of the theory plus the background knowledge, you can never actually go wrong. It's always a smarter way to think of it.

[00:23:56]  Peter: Okay, so you're saying that that's a Popperian refutation, but it's not actually what how Popper conceived of a refutation.

[00:24:05]  Bruce: Well, see, he did, but only if you look at everything he said. If you look at individual statements, you can find cases where he acts as if a refutation is just a single observation that refutes a theory.

[00:24:17]  Unidentified speaker: I see.

[00:24:19]  Bruce: So you have to kind of say, am I talking about any one particular quote from Popper? Am I talking about Popper as a whole?

[00:24:25]  Peter: Well, I think that's what I found very convincing about your last podcast on this, how you really went through the actual text where Popper was talking about it, and you had some pretty strong examples there.

[00:24:38]  Unidentified speaker: Yes.

[00:24:39]  Bruce: They're not that hard to find. That was the thing that surprised me. You know, it's funny, the debate I had with Danny Frederick on Facebook led to all this, right? Him criticizing my views, ultimately misunderstanding what I was saying, actually led to me clarifying my own views and realizing that I had in fact misunderstood something, even if it wasn't the thing that Danny and I were actually debating. It's a really good example of how criticism is just valuable, right? But okay, so now let's go back. So I think I've now explained, basically I'm claiming everybody's getting something wrong. Okay, it's Popper had the correct idea overall, but he sometimes delved into ways of thinking where you could use a single observation to refute something. So I think he sometimes was wrong, right? I think you could make the argument, well, in context, he's making a certain point, and I think that's always true.

[00:25:46]  Bruce: It's impossible to always say the correct thing because you're always trying to narrow your language for a specific point.

[00:25:53]  Peter: I guess that relates to also what I was wondering. So Deutsch has his take, it's a little different, Popper varied, and exactly how he used these words. Maybe David Miller has a middle ground. In the real world though, doesn't really matter that much.

[00:26:10]  Unidentified speaker: It doesn't.

[00:26:11]  Peter: Pretty easy to determine what someone means in actual communication.

[00:26:18]  Bruce: Here's where the problem comes. So David Deutsch, he has written two very excellent books that include a significant amount about Popper. But he never at any point uses the word refutation in the way Popper actually does. And in fact, he downplays the idea that we even particularly care about the demarcation that Popper was talking about. Because correctly, he points out that we can generalize Popper. We can think of it not just as, hey, I need an observation to refute this theory. But we can say, hey, the vast majority of theories, we criticize them. And if it doesn't survive criticism, never mind whether it's an observation or not, that theory will die. And basically what he, what Deutsch does is he generalizes Popper to the idea that you have a number of competing theories, and you criticize the theories, and the one that survives is your best theory.

[00:27:24]  Unidentified speaker: Okay.

[00:27:25]  Bruce: So now Popper isn't against that. You can find places where Popper talks about such a generalization. So this isn't an idea that Deutsch made up, lock, stock, and barrel on his own. He's getting it from Popper, okay? What Deutsch is adding to it, what he's bringing to it, is he's saying this is actually a better way to think of Popper's epistemology.

[00:27:45]  Unidentified speaker: That's not true.

[00:27:46]  Bruce: And I'm going to explain why. It's not that that David Deutsch is wrong. You can generalize Popper in that way, and that's actually a completely fair point. But when you say it in that way, when you say, oh, all criticisms, we're open to any criticism. An observation is just a special example of a criticism. And when you do that to downplay the demarcation between science and non-science, or as I prefer to say, between empirical theories and non-empirical theories. When you do that, you have lost something incredibly significant out of Popper. So significant that in some ways, you're no longer doing Popper's epistemology. Here's what you lose. Popper, the genius of Popper was that he came to his epistemology in terms of the question of what counts as science, which keep in mind means empirical theories. I've argued science in a more general sense includes metaphysical theories, and Popper never really denied that.

[00:28:49]  Bruce: He usually, when he says the, talks about the demarcation, he doesn't usually speak of science versus non-science. He usually speaks of empirical science. He's usually pretty specific. I found a couple of counter examples to that, where he says science instead, and there you've got some potential for misunderstanding, I admit. But when he talks about this, what Popper was really trying to do, what he was trying to work out, what led him to trying to solve the problem of induction, was these theories out there that claimed to be science that weren’t. Freudian theory, communism, things like that. And keep in mind, Popper was a communist at one point.

[00:29:30]  Unidentified speaker: These were theories that he found very tempting.

[00:29:35]  Bruce: So he really gave some thought to the question, what makes empirical theories special? And when he started thinking through what makes empirical theories special, now that's naturally the demarcation criteria, is, hey, I've got certain kinds of theories that are special because they're empirical and other kinds of theories, which aren't bad. They may be true. They may be very important, but they lack something. What is that thing that they lack? This is what Popper was trying to actually get at. When he talks about, now remember, a refutation to Popper is always an observation, not just a general criticism. This is true. I cannot find a single counterexample to where he uses the word refutation specifically as a kind of observation, all right? If maybe somebody else who's read more of Popper than me can give me an example, but like I've gone through a whole bunch of books and I just can't find counterexamples.

[00:30:31]  Bruce: He really seems to have been pretty consistent on this one.

[00:30:34]  Peter: The whole empirical versus non-empirical theory, is that Popper's language?

[00:30:39]  Unidentified speaker: Yes.

[00:30:40]  Unidentified speaker: Specifically?

[00:30:41]  Peter: Is that confusing though, because part of what his thing was, or is that just Deutsch, but to criticize empiricism with the whole theory-impregnated approach.

[00:30:55]  Bruce: As it turns out, empirical and empiricism aren't the same thing.

[00:31:00]  Unidentified speaker: Oh, good point.

[00:31:01]  Bruce: However, fans of David Deutsch, who have just read David Deutsch's books, but haven't tried to research this deeply, don't understand those aren't the same thing. And so they get tied in knots over this, and it's really bad at times, particularly on Twitter, where they just really don't get it.

[00:31:20]  Unidentified speaker: I see.

[00:31:21]  Bruce: So, empiricism is a specific bad philosophy. It's the philosophy that our senses can't be wrong. Okay, so it's an absolute source of truth. That is obviously not something Popper would agree with. Okay, because he didn't believe there were any absolute sources of truth. But Popper was absolutely arguing empirical theories are special. They are important. They are distinct from non-empirical theories in important ways. He was absolutely arguing that. So Popper's epistemology is deeply, in fact, exclusively empirical.

[00:32:02]  Unidentified speaker: I see.

[00:32:02]  Bruce: Okay, at least the version he writes in Logic of Scientific Discovery. As I mentioned, he does mention how you might generalize it. So if you want to talk about a generalized version of Popper, you'd be right to say it's not always empirical. Okay, but in terms of what he was writing about in his book, it was exclusively empirical theories he was talking about. Okay, and he makes this painfully clear when you read his book. Okay, and when you read the Logic of Scientific Discovery, if you read, say, Conjecture and Refutation, it's nowhere near so clear. Okay, I got all the way through Conjecture and Refutation without even noticing that because I hadn't read Logic of Scientific Discovery yet. And so I hadn't read the more clear statements that he makes. Okay, so what is it that makes a theory empirical? Basically, what makes a theory empirical?

[00:32:52]  Bruce: So think about how we theorize things. Now, Sadia's husband, Mark, he has offered a criticism of Popper that is actually kind of correct. Okay, although it's more a criticism of the Deutschian fan version of Popper than Popper himself. He would say, you know, I don't think I really buy there's anything special about Popper's epistemology. And keep in mind what he really means is the Deutsch fan version of Popper's epistemology. But to him, that's what Popper's epistemology is, because that's how he has interacted with him, right? He'd say, you know what? People have forever criticized things. And it's not like you can do something different than have a set of possibilities that you've conjectured, and then you criticize them and you end up with whatever's left. People have been doing that way before Popper. Popper hardly discovered that. And science was around long before Popper was.

[00:33:51]  Bruce: And so I don't really understand what's special about Popper because all he's really doing is describing what we all know we all do anyhow, okay? Here is what Mark is missing. Popper was about empirical theories.

[00:34:05]  Unidentified speaker: That's what he's missing.

[00:34:07]  Bruce: Yes, you can generalize it like Deutsch does to non-empirical theories. But when you do that, you've lost the importance of empirical theories. Here's what Popper said. Popper said, look, when you have a scientific theory, how do you do an experiment? Well, to be able to do an experiment requires that your theory locks down some sort of prediction that happens within some reasonable spatial temporal area, right? It has to happen within a reasonable period of time, and it has to happen within a reasonable locality that we can actually verify the outcome, okay? If that isn't true, then it's not empirical. Now, this is an incredibly sensible way to think of the term empirical, okay? To basically, to equate it with it, being able to do an experiment, okay? And I don't think there are many scientists who would argue with Popper over this, that empirical means able to do an experiment, okay?

[00:35:10]  Bruce: I will give you some possible counter examples to that, because I've attempted to criticize this. And I went out and found people criticizing it and took their criticisms. I will give you a counter example of that in just a second. But I think the vast majority of scientists would say that is a completely reasonable way to look at the word empirical, is that this theory says something about real life where I can actually perform an experiment, okay? Now what Popper is pointing out is that to be able to have it be constrained to a certain spatiotemporal area implies that the theory has some sort of universal law that constrains the outcomes. And that just logically speaking, just logically speaking, that implies that what we're really going to be—experiment—doing with the experiment is looking for a counter example. And the reason why is very simple.

[00:36:02]  Bruce: If I go out and I have a theory, and this theory has infinite empirical content, and I go and I confirm it, say 10,000 times, there's still an infinity of content that I can't confirm.

[00:36:14]  Unidentified speaker: Okay?

[00:36:15]  Bruce: So what I really want is I want to try to find a single counter example. Well, as it turns out, that's exactly what we do with experiments. And you really just cannot find counter examples to this because that really is what we do with experiments. When I say I'm gonna go test Einstein's Theory of General Relativity using the Arthur Eddington Expedition, and I'm going to go, quote, verify the theory of general relativity, what I'm really doing is I'm demonstrating that in that one particular case, it matches the stars in an eclipse, match what Einstein's theory said, and don't match what Newton's theory said. Now, I may word that as I verified Einstein's theory, and I'm not even wrong. In a certain sense, I am clearly verifying Einstein's theory, but nobody in their right mind would think I have now verified every possible outcome of the entire infinity of content that exists in Einstein's theory.

[00:37:17]  Bruce: In fact, in many ways, we're painfully aware of that. There's almost immediately a question, okay, sure, this experiment came out in favor of Einstein's theory, but does that really mean his theory is true? And we know it doesn't mean that, right? It really just means it beat Newton's theory. So Popper would say, look, what really mattered here was that you refuted Newton's theory, that you had a counter example to Newton's theory. Therefore Newton's theory now has a problem that Einstein's theory doesn't, that Einstein's theory even explains why it's a problem for Newton's theory. We now have a fairly clear-cut case that general relativity is something worthy of looking at further. In reality, it took decades between Arthur Eddington's expedition and science actually accepting Einstein's theory as better than Newton's. Part of that was just social stuff like Thomas Kuhn says, but part of it was because they couldn't figure out how to do better tests, and they really wanted to more severely test it than that.

[00:38:22]  Bruce: This one test, which by the way, at the level of technology that existed back then, wasn't really even a great test because everything was kind of out of focus, it was hard to know how to measure them. I mean, it wasn't the great test that we try to make it out today. In fact, you can put it into probabilistic terms. And we weren't, according to Deborah Mayo, we weren't even talking about a p-value of 0.05, right? I mean, it was really a fairly questionable outcome of the test. So it makes sense that we had to wait for technology to catch up before the theory was really going to be more strongly testable. Now, eventually they invented radioscopes. Deborah Mayo explains this, and I'm doing this off the top of my head, so I may be getting this wrong. They invented better ways to test the theory. And once we had much better ways to test it, all of them just landed right directly in Einstein's theory and against Newton's.

[00:39:19]  Bruce: And at that point, we drop Einstein's theory. [Wording as spoken.] It's now considered a rejected theory, although we still use it as an approximation. And we start to accept that Einstein's theory is the, quote, correct theory. Even then, have we proven Einstein's theory correct?

[00:39:34]  Unidentified speaker: Well, no, of course not.

[00:39:34]  Bruce: And no scientist would believe we did. When we say it's a correct theory, what we really mean is it's more correct than Newton's theory, right? So let me wrap what I just said together. We may speak in the language of verification. We may speak in the language of we verified this theory or we confirmed this theory, but we always, in context of empirical scientific theories, we always really mean we refuted the competitors, okay, by observation. The competitor made a prediction that didn't come true. This is what Popper noticed. And he's correct. No scientist would really argue this. Now, what arguments could you make with Popper on this? Well, one of them would be string theory. You could say, well, string theory makes predictions like the existence of super partners. But it doesn't tell you under what circumstances to expect those super partners.

[00:40:36]  Bruce: So we make bigger and bigger colliders and we may someday find a super partner. And when we do, it's gonna be a big deal. It's going to be something that really revitalizes string theory and gets us thinking harder about it. It will in some sense actually verify that something's correct about string theory. And yet you can never actually refute string theory because string theory, if you don't find the super partners, then the argument would just be, well, we need a bigger collider. You could in theory maybe need a galaxy-sized collider to be able to find the super partners, right? So you can't actually refute string theory, and yet it is still empirical because you can test that the super partners exist at least in principle, if not in practice, okay? Now what Popper would say to this, if Popper's here, if you're a Deutschian, I don't know what you say to that, because this is like a completely legitimate argument.

[00:41:38]  Bruce: If you're a deep Popperian, and you really understand Popper's argument, your argument is, yeah, you're right. That's an example of a theory that can only be verified, which is what we mean by non-empirical, right? Nobody really doubts that you can call that empirical if you want. You can say, yes, it's quote empirical because in theory we can test it, okay? But what experiment are you actually gonna perform? There's no experiment you can do that comes out with a definitive answer. That's the sense in which it's not empirical, is that you can't do an experiment that's definitive. If you really, really, really want to call it empirical, then fine, okay? Because we're not really arguing over words, we're arguing over concepts. If you really wanna say string theory is an empirical theory because in theory you could find the super partners and that would confirm or verify at least that one aspect of the theory.

[00:42:34]  Bruce: So therefore I'm going to call it empirical, then great, that, you know what? Maybe people even sometimes use the word empirical in lame terms [wording unclear] to mean that. Here's the problem with calling that empirical scientifically speaking. The following theory must now be considered scientific. Bigfoot exists, unicorns exist, because as it turns out, there's no way to perform an experiment and yet we may someday find a unicorn or we may someday find a Bigfoot. Popper points out you can't even claim it's improbable. If you try to take into consideration the entire universe rather than just the world, there could be a Bigfoot that evolved on some other planet out there in some other galaxy, right? So just in terms of practicality, science will never call a theory based around the existence of something rather than a universal law constraining something.

[00:43:29]  Bruce: It will never call those scientific theories. That's Popper's point. He's not making a different point. He's not trying to say something normative. He's not trying to dogmatically define a term. He's just trying to be practical. He's just trying to say, look, scientific theories will always be about experiments. Experiments will always be about constraints, and constraints, you only test them by violating them. Okay, that's Popper's epistemology in a nutshell.

[00:43:59]  Peter: Can I ask you, is this interpretation of Popper that seems pretty convincing to me, but is it controversial amongst these same kind of Popperians that we're pushing back on about refutation?

[00:44:14]  Bruce: Okay, let me get to that.

[00:44:16]  Unidentified speaker: So here's the question, though.

[00:44:18]  Bruce: Let's say that you buy my argument that that is what Popper meant by refutation. We could still take Deutsch's version of Popper and we could still say, but isn't Deutsch's version of Popper still correct, right?

[00:44:35]  Unidentified speaker: Sure.

[00:44:36]  Bruce: So it's tempting to say, and fans of David Deutsch on Twitter in particular, use the word refutation to simply mean, I have personally criticized this theory and I subjectively found a criticism that I subjectively found convincing. So let's take some real life examples. So Dennis Hackethal, who's been on this podcast, who I'm friends with, he considers himself a critical rationalist, but he's famously put up an article on how he knows that the best theory is that animals don't feel things. They have no qualia, they're just meat robots, is what he calls them. Now, I don't want to get too far into this. Let me just say that the arguments he makes are entirely outside of Popper's epistemology completely, if you understand Popper's epistemology as being about empirical tests. What he actually does is he, the arguments he uses are as follows.

[00:45:34]  Bruce: He says, well, let me, I can give you example after example of animals that don't understand things. You know, you've got the squirrel that you put it on the concrete and you give it a nut, and it goes through its little motions of trying to dig into the concrete and then tries to go through the motions of trying to bury the nut, even though there's nothing there for it to bury with. So this is a clear example of an animal not understanding something. Now you might ask, what's that got to do with them feeling anything? Which, by the way, would be a really good question. Dennis argues in his article, as far as I can make sense of his article, that we know that like robots today don't feel anything. So if an animal is acting like a robot, we ought to conclude that it's not feeling anything.

[00:46:19]  Peter: And this is on his blog?

[00:46:21]  Bruce: This is on his blog. Now, I think what this argument is, is it's an intuition pump. It makes you see animals' behavior as robotic, and it makes you try to associate them with robots, which you know don't feel anything. Now, in his mind, he has, quote, now refuted the theory that animals feel things, because in his mind, if animals really felt things, they shouldn't act like robots. At every level, this is a terrible, terrible Popperian argument. Like, it's the opposite of critical rationalism. First of all, it's induction. He's literally trying to say, I can give you an example of animals behaving automatically, and I'm therefore able to conclude and generalize to the idea that absolutely everything that animals do, they show no understanding. And I'm trying to generalize to just like, analogously, a robot doesn't feel things and doesn't understand things and has automatic movements.

[00:47:25]  Bruce: Therefore, you should conclude that animals are the same way. Well, why?

[00:47:32]  Unidentified speaker: Right?

[00:47:32]  Bruce: Couldn't it be the case that animals have automatic movements but understand some things and just have some automatic movements? If I knew for sure a squirrel was entirely automatic movements, would that tell me that an ape is entirely automatic movements? If I knew for sure that all animals used automatic movements, would that actually allow me to make an analogy that I therefore can, just because there's this specific case, robots have automatic movements and feel nothing, that I can therefore generalize to animals make automatic movements and feel nothing? There's nothing about this argument that makes rational sense from a Popperian standpoint. None of it survives critical rationalism. It is the opposite of a critical rationalist argument. In his mind though, because he came to critical rationalism through Deutsch, this is a solid critical rationalist argument.

[00:48:24]  Bruce: And the reason why is because in his mind, he thought through what he thinks should be an implication of the opposing theory. And then he came up with what he just subjectively felt like was a good argument. In his mind now, he's now criticized that argument. He's, quote, refuted it. And therefore, his argument is the only one that's remaining standing. Now, the reason why I keep emphasizing the word subjective is because that's what we've got here. I don't have any way to test any single aspect of his argument empirically. I don't have any way of knowing if just because robots don't feel things, animals don't because they use automatic movements. I have no way of knowing that just because a squirrel uses automatic movements, that therefore an ape is using automatic movements. For that matter, I'm not even using some of the arguments that are even better.

[00:49:17]  Bruce: Like, for example, the fact that humans use automatic movements. Like, we know that, right? Well, he acknowledges that in his article. He says, well, it's true, humans use automatic movements, but humans can change their automatic movements because they have understanding. So basically even your counter example that you're trying to use, you know doesn't apply to humans. The one example that we have. So there's just no reason to believe that the existence of an automatic movement therefore implies a lack of qualia. Humans have automatic movements, and they have qualia. So nothing about this argument works as a critical rationalist argument.

[00:49:56]  Peter: Meaning empirical argument?

[00:49:58]  Unidentified speaker: Yes.

[00:50:01]  Peter: So it's more of a non-empirical, metaphysical, philosophical theory?

[00:50:08]  Unidentified speaker: That is correct.

[00:50:09]  Peter: Are those all kind of the same?

[00:50:11]  Unidentified speaker: Yes.

[00:50:12]  Bruce: For the sake of argument, I'm not saying he's wrong. Maybe animals don't feel things. I don't know, right? What I'm assessing is the argument itself, that the argument is purely subjective. He's decided, this sounds like a good criticism to me, therefore, I have now refuted the theory, and I only have one theory remaining, and it's mine.

[00:50:32]  Unidentified speaker: If people can do...

[00:50:34]  Bruce: Now, this answers Mark's question. This is how people actually argue in real life, right? They may be using the Deutsch generalization of Popper. They're criticizing the other theories, and they're showing that their theory is the best one.

[00:50:50]  Unidentified speaker: Everybody does that, right?

[00:50:52]  Bruce: Even people who know nothing about Popper, that's just how we argue. There's no other way to argue. The mistake that's getting made here is that they're trying to act as if the existence of this subjective criticism that they feel very strongly about personally is somehow equivalent to a refutation. And it's not, right? Because a refutation is an actual observation that we can test that came out of an experimental test, the Popperian concept of a refutation, which is why I believe we should call them counter-examples instead of refutations. Because it's very natural for Dennis to think that the word refutation could just be a generalization that means any kind of criticism, even including just subjective ones that he personally feels very strongly about. That's not an unreasonable way for him to read the word refutation. It just isn't what Popper meant by it.

[00:51:47]  Bruce: Okay, so why does this matter though? Because we might argue it doesn't matter. Sure, Popper meant an observation, but an observation is, as Deutsch says, just a special kind of criticism. Okay, or just a certain kind of criticism.

[00:52:02]  Unidentified speaker: Yeah, it is.

[00:52:03]  Bruce: It's an objective kind of criticism. None of Dennis' quote refutations, his criticisms, none of them are anything but subjective. They are feel-good, pump-your-intuition types of criticisms, okay? Which once you're outside the realm of empirical theories, is what you mostly rely on, right? And so this brings us to a paper from Bruce Caldwell called Clarifying Popper, which I absolutely recommend people read. And he's trying to work out, how would you apply Popper to economics? Because economics famously can't be refuted by observation. And the reason why in this case, a little different than the case I'm using with Dennis, it's because there are so many forces at work in an economy, that we don't really expect any of our economic theories to make good testable predictions. Even if I know for sure that demand goes up and therefore prices go up, and that then creates additional supply.

[00:53:05]  Bruce: I totally buy that, right? Like I totally think that's a true economic principle. But it's dang hard to test inside of an economy, and you can almost always find counter examples to it. And then it's just because there's some other reason. There's some other force at work that was overriding that situation, whatever it was, right? I could make up an example if you want, but it should be easy to think of counter examples. Economies are complex things. We know that just because demand goes up, it doesn't necessarily follow that supply immediately follows. There could be some reason why it doesn't for some reason. Eventually, in the long run, it should. But for any one particular prediction, you could just be wrong, because there's something else causing a problem in the economy at the moment, right? Because of that, it's really hard to figure out how to apply Popper's.

[00:53:56]  Bruce: In fact, Caldwell argues, it's impossible to apply Popper's epistemology, by which I mean the empirical epistemology, not the Deutschian more generalized version, to economics. That you can't is fine, because you can then still criticize it. You can still then jump to the Deutschian generalized version, and you can try to still correct errors by other means through criticism. But you have lost something special. You have lost the ability to actually create experiments, and by experiment, work out which are the best laws. Now, why does this matter? And it's very simple. I wish I could find the quote. I don't have the quote handy, but he explains it very simply. He says the issue here is that while we can all agree with Popper that you should criticize theories and you should pick the best one, it's not at all obvious which criticisms are the better ones and which ones are the worst ones.

[00:54:59]  Bruce: And it turns out that's largely subjectively a matter of opinion. That's not true for empirical theories, for actual outcomes of experiments. And remember, in Popper, we're talking about repeatable outcomes, okay, not just some throwaway outcome, where everybody can go off and they can create the exact same criticism, the exact same reputation [apparently, refutation] of the theory through the exact same experiment. And they can all just for themselves go out and see, yes, this theory makes this prediction, and this experiment is a counter example to it.

[00:55:32]  Peter: So economics is a non-empirical area of human inquiry or whatever.

[00:55:41]  Bruce: So I think it might be more accurate to say it's less empirical.

[00:55:45]  Unidentified speaker: Less empirical, okay, yeah, that seems fair.

[00:55:48]  Bruce: So because like, for instance, you could almost assuredly take something like supply and demand, create a really simplistic experiment where there's no other forces at work, and you can probably show it actually is true, right? And so I don't think it's actually accurate to say economics has got zero empiricalness, because there are certain types of experiments you can do that you can work with. But they're not super convincing in the case of economics, because what we really care about is full economies, not simplified experimental outcomes.

[00:56:23]  Peter: Yeah, I guess that's one thing that people say the difference between Austrian and Chicago is that Austrians pretty much know it's a philosophy. They're advocating where the Chicago school with Milton Friedman try to make it into more of a science.

[00:56:41]  Bruce: That's right. Which is really hard in the case of economics.

[00:56:45]  Peter: Yeah, and some people say impossible.

[00:56:48]  Bruce: Okay, but this is Caldwell's point, and this is in essence what Popper was answering. The reason why empirical theories are special is because they're the only case where we have criticisms that lack the subjective element and where everybody can objectively look at the exact same criticisms and can all agree, yes, this is a problem. Okay, the moment you get outside of empirical theories, yes, of course, you should use criticisms on them and you should try to improve them through criticisms. Deutsch is completely correct. The mistake is not recognizing the importance and the specialness and why empirical theories are special, why they are honestly better, because they have this special condition where we have objective criticisms of them that we can all agree upon. Now, someone might argue with me a little here. Well, what about the fact that experiments aren't always repeatable?

[00:57:49]  Bruce: Well, if they're not, then that's a valid criticism of that experiment, right? We're talking about when it is repeatable. Or you might say, oh, what about the case that people get different outcomes from an experiment?

[00:58:02]  Unidentified speaker: Well, okay, that's a problem.

[00:58:04]  Bruce: That's a valid criticism of an experiment, is that I did this experiment and I came up with a different result, right? Reality makes it so that there are experiments that are just repeatable, and we get good enough at doing them and we get good enough at understanding them that we all can kind of agree this is a problem. And that's what makes empirical theory special. And that was what Popper was actually trying to say. Okay, so he's not against generalizing his epistemology. In fact, Donald Campbell goes on to try to generalize his epistemology, and Popper gave a glowing review of Campbell's attempts to do that, right? He talks about how Campbell anticipated stuff he was writing, and I'll have to find the quote for that, too. Okay, so Popper's totally in favor of taking his epistemology and generalizing it like Deutsch is trying to do. But we've lost something, and it is the realization that empirical theories are special.

[00:58:57]  Bruce: So now let's go back to, and we haven't done a podcast on this, but the question, do animals feel things? That's a completely valid scientific inquiry. And there's a whole body of research where they've tried to work that out, where they've worked out what would be an implication of that, and they've made attempts to do experiments. Now, you can always say, well, that experiment missed something, but you can always say that period. Popper points out you can always just dismiss the outcomes of an experiment and thereby not accept them. You can always just come up with some sort of ad hoc save for your theory, and you can say, oh, well, maybe this is what happened. If you're willing to do that, then you're violating Popper's epistemology, because Popper's epistemology includes not doing things like that. That's specifically what he's ruling out by convention.

[00:59:52]  Bruce: And to ignore that body of work, which would require a whole podcast to go into, even if you think it's wrong, it's still just a conjecture. I conjecture animals feel things. Now, how would I actually do an experiment with that? Okay, so I'm going to try to experiment, and I'm going to try to take fish which don't react to pain, and I'm going to try to inject into their lip, saline for one group and bee venom for another. And I'm going to take these fish that we know don't react to pain. I'm going to see if they even know the difference between these, whether they have saline solution in their lip or bee venom. Well, as it turns out, the ones with bee venom all swim over to the side of the tank and stuff their lip against the tank, and they try to deaden the nerves. Well, that doesn't prove beyond doubt that fish feel things, but that was absolutely what you would have expected if they did.

[01:00:48]  Bruce: So this was a case of where they could have refuted the theory, they didn't experiment [wording as spoken], and they failed to, and it corroborated the theory. That natural tendency to say, but you didn't prove it, that's justificationism, that's what we don't do in Popper. All we do is make conjectures, and then we say, what's the implication? How would I test that? How would I falsify that? How would I have a counter example to it? So my challenge to Dennis would be, okay, stop mucking around with little tiny subjective philosophical arguments that honestly I don't think mean anything. Instead, work out the implications of your theory, turn it into a testable scientific theory. If animals don't feel things like you think, that absolutely should have implications. What are they? If you're going to simply take the stance there are no implications, then I'm not saying you're wrong.

[01:01:41]  Bruce: I'm just saying you're not scientific. And that's enough for me.

[01:01:45]  Unidentified speaker: Well, what makes sense to me is that, yes, animals feel things.

[01:01:49]  Peter: They feel things as animals, though. I mean, they don't have self-reflection that humans do, which makes us feel things as humans. But maybe that's a non-empirical theory there.

[01:02:05]  Unidentified speaker: I don't know how you test that.

[01:02:07]  Bruce: We have to do this as a separate podcast. There actually are some really interesting competing theories here. Because I've actually tried to look up the competing theories, rather than just satisfied with my own subjective criticism. I actually looked up what's the science say, what have they attempted, what are the competing theories. I can actually tell you what they are, but I'd have to look them up, have a prepared explanation, and some of them are just cool. Some of the things they've come up with, one of the ones that I really think is particularly brilliant is animal grief. The fact that grief in humans is arguably against evolution, because you feel grief, you stop taking care of yourself, and you die, and then you can't replicate your genes. Okay, so how do you explain human grief, the fact that we feel grief, for humans? Well, the way we do it, and we do do it, right?

[01:03:03]  Bruce: We conjecture an explanation, but it has to be consistent with other good explanations, so it has to be consistent with evolutionary theory. So we say, well, evolution produced this feeling that you want to be near your kin, and you want to be close to your kin, and you want to take care of your children, and you have this fear of them dying, so that you will take care of them. That's the adaptive part. Once the child dies, there's this unfortunate feeling that you now have to experience, because otherwise it wouldn't have been adaptive in the first place. So while that particular outcome in that case, that is non-adaptive, it is overall adaptive. Well, this is a pretty good argument, right? Okay, and one of the things that's brilliant about it is everybody will tell you there's no way to actually test for the existence of qualia, not even in humans.

[01:03:53]  Bruce: This is where the idea of a philosophical zombie comes from. Well, Deutsch says, look, we just simply consider real anything that shows up in our best explanation. If our best explanation of why grief exists is that it includes an explanation that evolution produced a feeling, okay, now qualia is part of that explanation. We don't need to explain what qualia is. It's now considered real just because it is part of a good explanation. If you buy that argument for humans, you have to buy it for animals. And as it turns out, some animals die from lack of self-care due to grief, right? Their mother dies, and then they die, you know, et cetera, okay? Famously dogs, but like apes do too, right? Elephants do. Not all animals do, but some do. Okay, well, if you, unless you can actually offer a testable explanation, alternative to grief as to why this is happening, then that is the best explanation.

[01:04:54]  Bruce: It's not that we've proven it right. We don't even care if it's proven right, right? Because that's justificationism. It's just, can you offer an alternative that's equally testable? And this is a testable theory, because like, for instance, okay, we've seen animals die from lack of self-care. We're going to hypothesize that that's due to the feelings of grief. So what would we then find correlating with the feelings of grief? Well, we would expect that they would have the same sorts of hormonal surges that humans have. So they will actually track down the apes that they're following, and they'll look for which ones are showing these signs of grief, and then they'll look at the hormones in their feces, and they'll see if it correlates the same way as it does for humans. And as it turns out, it does. Now, if it hadn't, that would have refuted the theory, but it does.

[01:05:46]  Bruce: And its refutation is what matters. It's not trying to confirm something. We'll never prove beyond doubt animals feel things. We never will. We don't need to. There's not even a desire to if you're a critical rationalist. And it's not a matter of what you believe. I don't believe animals feel things or don't believe animals feel things. I simply know that if you want to explain animal grief, right now, the only possible [unclear word] explanation that is out there that exists at the moment is that animals feel grief. At least some animals feel grief. I had a friend who came back with, well, no, I can give you an alternative explanation. It's that the program that they're running in their head, which doesn't feel anything, it doesn't know what to do when it's supposed to follow this program or something. It's supposed to follow the child around, the child dies, and it just doesn't know what to do.

[01:06:45]  Bruce: So it's like, okay, great. I know you can always offer ad hoc saves like this. Just have no doubt about that. My question for you is, what are the actual implications of that theory and how do I test it? Well, I didn't have it. [Wording unclear.] Because it's a totally made up, easy-to-vary theory that tells you nothing about the world. Its sole purpose for existence was to try to ruin the good explanation, which is that the animals in question here, apes in this case, feel great grief. [Wording unclear.] Think about how this, quote, explanation has no empirical content of its own. It simply tries to offer a vague explanation meant to mimic the empirical content of the good explanation, namely that apes that seem to die of grief are in fact dying of grief due to the lack of self-care because they feel bad. That it has no empirical content of its own eliminates this theory from the critical rationalist contest.

[01:07:40]  Bruce: But notice also that its explanatory power actually comes from its competitor. It's basically saying a parental animal that loses a child will act in a way that looks identical to grief due to the vague idea that it doesn't know what to do. Clearly, that is so vague, it could never have actually had any empirical content of its own. But it doesn't even explain the actual case because the case that was in question was an adult child ape dying due to the loss of its mother. So it still isn't really explaining anything. And why would evolution evolve such a poor response in the first place that is clearly negative in value when it could have evolved something else? So it, quote, doesn't know what to do. How about evolving that it does know what to do? That seems like the far more straightforward thing that evolution would evolve. This whole explanation that's supposed to be an alternative just fails as an alternative explanation from a rational standpoint on every single level.

[01:08:37]  Peter: Can I ask you one more quick question that hopefully won't lead us into another long tangent here?

[01:08:45]  Unidentified speaker: Sure.

[01:08:48]  Peter: Could you explain the difference between an explanation and a theory in light of your ideas on refutation and falsification?

[01:09:00]  Bruce: So I actually use those two terms interchangeably.

[01:09:03]  Peter: An explanation and a theory?

[01:09:05]  Bruce: Yeah, so maybe that's a bad answer though, because maybe the two words don't mean quite the same thing. So let me say that both of those terms probably mean many different things in many different circumstances. Words tend to have multiple meanings, right? So if I were to say, look, I've got this theory that JFK was killed by a conspiracy. Well, I mean, like, that's not a theory in the scientific sense, right? I mean, it's vague. It doesn't make any testable predictions I can go try to experiment with. So we wouldn't call that a theory in the scientific sense. And yet using the word theory is non-problematic, because you know what I really mean is I've got this idea that I think is true.

[01:09:52]  Peter: It seems to me, Deutsch uses it quite broadly. Like when he talks about a theory, he's not just talking about a scientific theory.

[01:09:59]  Unidentified speaker: Yes.

[01:10:00]  Unidentified speaker: At least that's what I get.

[01:10:01]  Bruce: And then let's use the example of Bigfoot. So I've got this theory that Bigfoot exists. Well, that's clearly not a scientific theory, because there's no way to do an experiment to test it, right? On the other hand, because it's open-ended, I don't know where... There's no experiment I can do where I can say, now I've confirmed that there's no Bigfoot, right? On the other hand, the Loch Ness Monster has to exist within a specific lake to be considered the Loch Ness Monster. So let's say that I've got this theory that the Loch Ness Monster does not exist. Okay, now notice that I've flipped it, okay? Because now I'm going to try to refute the theory by counterexample. So I go out, I take a bunch of boats, and I do a sonar mapping of the entire lake, and we find no Loch Ness Monster. Well, that still doesn't prove beyond doubt there's no Loch Ness Monster.

[01:10:58]  Bruce: Maybe the Loch Ness Monster is a super intelligent alien that knew how to swim around the boats or something like that, right? But what I did is I had a scientific theory that within this one particular lake, there is no Loch Ness Monster, and then I strenuously, severely tested it by going out and taking all these boats with sonars and scanning the whole lake, and we did not find the Loch Ness Monster, okay? So the theory, there is no Loch Ness Monster, continues to hold and is a valid scientific theory, okay? Whereas the theory, the Loch Ness Monster exists, is not a valid scientific theory because there's no way to actually do an experiment for it. Does that make sense?

[01:11:38]  Unidentified speaker: I see, yeah.

[01:11:39]  Peter: Well, I don't feel bad for being a little, suddenly a little confused on the difference, but I get what you're saying.

[01:11:47]  Bruce: Okay, now let me tie this to explanation.

[01:11:49]  Unidentified speaker: Okay.

[01:11:50]  Bruce: This is another one where the fans of David Deutsch, particularly on Twitter, misunderstand something important. So in science, Popper argues that an empirical theory, that what you want is you want to, you want to replace the theory with a better empirical theory. So that some theories have more empirical content than others, and that all things being equal, meaning both theories have survived all tests, you would always prefer the more empirical theory over the less empirical theory. So let me give you an example of that. The example Popper uses, I don't like, but it gets, it makes the point, is let's say you have a theory that the planets followed ellipses, or a theory that the planets followed circles. All things being equal, you would prefer the theory that the planets follow circles, because a circle is a type of ellipse. So ellipse, so a circle has more empirical content, it's easier to falsify, easier to find a counter example, which is what I mean, than the theory that it is an ellipse.

[01:12:53]  Bruce: Now this is, the problem is this is a bad example, because first of all, the planets don't follow either ellipses or circles. And because of that, it's hard to really grasp why we would prefer the circle when we know it's a false theory. So let me give an example of this that's better because it's a true theory. So let's start with the theory, oranges stop scurvy, okay? Well, is that an explanatory theory? Well, the twit rats on Twitter, they have argued for these types of theories that they don't count as explanatory theories because they don't explain anything. They're just talking about a correlation. Okay, well, that's wrong because this theory does explain something. It explains that if you eat an orange, you won't get scurvy. That's what it explains. It doesn't have any empirical content or explanatory content outside of that, but it's not [zero?].

[01:13:45]  Peter: And it's hard to vary as well.

[01:13:47]  Bruce: It is hard to vary.

[01:13:48]  Unidentified speaker: If you eat a banana, you're not going to slice. [Wording unclear.]

[01:13:53]  Bruce: Okay, moreover, it can be falsified. What I mean by that is, you can find counter examples to it. You could actually go out, have people eat oranges, and if they get scurvy anyhow, then that theory is false, right?

[01:14:07]  Unidentified speaker: And you're going to need a better theory.

[01:14:08]  Bruce: Okay, now consider the following theory, and I know I used this example in a past podcast.

[01:14:13]  Unidentified speaker: Vitamin C stops scurvy.

[01:14:15]  Bruce: Well, this is a better theory than the oranges stop scurvy theory. Okay, in this case, they're both true, but the vitamin C stops scurvy is far more precise, and therefore has higher empirical content. And it's easier to figure out which experiments to do. Now we can go, you told me I got this wrong before, but you said lemons have vitamin C, and therefore we can go try lemons and see if those also stop scurvy. And we can go try fruits that don't have high contents of vitamin C, which happened in real life, by the way. They thought that there was certain kinds of fruit, so they would bring these fruits with them that didn't have vitamin C, and they would all die of scurvy. So it was a natural experiment, right? So we now have a much better theory. Now let's move on to vitamin C stops scurvy by causing the growth of connective tissue. I'm not a scientist on this.

[01:15:08]  Bruce: Okay, like I looked this up on Wikipedia or something. And there's this detailed explanation for how vitamin C has these mechanical effects inside the body that cause it so that scurvy isn't created. Well, this is a way better empirical theory. And the reason why each of these theories is better is because we're maximizing explanatory content. And we're doing it in a way where it increases the empirical content. Now, one of the things that the Deutsch fans on Twitter get wrong is that they'll say, this theory, pet theory of mine, maximizes explanatory content by which they mean does not maximize empirical content. They just personally subjectively feel like it's a really good explanation. So they've moved us away from the objectiveness of empirical theories and back into the subjective realm. Again, they do this all the time. They're always trying to find ways to immunize their theories through by using subjective arguments instead of objective arguments.

[01:16:09]  Bruce: And the example of Brett's theory with IQ, the fact that I argued it actually is a scientific theory, and I think it is. I think it's a good scientific theory, one worthy of respect. But they were trying to immunize it as, no, it's just a philosophical theory, because they wanted to apply, they wanted to call it explanatory, but they didn't want to subject it to experiments, because then it would be a refuted theory. So what you end up with is an attempt to maximize explanatory content while minimizing empirical content. Well, that's the problem with communism. That's exactly what Popper was trying to explain, is wrong with an explanation like communism. Sure, people love communism because it's, quote, highly explanatory, right? It explains everything. In fact, no matter what you come up with, it explains it. It explains every possible outcome of every possible situation, in which case it really means it explains nothing.

[01:17:07]  Bruce: So what we're really looking for is we're looking for the way you maximize empirical content is by maximizing explanatory content, but we're interested in explanations that are empirical.

[01:17:19]  Unidentified speaker: Okay.

[01:17:19]  Bruce: So we want those two to always be together, that you're maximizing explanatory content and empirical content at the same time, in which case, yes, maximizing explanatory power is always a benefit, right? Even if the theory is wrong, it's still good, because then you can show there's something wrong with the theory, and you can figure out how to try to tweak it and how to try to improve on it, how to error correct it, things like that, right? So we want maximum explanatory content pretty much no matter what, even if the theory is false, right? So in some ways, that's more important than actually refuting the theory. Now, obviously, it's probably better to think of it not as one's more important than the other, but these are two different things we want at the same time. We want the theory to pass all the tests we can think of, and we want it to have lots of tests, because it has high explanatory content and thus high empirical content.

[01:18:11]  Bruce: But if we're doing this in context of Popper's epistemology, we're expecting explanation and empirical content to be the same thing. So yes, one might argue that communism or Brett's theory on intelligence, that those are explanations and that they have good explanatory content, but not in the sense of actually being able to do an experiment and have empirical content by which we can judge it. So in that sense, we're not maximizing. When we talk about we want to maximize explanatory content in Popper's epistemology, we're always assuming that we're doing so in a way that maximizes empirical content at the same time. When thought of in this way, if we're using the word theory in the scientific sense or the word explanation in the scientific sense where we're maximizing explanatory content so that we can maximize empirical content, then I think those two words really mean the same thing.

[01:19:08]  Bruce: You can say the theory of general relativity or the explanation of general relativity, and they're like the same thing. But I definitely think that the way the words just get used in general, they can mean all sorts of different things, and they don't necessarily mean the same thing. So a theory would be an idea about something, an explanation would be trying to describe in detail, in empirical detail, why that theory is true, right? I have a theory that there's this force called gravity, and I've got this explanation that it's because of the curvature of space as per Einstein's theory. But in reality, we don't have a scientific theory that there's a force called gravity. We have Newton's theory of gravity, or Einstein's theory of gravity, which include explanatory content, which allows us to make it empirical. So I think that as long as we're sticking within empirical theories, they're kind of the same thing.

[01:20:11]  Peter: Well, very clear—clarified a lot there and throughout the our entire discussion here. And I thank you for that, Bruce.

[01:20:22]  Unidentified speaker: You're welcome.

[01:20:23]  Peter: And I'll look forward to doing it again soon.

[01:20:27]  Unidentified speaker: All right.

[01:20:28]  Unidentified speaker: Thanks, everybody.

[01:20:29]  Unidentified speaker: Okay.

[01:20:29]  Unidentified speaker: Thank you.

[01:20:30]  Unidentified speaker: Bye-bye.

[01:20:31]  Unidentified speaker: Bye-bye.

[01:20:34]  Bruce: The Theory of Anything podcast could use your help. We have a small but loyal audience, and we'd like to get the word out about the podcast to others so others can enjoy it as well. To the best of our knowledge, we're the only podcast that covers all four strands of David Deutsch's philosophy, as well as other interesting subjects. If you're enjoying this podcast, please give us a five-star rating on Apple podcasts. This can usually be done right inside your podcast player, or you can Google The Theory of Anything podcast, Apple, or something like that. Some players have their own rating system, and giving us a five-star rating on any rating system would be helpful. If you enjoy a particular episode, please consider tweeting about us or linking to us on Facebook or other social media to help get the word out. If you are interested in financially supporting the podcast, we have two ways to do that.

[01:21:20]  Bruce: The first is via our podcast host site, Anchor. Just go to anchor.fm slash four dash strands, F-O-U-R dash S-T-R-A-N-D-S. There's a support button available that allows you to do reoccurring donations. If you want to make a one-time donation, go to our blog, which is fourstrands.org. There is a donation button there that uses PayPal.

[01:21:45]  Unidentified speaker: Thank you.