Episode 42: Popper without Refutation & Resolving the Problems of Refutation (part 2)

• 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

[00:00:11]  Bruce: Welcome back to The Theory of Anything podcast. We're continuing our discussion about Popper without Refutation and related subjects. How's it going?

[00:00:20]  Cameo: It's going well, Bruce. How are you?

[00:00:22]  Bruce: Good. I just finished my 15-minute presentation for the Popper Conference.

[00:00:27]  Cameo: And did they kill you?

[00:00:29]  Unidentified speaker: Did they try to?

[00:00:30]  Bruce: Yeah, they tried. You had mentioned in the last podcast that they're going to be coming at me with lightsabers and at least one person was, and it was David Miller, very probably the most well-known and best Popperian alive today. So I'll actually talk about what he had a concern with. I think it was more a concern with something David Deutsch said that I had printed. Or maybe he had an accent, so it was a little bit hard to understand what his concern was, but I think I understood what it was. So David Deutsch specified in our last podcast, I gave the quote where David Deutsch specified that to have a refutation requires a second theory.

[00:01:14]  Cameo: Okay.

[00:01:15]  Bruce: Now, in context, what I actually was saying was that's a different way of thinking about the word refutation. That is not the way Popper thought of the word refutation. And I think maybe he missed that I said that. So he came back with what Deutsch is saying might be true for physics, but it's not true for every universal statement. I think what he's trying to say is, and maybe this is where it ties into what I was saying, you don't always have to worry about background knowledge for every single universal statement. If you have the universal statement, all swans are white, and you have the basic statement, here I have a black swan, that refutes that statement. There's no actual background knowledge. Well, now there is. And I brought this up to him and he didn't like it, but I'm right. But he's right, too. We're not really disagreeing. I said, part of a basic statement is you're basically just assuming this is an empirical thing that we have to deal with.

[00:02:14]  Bruce: Typically, a single observation does not count as a basic statement. Typically, it has to have been repeated by other groups. Everyone can see that this is true. It's not just you. So one of the first things you're going to do when somebody observes a black swan is they're going to go check to see if it's not a white swan that's been painted black, right? They're going to do their due diligence on their background knowledge that they've kind of eliminated every obvious possibility first, right?

[00:02:46]  Unidentified speaker: Yes.

[00:02:46]  Bruce: And that would be an example of how you're always dealing with background knowledge. You can never not deal with background knowledge. Now, his point is, look, that's all part of the basic statement, and he's right about that. The basic statement: here is a black swan. The assumption is you've already done all that. Everybody is seeing this black swan. That's an empirical thing we now need to deal with. Now, it could be wrong because everything is tentative, right? Nothing is ever for sure. But at this point, we're assuming we've got a black swan because—that is, we've already tried everything else. So now at this point, we've now refuted the statement, all swans are white. And insofar as that's what he means, I completely agree with him. Now, a completely fair question would be, do actual scientific theories ever take that form? No. I mean, they're always more complicated than that.

[00:03:37]  Bruce: I guess one way you could look at this is Deutsch was talking about real scientific theories, and David Miller is talking about universal statements, and that the two aren't quite the same because a theory could have multiple universal statements involved with it.

[00:03:52]  Cameo: Yeah, there's a level of complexity that's different between those two inherently.

[00:03:59]  Bruce: I think the thing he was trying to clarify is, look, if you're talking about just a single universal statement, a refutation could be against that universal statement. It does not necessarily a part of some other theoretical system. [Wording unclear.] I think my point was, well, in real life it's always part of some theoretical system, but the basic statement may be taking that into consideration. I don't know that we were really disagreeing, but he certainly didn't like the way it was worded.

[00:04:25]  Cameo: In a 15-minute talk, though, how does he even have time to ask you that kind of a question?

[00:04:30]  Bruce: After 15-minute talk, there's 15 minutes of questions. I got quite a few questions.

[00:04:35]  Unidentified speaker: Okay.

[00:04:36]  Cameo: Did he like your answer? Once you started discussing it with him, was he like…

[00:04:41]  Bruce: There wasn't—yeah, he kind of was. But we were agreeing, so the moderator said, you guys are agreeing, let's just move on. That was kind of where it ended. You guys are agreeing.

[00:04:53]  Unidentified speaker: Okay, enough fighting. You guys agree.

[00:04:55]  Bruce: Well, he had to let other people ask questions. It wasn't possible for us to talk further and to work out if there was a difference of opinion or not. I don't think there was. There's not much I can explain in 15 minutes, and there wasn't much chance to respond to him. My guess is that if we actually could sit and talk, we would realize, in context, we're really just saying the same thing. It's not really at odds with each other. It's a good distinction though. The fact that when I say it's always refutes the theory plus the background knowledge, I am talking about scientific theories, complicated scientific theories. If you're talking about a single universal statement, yes, you can think of it as all the background knowledge is now contained within the basic statement at this point, and therefore you can abstract that aspect away. I'll admit that upfront as part of this podcast as a good clarifying statement.

[00:05:51]  Cameo: All right. I think that's accepting your own refutation.

[00:05:57]  Bruce: Yeah, absolutely. I've tweaked my theory to now I've refuted the old version. I've tweaked my theory, so now we've got a better version. That's exactly what we want.

[00:06:07]  Unidentified speaker: A very Popperian of you.

[00:06:09]  Cameo: It's almost like it's a Popperian podcast.

[00:06:14]  Unidentified speaker: All right.

[00:06:15]  Cameo: I just want to point out because I googled David Miller. I didn't know who he was. He has written four books, The Most Recent, Out of Error, Further Essays on Critical Rationalism. Prior to that and his most notable work is Critical Rationalism, A Restatement and Defense, and then Popper Selections. Then his very first book was Croquet and How to Play It.

[00:06:39]  Bruce: I wonder if that's a different David Miller.

[00:06:41]  Cameo: No, it's not. I'm looking on his Wikipedia page.

[00:06:44]  Bruce: Oh, really?

[00:06:45]  Unidentified speaker: Yes.

[00:06:45]  Bruce: That is awesome.

[00:06:47]  Unidentified speaker: That is awesome.

[00:06:49]  Bruce: I'm planning to read David Miller. I've only read one of his papers and it was really awfully good. So he is really someone who knows what he's talking about. Let me just say that. The paper I did read was Problems with Critical Rationalism, where he's actually trying to look at what are things that we have not resolved, that really are problems.

[00:07:09]  Unidentified speaker: Right.

[00:07:10]  Bruce: Some of them were too complicated for me to understand, but the ones I could understand, they were all spot on. I mean, he's really thought this through well.

[00:07:18]  Cameo: I don't even know what it might look like if it's too complicated for you to understand.

[00:07:27]  Bruce: It's good for me to remind people that I'm really and truly not an expert.

[00:07:32]  Unidentified speaker: I know you're not.

[00:07:33]  Bruce: I'm just an interested layman. I'm neither a scientist nor a philosopher, and I'm not trying to be either. It's important that we as laymen that we try to understand things. That's what I try to do. I'm very interested in trying to understand what is this, and I'll put some effort in to figure it out. If it's something I'm interested in, I'll put some effort in to figure it out. We laymen, we have things intelligent to say to experts, and that's just the way real life works. You cannot just trust experts because I actually had an interesting, I was going to bring this up in a future podcast. Let me mention it, I'll re-mention it again. I had an interesting conversation with a guy who's studying philosophy, and he asked me if it was actually true that the Church–Turing–Deutsch thesis really implied that we could create an AGI. I said, yes.

[00:08:28]  Bruce: Now, I had certain things in mind. Just as I was saying, it's always based on your background knowledge. I was making certain assumptions that maybe he wasn't. If the conversation had gone further, I could have clarified that. For instance, I was assuming the laws of physics were correct. If he's not assuming that, then obviously everything goes out the window. So what I had in mind that he was asking, usually when someone asks me a question like that, they are assuming the laws of physics are holding. They're not assuming there's a new set of laws of physics. So I was making certain, and you always have to, until someone clarifies, it would be natural for me to assume that. What I was trying to explain though, is that even if the laws of physics are false, that would really just mean that the Turing-Deutsch thesis is wrong, in which case it still implied it.

[00:09:18]  Bruce: I was trying to explain that to him. This isn't some dogmatic statement, this isn't some thing that it's not even that big a deal, it's just the realization that if that theory happens to be true, then that's part of what that theory actually implies under our current understanding of the laws of physics.

[00:09:36]  Unidentified speaker: Okay.

[00:09:36]  Bruce: And he said, here's this guy who's an expert and he gives me a book. And he says, he's disagreeing with you, so you must be wrong. So I read part of the book. And the guy who's this expert, who's written multiple books on the subject, by the way, he starts off explaining that the Turing–Deutsch—sorry, the Turing thesis, he wouldn't have known about the Deutsch thesis, that it was describing what a human computer could compute, and that the class of computations that a Turing machine can do would be different. That is false. That is completely not what they teach you in computational theory school, which I've been in twice now, both at the undergrad and the grad level, right? That is absolutely just not the case. So I whipped up Turing's paper, which he was quoting. And I said, you don't have to take my word for it. Go read Turing's paper.

[00:10:29]  Bruce: You will see that this expert is wrong on what he said right here. And the guy at this point says, I can't believe that you would claim this expert's wrong. There's no way that's really true. He's written all these books. I'm not going to talk to you anymore. And the conversation ended.

[00:10:45]  Unidentified speaker: It—seriously… [Overlapping fragment unclear.]

[00:10:46]  Bruce: Yes. You know what? I understand where he's coming from. Okay. This guy's an expert and I'm just a layman. And I'm not even trying to be anything but a layman. You know, if you were a betting man and you, if you didn't have the ability to actually go look at Turing's paper and resolve it by looking at Turing's paper, which he refused to do, unfortunately. If you were a betting man, you should bet on the expert, right? But in this case, I'm right, and the expert's wrong. And that's a fact, right? I mean, all you have to do is go look at the paper and you can see it's wrong.

[00:11:15]  Cameo: The interesting thing about that, though, is you didn't, you didn't try and talk him out of his belief in that expert by showing your own expertise. You gave him the expert that the expert was citing. And you gave him the opportunity to look at this source material. This actually, I don't even want to talk about this anymore. I think we should totally save it for a future episode. It's a fascinating conversation, and we will never get through the end of Refuting Popper or Popper's Refutation.

[00:11:52]  Bruce: You're getting at exactly what's so interesting about it, is that there was no reason to take my word for it. There was no need for me to assert my expertise, because all I needed to do was point to the actual paper.

[00:12:06]  Cameo: Yes.

[00:12:06]  Bruce: And that's actually the right way to go about this, right, is experts' appeals to authority mean nothing. They literally mean nothing. And that's a hard thing for us to accept, especially since we live in a world where you just don't have time to go research everything. You almost have to assume authority means something, right? You're playing this kind of gamble with it. But at the end of the day, there's just a factual statement being made, and it's either true or it's false, and we can actually determine if it is or not, right? And of course, I was based on certain assumptions that I was assuming he was making. Maybe if we did talk, he would have said, oh, I'm not assuming this and you are. And I said, oh, yeah, if you're not assuming that, then this would be true instead, right? And I've got no problem explaining that. But everything I said was correct within the set of assumptions that I thought he was asking me.

[00:12:53]  Bruce: So anyhow, and it didn't matter what this expert said. It like literally, and I think that's a hard thing for us to accept, is just how often experts are completely wrong. They come across so knowledgeable, but they are often completely wrong.

[00:13:09]  Cameo: Right. So we're going to put a pin in this and get back to, how did we even get here?

[00:13:17]  Bruce: All right.

[00:13:18]  Unidentified speaker: So let's talk about David Miller, right?

[00:13:20]  Bruce: Yes. Yes. So let's do a quick review here. So when Popper speaks of a refutation, he really means a counter example or a problem. Okay. And here, I will allow for David Miller's correction. If we're talking about a single universal statement, this is still true, but it's all part of the basic statement that we're talking about at this point, so you can ignore it. But we're talking about a real theory. We're talking about some combination of the theory plus the background knowledge. You don't have a true refuting case if you mean the theory itself. You're really refuting the combination of all the theories that you are dependent upon. Theories about your instruments, assumptions about how many planets there are, everything in your background knowledge, some of which you may not have even explicitly thought of. It's just a tacit assumption that you're making.

[00:14:09]  Bruce: You're actually refuting that whole combination. Therefore, it might be better to refer to them as problems or counter examples.

[00:14:17]  Cameo: I like the word counter example, I think.

[00:14:19]  Bruce: Yes. It can be thought of as a refutation of the combination of the theory plus the background knowledge. If it's a single universal statement, and if we're accepting David Miller's adjustment, then there is no background knowledge because in that case, it's part of the basic statement. Therefore, the theory plus the background knowledge equals the theory. But other than that, everything I'm saying is still correct. It would be very common for there to be background knowledge that you have to carefully think about. It would probably be better not to call such problems a refutation at all, as the words carry too much baggage. This is one of the main sources of misunderstanding that exists about Popper that has kept his philosophy from catching on in a lot of quarters. I gave examples of that. I give examples of people who clearly had misunderstood this aspect of Popper, and therefore thought they had disproven Popper when in fact they hadn't.

[00:15:08]  Bruce: And even a famous Popperian like David Deutsch clearly is using the word refutation in at least somewhat different way than Popper himself. And that ought to give us a sign that even committed Popperians sometimes, don't make this distinction correctly. And then I listed out several problems with refutation that are actually, we're going to prove this time around that they're pseudo problems, but at least they sound like legitimate problems the way I worded them back then. Okay. So now let me say, okay, is this really true? I had people challenge me on this, not in the conference, but Sadia and her husband were talking with me about this and they said, okay, is this really true? Like with Jupiter's orbit, when you see that Jupiter's orbit doesn't match, does that really refute Newton's theory? And I said, no, it doesn't refute Newton's theory, but it is a Popperian refutation.

[00:15:59]  Bruce: And they're like, okay, explain that to me. So, okay, okay, here we go. So with that theory, consider that the two competing theories are Newton plus seven planets versus Newton plus eight planets. That's really the two competing theories that we're going to ultimately end up with. Initially, we have other theories like possibly Newton's wrong. We eliminate those as we go. Okay. And we come down to Newton plus seven planets versus Newton with eight planets. Once we know those are the final two competing theories, the fact that Jupiter's orbit doesn't match Newton plus seven planets is a refutation of that theoretical system. Okay. So it is a refutation even though it didn't refute Newton. And this is the distinction that I'm trying to make, is that the fact that it turned out to be a part of the background knowledge doesn't make it not a refuting case, which is why probably we should call them counter examples instead.

[00:16:55]  Bruce: So it's still a completely fair, Popperian refutation. It doesn't change Popper at all the fact that Jupiter's orbit didn't match Newton. Okay. Because it still refuted the theory that there was seven planets. So now what about Mercury? So with Mercury, we have this orbit that doesn't match. And they already had this history with Jupiter. So they thought, that means there's another planet. So they named this planet Vulcan, which is from Star Trek, that's where the term actually comes from. And so they named this planet Vulcan. They calculate where the planet must be for the orbit of Mercury to match. Then they go out and they look for it and it is not there. They just eliminated the theory Newton plus nine planets. So in this case, here's the competing theories. Newton plus eight planets versus Newton plus nine planets versus general relativity plus eight planets.

[00:17:50]  Bruce: So the fact that Mercury's orbit doesn't match prediction, refutes Newton plus eight planets. The fact that we then said, let's look for this ninth planet and we don't find it, refutes Newton plus nine planets, leaving general relativity plus eight planets. So these are still fair refutations. Once you realize that you're doing the combination of the theory plus the background knowledge, the idea of a Popperian refutation will always work. There's never really a problem with it, other than the fact that the word refutation happens to bring the wrong thing to mind. But in terms of the way Popper intended it, it's just non-problematic. Okay. Do you see what I'm saying here?

[00:18:33]  Cameo: Yeah. And was this part of your presentation?

[00:18:36]  Bruce: No, this was not part of my presentation. I did use this as one of my examples during the discussion afterwards, though.

[00:18:43]  Cameo: I see. Because I think this does bring a lot of clarity to what that background knowledge portion is and the impact of it.

[00:18:55]  Bruce: Yes. Okay. And this is why I say they're counter examples. We've got this counter example. We don't know yet what it refutes, right? We have a guess, we have a theory. A lot of times we have a very good theory. If I measure something and it's totally off from what theory says it's going to be, the very first thing I think of is, oh, something's wrong with my instrument. So I go calibrate my instrument or I go try a different one. Everybody does that. You never think of that as a refutation, but it is. It's refuting the part of your background knowledge that was the assumption that your instrument worked.

[00:19:31]  Unidentified speaker: Right.

[00:19:33]  Cameo: It's really great clarification.

[00:19:36]  Unidentified speaker: Yeah.

[00:19:37]  Bruce: Okay. Here's the key thing, though. Problems are objective. Okay. There's never a case where a problem isn't a problem. Now, again, let me make sure we're always realize we're always talking about theory plus the background knowledge. The fact that Jupiter's orbit didn't match, that was a problem that had to be resolved. And it had to be resolved by adding an eighth planet to the solar system. The fact that my instrument is off and I have to recalibrate it, that is a problem and I have to change my theory by recalibrating my system. That changes that part of that theoretical system. There's never a case where a counter example or a problem isn't an objective fact that you have to deal with.

[00:20:19]  Unidentified speaker: Right.

[00:20:19]  Cameo: You don't just get to like brush it under the carpet and say, hey, that's not a, just don't worry about the fact that we're questioning whether our tooling works.

[00:20:30]  Bruce: Right. You will always go and you will resolve that problem in some way. I guess the one counter example you could come up with is something, like you repeated it, it doesn't happen again. At that point, that repetition, you make the assumption, I did something wrong and I don't know what it is, but I'm not going to worry about it. But that is still dealing with the problem objectively. I mean, that's still you making now a conjecture, I did something wrong and I don't know what it is, but now that I'm getting the right answer every single time, I'm just going to go with that. That is still a case of you having some part of your background knowledge that got refuted and you've now replaced it with something else. There will never be a case where that isn't what you're doing, so long as you are looking broadly at what we mean by theoretical system.

[00:21:16]  Bruce: This is why this matters, is because the fact that problems are objective is precisely why Popper's epistemology is correct and why progress is always possible. Because you always have something you can hang your hat on and say, okay, now I've got to deal with this problem. I can now conjecture how to deal with that problem. All right. Let's talk about now a bit more about Popper's epistemology, specifically the asymmetry. Now, we talked last time about the fact that there's at least a common misunderstanding amongst lay Popperians like myself, at least. I don't know this misunderstanding exists amongst like true Popper scholars, but amongst laymen like me, there's this misunderstanding that we call the absolute verification fallacy. It's the idea that when Popper talked about refutation, we refute theories but we can't verify them or support them.

[00:22:04]  Bruce: That what he really meant was that refutation is always tentative, but verifications imply certainty and certainty is impossible. Well, if that were really the case, if that were really what Popper meant, then all we would need to do is replace absolute certainty with tentative certainty, and suddenly we've restored the symmetry between refutation and verification. That can't possibly be what Popper meant. Well, what did Popper mean? Well, I'm going to quote him here. He says, this asymmetry that he's referring to has a purely logical and also methodological or heuristic aspect. As to its logical aspect, there can be no doubt that a unilaterally falsifiable universal statement is logically much stronger than the corresponding unilaterally verifiable existential statement. Remember, we talked last time about the fact that a universal statement can logically only be refuted and an existential statement can only be verified.

[00:22:59]  Bruce: For the following is a well-known logical rule. From universal statement pertaining to all things of a certain kind, or all elements of a certain kind, non-empty universe of discourse. One, all things have the property P, we can derive for any individual thing, A belonging to this kind of universe. Two, the thing A has property P, and from two in turn we can derive three, there exists a thing that has property P. Thus, one entails two and three, two entails three, but three does not entail either one or two, and two does not entail three, or one. [Numbered relations retained as spoken.] In other words, one is logically stronger than one and two and three [as spoken], and two is logically stronger than three. Of course, there's no way anyone just listening to this is going to follow anything I just said.

[00:23:40]  Cameo: I'm looking at the screen.

[00:23:42]  Unidentified speaker: I'm not following.

[00:23:45]  Bruce: And this is the source of the important asymmetry in the case of unilaterally falsifiable, universal, and unilaterally verifiable, existential statements, and the situation is the same for more complex statement. This is from Logic of Scientific Discovery, page 184.

[00:23:59]  Cameo: So I can see why there is some disagreement as to his intent.

[00:24:07]  Bruce: So let me explain this kind of in plain English, okay?

[00:24:11]  Cameo: Before you do that, can I ask, since I am looking at the screen, this is just a direct extract from...

[00:24:20]  Bruce: That is correct...

[00:24:20]  Cameo: …184.

[00:24:21]  Unidentified speaker: Okay, thank you.

[00:24:22]  Bruce: That's correct. An existential statement would be like Cameo, there exists a woman named Cameo. Something along those lines. Okay. We might even, if we move on to any type of singular statement, we might say Cameo is a woman or something like that. So a universal statement though, would be like there are no such thing as women. One of those two statements tells you a lot more than the other, is what he's trying to get at. So if I wanted to verify that Cameo exists, I go meet Cameo, right? And then I verify Cameo exists. If I want to verify Cameo is a woman, maybe I take a DNA test, how we happen to be defining woman, I guess. There are ways you could go about verifying that basic statement. If I say there are no such thing as women, all right, that tells me that the statement Cameo is a woman, if we're assuming that statement is true, that tells me the statement Cameo is a woman is false.

[00:25:22]  Bruce: In fact, it also tells me the statement that Margaret Thatcher is a woman is false. It tells me all sorts of things because it's just a logically stronger statement. This is what he's getting at is that science is built on universal statements like this constraints, things that say this cannot be true or this is always true or something along those lines. Because those are the statements that we are interested in in physics, sorry, in science, particularly in physics, but all science, it makes sense that those are only logically refutable. Now, we're going to see that there are some theories that challenge this slightly, and I'll explain how to deal with those. But just at a general level, hopefully you can already see what I'm talking about. That universal statements are just logically stronger, and because they are logically stronger, that is where scientists are going to be concentrating, and because that's where scientists are going to be concentrating, they will only be able to deal with those statements by refuting them.

[00:26:20]  Bruce: The way you refute the statement, there are no such thing as women, is by demonstrating that Cameo is a woman.

[00:26:26]  Cameo: Yeah, I can accept that.

[00:26:27]  Bruce: This is the asymmetry. Now, it's a logical asymmetry. It's got nothing to do with certainty or lack of certainty. Okay. That this does not even play a role in Popper's asymmetry. And this is what the absolute verification fallacy has misunderstood. This is also why he used the word refutation, because when you're dealing with logic, the correct term in English for how you show that a universal statement is false, is called a refutation. It just so happens that that's not the best word in the scientific realm for it, because typically we're talking about an individual theory at that point. But in terms of the logical realm, it's exactly the right word, and that's why Popper used it. Also, I would argue that Popper English was a second language for him. So I think some of these misunderstandings come from the fact, and I've heard, I don't know German, but I've heard that some of these problems don't exist in German.

[00:27:18]  Unidentified speaker: Oh, interesting.

[00:27:19]  Bruce: So it may be he did his own translation for what I understand. Someone correct me if I'm wrong in the audience there. But it makes sense that he picked the best words he could in English, but that maybe there were better words to pick.

[00:27:34]  Cameo: Or maybe our language doesn't even-

[00:27:38]  Bruce: Have the concept, right?

[00:27:39]  Cameo: Have the concept. Germans have, I think, a more negative… Americans have this kind of artificial over-emphasis towards positiveness that I don't think you see in a lot of European countries, in Germany. So it's fascinating. There may not have been the language for him to describe the nuance he wanted to describe.

[00:28:08]  Bruce: Without using lots of words, yeah.

[00:28:10]  Unidentified speaker: Yes.

[00:28:10]  Bruce: Like I'm trying to do, right?

[00:28:12]  Unidentified speaker: Yes, yes.

[00:28:12]  Cameo: I mean, you can see how much work it is to get to that precision of what you want to describe.

[00:28:19]  Bruce: Yeah. This is Popper again. Owing to the logical power, universal statements may be important as explanatory hypotheses. They may explain, especially in conjunction with singular initial conditions, singular events or statements. Purely existential statements, on the other hand, in isolation or even in conjunction with singular statements, are usually too weak to explain anything. This is why scientists are interested in universal hypotheses rather than isolated existential hypotheses. Notice that this is almost more like a convention. Popper's not really saying you're never going to be interested in a theory that has an existential statement. And this is a good nuance to understand that he's talking about a certain convention. It's possible to word things in such a way to make Popper sound wrong, but then you're just misunderstanding his point.

[00:29:11]  Bruce: I'll give an example of that in a second. This leads to this methodological or heuristic aspect of asymmetry due to the difference between the critical or falsificationist attitude and the verificationist attitude. This is Logic of Scientific Discovery, page 184. And then he talks about how predictions are verified but not refuted. And this is something that I think a lot of lay Popperians like myself miss, initially, I know I did, is that verification plays this huge role in Popperian epistemology. It just is in a different, it's just in terms of experiments. Testing via prediction is the main case where Popper's epistemology cares about verifications. This is because predictions take the form of singular statements, which are a form of existential statements and thus have to be verified, not refuted. So here's Popper again. Certain singular statements, which may be called predictions, are deduced from the theory being tested.

[00:30:01]  Bruce: That's me saying being tested. If the singular conclusions turn out to be acceptable or verified, then the theory has for the time being passed its test, we have found no reason to discard it. Notice that Popper does not avoid the word verified when we're in this context. Okay. And for those who think I'm doing this wrong, I looked up in the index and it said existential statements can only be verified in the index. So Popper does not avoid the language of verification. He's just using it in a different context, specifically how we test things, how we test predictions. Okay. And they still play a big role there. So to really understand this distinction now and to get this kind of driven home what Popper was getting at, let's talk about his demarcation criteria. So now famously his demarcation is between science and non-science. I'm going to make the argument, and again people will come out with lightsabers, but I'm going to make the argument that this is really about the worst way he could have said it.

[00:31:02]  Bruce: He's right in the way he's thinking of the word science, but not in the way most people in the English language think of the word science. What he really means is that his demarcation is between empirical and non-empirical. That's what he really means. And in fact, here's a quote from Popper where he says it outright. The criterion of demarcation between empirical and non-empirical theories, I have called the criteria of falsifiability or the criterion of refutability. It does not imply that irrefutable theories are false, nor does it imply that they are meaningless. But it does imply that as long as we cannot describe what a possible refutation of a certain theory would be like, the theory may be regarded as lying outside the field of empirical science. So notice the first part of the quote, he doesn't even use the word science. He just says empirical and non-empirical theories.

[00:31:49]  Bruce: And then when he does mention the word science, he qualifies it as empirical science. Now, Popper's pretty good about qualifying that he means the boundary between empirical science and non-empirical science, which would be identical to saying the boundary between empirical and non-empirical theories. Occasionally, he doesn't though. So if someone wants to find me a quote and says, here's where Popper says it's the difference between science and non-science. Yes, I know plain well it is possible to find quotes that say that. But they're rare. He almost always clarifies he means empirical science. So when I mentioned this, I say really the demarcation is between empirical and non-empirical. And the reason why Popper says between science and non-science is because to Popper, the word science meant the set of empirical theories. Therefore, he meant the same thing.

[00:32:40]  Bruce: He meant empirical and non-empirical. When I tell that to Popperians, mostly lay ones that I'm talking to like on Twitter, the reaction is always a strong, oh, that's not what Popper said. And sometimes they'll stop and think about it. And they'll say, I'll say, okay, if you don't think his demarcation is between empirical and non-empirical. So first of all, let me clarify that if in your mind, science equaled the set of empirical theories, then you should not have challenged me. You should have said, oh, well, those two are the same.

[00:33:13]  Unidentified speaker: That's what you should have said.

[00:33:16]  Cameo: I can agree with that.

[00:33:17]  Bruce: Okay. So the very fact that you're challenging me at all means that in your mind, science does not equate to the set of empirical theories, that you had something else in mind when you were thinking about the demarcation boundary. What is it? And that's what I ask people. What is it? What is it that you think the demarcation is if it's not empirical versus non-empirical? Well, that turns out to be a basically impossible question.

[00:33:42]  Cameo: Yeah, but even impossible questions, especially on Twitter, people will try and give you a, you know, a confident answer.

[00:33:51]  Bruce: So I had someone give me a good answer. He said, I can't remember exactly what it was, but he came back and he said, I think science means this and empirical means that. So then once he had, I said, which of those is Popper's demarcation? And suddenly realized, oh, it's actually the empirical. It's not the one I was, it's not between science and non-science. It's between empirical and non-empirical. And that was my point, right? Is if you do come up with some sort of definition of science, that's distinct from empirical theories. That will not turn out to be the demarcation Popper was talking about because he was actually talking about empirical versus non-empirical.

[00:34:26]  Cameo: So this quote you have here, why is this not a more appreciated statement? Does he counter this more than—I mean, you make it sound like… I'm just surprised people are fighting back against this so much.

[00:34:41]  Bruce: You know, keep in mind that this is probably mostly laymen, like myself, that it's just a bunch of laymen talking.

[00:34:47]  Cameo: Just a bunch of laymen arguing.

[00:34:51]  Bruce: But understand that there's a legitimate problem here. And let's talk about what the problem is so that we can make it very distinct. The problem is that string theory is part of science in most people's minds, and yet it is not empirical. Nobody doubts string theory is not an empirical theory. Nobody is arguing—there's no scientists out there arguing: “String theory in its current form is an empirical theory.” But there are many scientists out there arguing, as we saw with Penrose, by the way, using Penrose as an example, which was one of the problems I raised in the previous podcast. There are many scientists arguing, Popper is wrong that the only scientific theories are ones that you can refute, because we have theories like string theory or supersymmetry that can't be refuted, and they are part of science. So we have many scientists who are just familiar enough with Popper to know that it's supposed to be the boundary between science and non-science, and that it's based on refutability.

[00:35:47]  Bruce: What they've missed is that Penrose is not going to tell you that string theory or supersymmetry are empirical theories because he knows they're not. And this is actually one of the resolutions to one of the problems that I raised. This one's a little bit more complicated. I'll get to it in a second. It's the fact that the word science, to most people, means something different than the set of empirical theories. Because we have basically any early theory in science before you can make it. The goal of science is usually to make a theory testable. So the gold standard of science is that the theory is empirical. But when you're initially coming up with a theory, it's often not empirical. It takes you a while to figure out and refine that theory to the point that you can figure out tests for it. And sometimes it takes a long time. String theory still has no empirical consequences that we know of.

[00:36:39]  Cameo: And sometimes we just might not have the technology yet to test the theory.

[00:36:46]  Bruce: By the way, that's a good example of how you can't easily differentiate between the theory and the theoretical system that's attached to it. Because a theory may be non-empirical for its current theoretical system, but as instruments change, that part of the theoretical system changes, and now that original theory that was non-empirical has now become empirical. There's no real distinct boundary between the theory proper and the rest of the theories that it relies on. So this is actually the answer to Penrose's problem, is that when he said Popper's wrong about refutability because there are theories in science that aren't refutable, but they're legitimate science, he's simply misunderstanding that Popper was actually talking about empirical theories, in which case the two are in agreement. But because that is what people know Popper by rumor, if you will, right?

[00:37:37]  Bruce: I mean, this is all a little hard to tease out when you're just trying to read through the book. And most people don't read his logical scientific discovery. They start with a different book or they go off of some summary of someone.

[00:37:50]  Cameo: What book do they read?

[00:37:51]  Bruce: You know, I read David Deutsch first, and that was my introduction to Popper. And at the back of his book, he recommended, I think, Conjecture and Refutation and The Myth of the Framework. And those were the first two I read. I did not read Logic of Scientific Discovery until years later. In fact, fairly recently, in fact, because I'd always heard it was hard. And by the way, it is.

[00:38:10]  Cameo: Well, just that one thing that you quoted on the last page is very difficult to read and parse through. That's heavy reading.

[00:38:24]  Bruce: However, the Logic of Scientific Discovery is the only actual book that Popper wrote. The rest of his books are actually collections of speeches or papers. They don't have a centralized theme usually. Whereas the Logic of Scientific Discovery was Popper saying, I'm going to lay my theory out in full. And it's the only place he ever does. The science and the realism and the aim of science is him clarifying things, so it's another good one. But there's only one book where he actually lays out his theory in full, and that is The Logic of Scientific Discovery. And I would suggest as a layman Popperian who only recently read Logic of Scientific Discovery, everything we're talking about now is me having read that book and going, oh, I misunderstood that. Oh, I misunderstood that. Oh, I misunderstood that. Right.

[00:39:10]  Cameo: Either you misunderstood because you were reading a summary that somebody else who also misunderstood or-

[00:39:18]  Bruce: I was reading Popper talking about stuff more in the abstract. I didn't have some of the details I needed to- And… Because he picks words that are a little misleading in English, and that's the main point I'm making, it's natural that I misread him.

[00:39:34]  Cameo: Yeah, absolutely, because you needed to do a deep study of his actual philosophy and his actual theories.

[00:39:43]  Bruce: Let's go back to Penrose's example of supersymmetry. His argument was supersymmetry, and he also mentioned string theory, are important parts of science, yet they cannot be refuted, ergo, Popper was wrong. But if the boundary is empirical versus non-empirical, then actually Popper and Penrose are agreeing. Those are both non-empirical theories, period, end of story. So this is no longer a disagreement between them, it was just a misunderstanding on Penrose's part, in part because of the fact Popper chose to refer to it as the boundary between science and non-science. Even when he clarified empirical science, I think people just missed that clarification, right? It's so easy, but the word science just has a broader range of meanings. It's not always the set of theories that are empirical. Yes, string theory is a part of science. Yes, supersymmetry is a part of science.

[00:40:33]  Bruce: They're just not part of the empirical part of science.

[00:40:36]  Unidentified speaker: Yeah.

[00:40:37]  Bruce: Okay. So by insisting on the demarcation science versus non-science, we set up a war of words. And this is why I don't think we should do that anymore. Where really Penrose is always going to claim, string theory and supersymmetry are a part of science. I've seen Popperians try to say, and non-Popperians too, people that agree with Popper but not necessarily Popperians, say, oh, string theory is not science. I've seen people say that. It's like fighting words. There's no point in saying that. Of course, it's part of what we call science.

[00:41:06]  Cameo: We're just saying it's non-empirical.

[00:41:10]  Bruce: If you clarify what I mean is it's non-empirical, no one will be able to argue with you.

[00:41:17]  Cameo: Show me how you're testing your theory. Do you have a special string theory machine?

[00:41:25]  Bruce: Now, I just said there's no one who can argue with you, but actually, you could make one argument here. And I'm going to now clarify that argument. So Sean Carroll, he gave a, and I put this out on Twitter for comment from people. He attacked Popper's theory, and he attacked it on exactly the grounds I just gave, that there are theories like string theory, that in theory we can verify them, but we can't refute them. And then he said, but what if we actually do find one of these particles? What if, to use the Penrose example, I don't think this is the example he was using, very similar. Let's say that we get bigger and bigger colliders, and then we actually find super partners. Okay. Wouldn't that then verify supersymmetry in some sense? I mean, obviously it can't in any absolute sense because there may be some other theory that also predicts super partners.

[00:42:15]  Bruce: So you can't verify it as completely correct. But wouldn't it give us some sort of support for the theory, is what I'm trying to say? Yeah, of course it would. Okay. And this is what Sean Carroll then used to say, this proves Popper is wrong. And he would go a step further. That was empirical, Sean Carroll would say, because we actually did an experiment and we found it. Okay. Well, now it depends on what you mean by empirical. But let me make a case that's not a great way to use the word empirical. I'm not going to deny the word empirical could be stretched to include that case. And Popper doesn't deny that either, by the way. I can give you quotes specifically from Popper where he admits this. So he says, I admit a system as empirical or scientific only if it is capable of being tested by experience. The key thing here that we're talking about is when Popper says testable by experience, he really means I can set up a program for test and get a definitive answer.

[00:43:07]  Bruce: So yes, I could build larger and larger colliders. And in theory, maybe I find super partners at some point. And yes, in theory, that was a test in some sense of string theory or supersymmetry. But really, nobody is going to think that these are fair tests because every single time you don't find something, you just wait for the next collider and then you see if it finds something. And if that doesn't, then just wait for the next collider. There's no definitive answer ever until you actually find it. So it's not a definitive test. It's not a test in that sense. Popper, when he talks about a test, he means it in the normal scientific sense of I'm going to do this test and it's going to distinguish between theories. That's what he means. He doesn't mean test in its looser sense. Who knows, maybe we'll find it by chance. And this is really what he means by empirical, is we can set a program for tests, we can do this test, we can definitively decide was this correct or not.

[00:44:03]  Bruce: And that will always be only on universal and thus refutable theories. And that's not normative. That's just a statement of fact as to how these things work, logically. You always would have, if you're going to have a definitive test, then you have to have some sort of universal statement that says, okay, under these very specific spatiotemporal conditions, you will find this result. And if you don't, the theory is wrong. That's an empirical test in Popper's mind. So an existential theory that has no attached constraints like that can never have a program for test. It just can't, just it's logically impossible for it to have it. So constraints by definition imply universal laws. And universal laws have to be refuted because that's the way logic works. Ergo, empirical equals logically refutable, done. Everything else you say may be true. It may be true that you could find super partners by chance.

[00:45:04]  Bruce: It may be true that when you did, that would be an important discovery. I'll talk about what that means in a second. But you're not talking about what Popper meant when he said, empirical equals logically refutable. What he said, what he meant when he said that is a statement of logical fact. And there is no way around it. It's not normative. It's not science should be refutable. It is every time you do science, or rather you do a test in science, it will be because you have some sort of constraint. That constraint is a universal law. Therefore, it will be refutable and not verifiable. That's what he means. And he is right. And there is no way around that fact. Now, let's talk about how this gets a little weird. Okay, why he's right, but why it feels like he's wrong sometimes. Popper gives the example of the theory that the element with atomic number 72, which is hafnium exists.

[00:46:00]  Bruce: When stated in that way, that is an existential statement and it can only be verified. Popper points out though, that this theory was never actually verified, until the theory had developed enough that there was reasonable constraints that helped us understand enough about the element 72 that we could reasonably go look for. He says all attempts to find element 72 were in vain until Bohr succeeded in prediction several of its properties by deducing them from the theory. But Bohr's theory and those of his conclusions which were relevant to this element are far from being isolated purely existential statements. They are universal statements. Logic of Scientific Discovery, page 49. This creates this weird circumstance where we speak of a theory, we use linguistically, we may speak of the theory that atomic number 72 exists. But what we really mean is something more complicated than that.

[00:46:54]  Bruce: We mean we're going to go find atomic number 72 and part of the theory are the following constraints and therefore we should do this following test to find atomic number 72. Well, the part of the test we're doing, that's going to be a series of universal statements because there are constraints, because that's what universal statements and constraints are, and the same thing. [Closing wording unclear.] What you're really going to do is you're going to try to refute that part of the theory. That's what you're really going to do. When you then do that, find element number 72, you're then going to announce to the world, we verified the theory that element number 72 hafnium exists. There's nothing wrong with that, and it's not counter to Popper. And this is where I think Popperians make a mistake. They've downplayed the language of verification so much that they've missed the fact that Popper was never talking about the language of verification or even had a problem with it.

[00:47:41]  Bruce: We've already seen multiple examples of where Popper uses the word verify. He just didn't have an issue with it. He's saying something very specific, which is, look, if you're going to try to test this theory, it has to have constraints. Those are universal. Those will have to be refutable. And that's how you're going to set up your test. And once you've done that, yes, then you can announce to the world, we verified atomic number 72 exists. Nothing wrong with that. That's not counter to his theory at that point. You've simply, if you think it is, you've simply misunderstood what he was actually saying. And in this case, though, what Popper would say is, look, even in that case, what was the alternative theory? Since this theory is being worded as an existential statement, even though it's not really, it's more complicated than that. Its counter theory would be, there is no atomic number 72.

[00:48:25]  Bruce: When you found atomic number 72, you refuted that universal statement. So even in a case like this, you're really still always refuting things. And it's really the refutations that matter. What mattered here was you killed the alternative theory through a refutation. Now, the reason why this is weird is because existential statements, when we word a theory that way, their competitor is basically just the inverse of it. The theory, atomic number 72 exists, its competitor is that it doesn't exist. That's just not true for most scientific theories. The theory, Newton's theory, its inverse is not Einstein's theory. This is something that David Deutsch always points out. So in the scientific realm, except when you happen to be wording things like an existential statement, the inverse of the theory is not its competitor. In the one case where we're wording the theory as an existential statement, which sometimes that is just the most convenient way to do it.

[00:49:21]  Bruce: So you should. In a certain sense, the theory you really care about is its inverse, that atomic number 72 does not exist and you're refuting that. In other words, Popper's theory still works. There's never a case you're going to come up with where Popper's theory doesn't work. What you will come up with though, is cases where it's much easier to word things as a verification. And there are probably plenty of cases like that. But those aren't refutations of Popper's theory. That's just a linguistic misunderstanding of Popper's theory. Are you with me on that? Do you follow what I'm saying here?

[00:49:55]  Cameo: I do, yes.

[00:49:56]  Bruce: So we already kind of talked about the fact that I get a lot of pushback on this. But I do think what people are going to find is you just can't make sense of the boundary condition, the demarcation, unless you're assuming science equals empirical equals testable, equals the test gives a definitive answer. Those have to be understood as the same thing. If you understand one of those is not the same thing, I don't think Popper's epistemology makes sense anymore.

[00:50:20]  Unidentified speaker: Right.

[00:50:22]  Cameo: You almost throw out the baby and the bathwater with the whole thing. If you can't resolve that problem and that.

[00:50:31]  Bruce: Let me use an example that I think I have in our coming up slide, but I think would fit better here. Let's take the theory Bigfoot exists. I should probably note here that that wouldn't even be considered a theory under the way Popper would normally think of a word theory, but I'm using theory in its looser sense, and there's not really a problem with doing that. Popper's theory still applies. And I intentionally word that theory as a purely existential statement. First of all, that's a non-empirical theory. Bigfoot might exist or might not exist, but there is no program for testing if Bigfoot exists or not as of today. Okay, because right now that theory is just an existential statement. So that would be an example of why that theory is not part of empirical science, is because in its current form, it simply can't be tested, so it's not empirical.

[00:51:16]  Bruce: Now, let's say we had a theory of Bigfoot that told us enough about Bigfoot, that we could say Bigfoot exists and he will be found within this two-mile square radius of this forest. That's dumb, but let's say we could based on what we knew about Bigfoot. At that point, the theory Bigfoot exists now has enough universal statements attached to it that we can realistically go test that theory. At that point, that theory becomes empirical. We would then go out and we would go search that two-mile square radius. Now, let's say we found Bigfoot. That would then refute the theory that Bigfoot does not exist, leaving only one theory that Bigfoot does exist. Let's say on the other hand, we didn't find Bigfoot within that two-mile square radius. At that point, we would have refuted the theory that Bigfoot exists because that current theory included the empirical facts that it should have been within that two-mile square radius.

[00:52:11]  Bruce: Do you see how there's two different possible types of refutation at that point? One where we conclude Bigfoot does not exist, and one where we conclude that he does exist. Now, this may sound like a silly example, but this actually happened with the Loch Ness Monster, believe it or not. The Loch Ness Monster, by definition, has to be in Loch Ness. Yes. So that means that now we have a specific temporal spatial area to search that's at least somewhat reasonable. So they took out boats with sonar mapping and they mapped the whole thing, and there was no Loch Ness Monster. Now, maybe we can make a little bit of wiggle room here for true believers of the Loch Ness Monster, that the monster was super smart, it knew how to swim around the boats that were doing the... Maybe we can make that much of a room for them. Okay. That's definitely at this point a complete ad hoc save, which by Popper's epistemology, we always discount.

[00:53:06]  Bruce: The fact is, though, is that we actually have refuted the idea of the Loch Ness Monster exists. And the reason why we were able to do that was because we found a way to make the theory testable, which meant it had to have universal laws attached. Yes.

[00:53:23]  Cameo: Poor Loch Nessians.

[00:53:25]  Bruce: How is this tied to induction? Everything Popper talks about is all tied together. This is something that I think a lot of people miss, is they got this idea of his epistemology, and they think it's distinct from the fact that he has this boundary condition on science, and I think that's distinct from how he solved the problem of induction, but they're all really all part of one program. So how does this all relate to induction? So what we need if we're going to do science and make it empirical, keeping in mind that in Popper's mind science equals empirical, is we need a way to learn about the world through observations. Now, David Deutsch would be the first to say that you can't learn about the world through observations, that's empiricism. It's not quite true, it's almost true. We do need a way to learn about the world through observations, because we've got observations available to us.

[00:54:09]  Bruce: Empiricism, by the way, is really the idea that observations are infallible, which is not true. That's the part that's really false about it. So induction was this fictional form of logic that allowed you to flow from observation to general law. So I've seen 100 million white swans, I can now deduce that all swans are white. Now, obviously, that's just logically false, because you can't make that assumption no matter how many white swans you see. So what is the relationship then between observation and general laws? This is really what Popper was trying to solve. So on everything I just said, I'm going to say it again, but in a different way. The idea here is that no truth flows. If what you're trying to do is you're trying to demonstrate that all swans are white, no truth flows from the observation that you have a white swan to the rule all swans are white.

[00:55:04]  Bruce: You cannot deduce anything that way. But you can deduce that that theory is wrong by having a non-white swan. And that is also an observation. So that's a case where we can deduce something about universal laws through an observation. And that's what science needs to do. So science needs a way to do that. Therefore, it has to be refutation. That was how he figured this out. So here is Popper's statement from page 19 of Logical Scientific Discovery. Such an argument to the falsity of universal statements is the only strict deductive kind of inference that proceeds as it were in the inductive direction that is from singular to universal statements. This is the relationship between induction and how Popper addressed induction and the rest of his theory. He was trying to work out how do I learn something about universal laws through an observation?

[00:55:52]  Bruce: Oh, I see I can do it through refutation, through a logical [refutation]. So Popper's epistemology captures the only logically deductive way we can learn from observations. Okay, now, can you strengthen the theory? Many Popperians claim you cannot strengthen the theory. I think they're technically correct if you happen to understand the word strengthen in a certain peculiar way. By now, you should probably be used to me the fact that it depends on what you mean is always my answer. I think that when understood as can the theory become more sure? No, it can't. There is no way to make a theory more sure. But I would have to ask the question, what do we really mean? When somebody says, I saw this observation and it strengthened this theory. What do they mean? Are they meaning this in the inductive sort of way? I don't think so. So Mark, which is Sadia's husband, he threw out the example on social media.

[00:56:45]  Bruce: He said, listen, what about Brownian motion? When we discovered Brownian motion, how can you say that didn't strengthen the theory that atoms really exist? Because at the time, there was the belief that atoms were actual physical things and the belief that they were just kind of instrumentalist, numerical things that we use, but they didn't actually exist in real life. And Mach was a super famous scientist that backed that idea. And he said to Andrew Croshaw, I said Crenshaw last time, his name’s actually Croshaw, how can you say that doesn't strengthen the theory? Well, they were talking past each other. What Andrew meant was nothing about that theory changes just because Brownian motion was found. What really changes is that its competitor is no longer viable. And it's now the last remaining theory. My point of view is, is that when Mark says it strengthens the theory, that's what he meant.

[00:57:38]  Bruce: And so trying to tell him, you can't strengthen the theory when clearly you can, even though what we want to do is we don't want to tell Mark, you can't strengthen the theory, which is what Andrew was saying. We want to tell Mark, yes, you can strengthen the theory. But what you mean in this case is that Brownian motion made Mach's version of atoms non viable. So there was only one competitor left. There was only one theory left that atoms were real things, that they physically existed. That's what we mean by strengthen the theory. I think there's one other thing we mean by strengthen the theory. I think you can also strengthen the theory by making it more testable. Now, this depends on what you mean by theory. But if I were to look at Darwin's original theory of evolution and then compare it to modern synthesis today, the way it ties in with DNA theory and just all the things we can do today, one could argue it's the same Darwinian theory, or one could argue that we've long since refuted Darwin's original theory and replaced it with something far better that still had the heart of Darwin at it.

[00:58:39]  Bruce: It depends on what you mean by theory. If we're going to be really technical in the logical sense, which is what Popper was doing, then we have to understand that every time we improve a theory to make it more testable, even if that means we've improved the instruments that are attached as part of the theoretical system, that in a certain sense, we refuted the old theory and we now have a better one. Because of that, the idea of improving a theory by making it more testable is actually a special case of case number one, which is strengthening a theory means that you've killed off its competitors. However, because that feels different enough to most people, I'm not going to insist on that. So let's just say there are two ways to strengthen a theory. One is you kill its competitors, and the other is that you make your theory more precise so that it's more testable, which means it's more empirical and has more empirical content.

[00:59:28]  Bruce: That these are the two senses in which you can strengthen a theory. I don't think any Popperian would argue with me that you can do those two things, that they're important. I think what they would argue with me is that they would tell me I shouldn't use the word strengthen to call those two things strengthening of a theory. I'm simply disagreeing with them. That's essentialism. They can choose what words they want. I can choose what words I want. I think mine are easier to understand, so I'm going to go with mine, especially when I'm talking to non-Popperians. Yes, you can strengthen a theory, but you must understand that what you're really doing is you're either increasing the empirical content of the theory, or you're killing off its competitor. And I can give an example. And I can always give an example because it's always one of those two things.

[01:00:06]  Bruce: That's all ‘strengthen’ [unclear] actually mean. So we gave the example of Brownian motion. Now, it may be that some people, when they say strengthen the theory, they are thinking in terms of inductivism or surety. They may be thinking, we've now proved, Brownian motion happened, we've now proven 100% that atomic theory is correct. If that's what they mean, then yeah, they're wrong. And maybe some of them do mean that. So it may be worth clarifying what they mean. When you say strengthen a theory, do you mean this or do you mean that? I think strengthen the theory means that we've killed this competitor. Is that what you mean? We can talk to people. We don't have to jump down their throats over what word they used. If that makes sense.

[01:00:46]  Cameo: Are you sure we can talk to people?

[01:00:49]  Bruce: It's sometimes quite hard, I admit. But for me, this is what it means. Okay, so now with all this in mind, I'm now going to resolve all the problems and then we'll be done. Okay, so let's go through each of the problems that I originally mentioned in the previous podcast. So, Kuhn, nevertheless, anomalous experiences may not be identified with falsifying ones. Indeed, I doubt the latter exist. If any and every failure to fit were grounds for theory rejection, all theories ought to be rejected at all times. Can you, based on what we've said, can you explain how to resolve this problem using Popper's actual theory? And hint, it's right on the screen, but anyhow.

[01:01:24]  Cameo: Anomalous experiences may not be identified as falsifying ones.

[01:01:29]  Bruce: So when he says anomalous experiences, he means counter-examples, falsifying ones. He means ones that target a specific theory, which, as per [Duhem and Quine?] don't exist. In other words, he's just misunderstanding Popper's theory, because Popper's theory is actually about what you can do with anomalous observations with counter-examples. What he's calling anomalous experiences is what Popper meant when he said refutations.

[01:01:52]  Cameo: It's just the same problem with understanding those particular words.

[01:01:59]  Bruce: Yes. If you understand Kuhn as meaning by refutation, the theory itself, then he's actually right. But that isn't what Popper meant. So the two men are not actually disagreeing. They're actually saying the same thing. Tentative refutation versus tentative support. So Deutsch said the asymmetry between refutation, tentative, and support non-existent in scientific methodology is better understood in this way by regarding theories and explanations. First of all, Deutsch is using his terminology different than Popper's. To Deutsch, a refutation was a combination of an anomalous observation, a counter example, plus a rival theory. So it's something that happens later in the process. Whereas to Popper, it was the original anomalous observation, but the theory that was being refuted was the entire theoretical system, the theory plus the background knowledge.

[01:02:48]  Bruce: That's the first thing you have to understand, to understand what Deutsch is saying here and how it differs from what Popper is saying. Now, Deutsch is also mistaken a little bit here. He says you cannot support a theory. Now, the way he's taking the word support, maybe that is true. He never clarifies what he means by support, but you can absolutely support a theory through corroboration. Yes, there is such a thing as support. If someone were to read this statement, I think what I would suggest to them is Deutsch is correct, but only if you understand support as meaning prove a theory completely true. That is truly non-existent. And that's what I think he had in mind at the time. And you have to also understand that he's talking about a different sort of, it's related but different asymmetry, and that Popper's original asymmetry was actually a logical one, and that there's actually a little bit of a disagreement here between Deutsch and Popper over this.

[01:03:42]  Bruce: I won't get into the disagreement. I have a blog post where I talk about it. But Deutsch is taking this a somewhat different direction. He's not necessarily wrong. And there's not necessarily a disagreement at the conceptual level, but Deutsch is downplaying the logical aspects of Popper in this paper. And it even says that we shouldn't understand it that way. Where I don't think you can understand Popper's original epistemology without understanding it in that way. I think Deutsch is getting there by a different path, though. And I think ultimately you end up with the same results, no matter which way you try to conceptualize it. So now let me also point out that I think when he uses the word support, that maybe wasn't the best word. If he had used the word verify, that would have been a better word. Support and verify aren't the same thing.

[01:04:27]  Bruce: You may think of them as the same, but they're not. So let me give an example of light wave theory versus light particle theory. One time we thought we had a theory that light was waves, and we had a theory that light was particles. We went out and we found a crucial test between those two theories. And what we did is we showed that there was a refraction effect that was an interference pattern. And we saw that as verifying that light was a wave, because waves have interference patterns and particles don't. So we thought of it as a verification of light wave theory. And at that point, because we had “verified” light wave theory, we saw light particle theory as refuted. And so it died. It almost died. It started to disappear as a theory, until Einstein showed the photoelectric effect. Now the photoelectric effect demonstrates that light behaves in quanta, which is the same as saying it's in discrete particles.

[01:05:23]  Bruce: And so he had now come up with an experiment that quote verified the light particle theory and supposedly refutes the light wave theory. So they had to resurrect the light particle theory. And now they had two experiments, one that quote verified one theory and one that quote verified another. I want to suggest that in reality, we verified neither at all. What we really did is we refuted both. And we know that's true because the correct theory, the more correct theory anyhow, is quantum mechanics, which has some aspects of wave and some aspects of particle, but under different circumstances. And that's the true nature of light. The whole idea of superposition plays a role here. So we may have thought of it as this experiment verifies light-wave theory or this experiment verifies light-particle theory. But when we look at it in total, really what we did, the theory that showed the refraction effect, the interference pattern, that refuted the theory, light-particle theory.

[01:06:18]  Bruce: And then we had photoelectric effect that refuted the light-wave theory. Once we had neither theory, we were forced to come up with a new theory that was not refuted by either experiment. And this is how we may speak of verifications, but we really mean at best support. We could say that one of those supports those and one supports the other. That would be a true statement. But what we really mean is we mean refutations. And this is a weird thing for people to wrap their mind around, that we may use words like support and verify, but what we really mean when we really get down to it, when we're looking at it at the logical level, we're really talking about refutations. And we're always talking about refutations. We're always talking about a competitor dying, even if the competitor is an improved version of itself versus a less improved version of itself.

[01:07:02]  Bruce: Okay, corroboration and support. Support, so I raised the issue that this is related to the Deutsch one. Really, Deutsch is just picking a bad word here, in my opinion. Corroboration is a kind of support. Support is always just tentative though, as Popper says. It's always temporary until we think of a good [refutation]. We go out and we test our theories as best we can. We try to refute them. We may fail to. That we call that corroboration. That plays an important role in Popper's epistemology. You can't really say support plays no role in Popper's epistemology. It's just not true. Now, what is it that it's doing? Now, this is where Bayesians make a big deal. They say, look, you're claiming as a Popperian that the epistemic status of the theory can't change. It can't strengthen. I just admitted that that's not true, but often Popperians claim that wrongly.

[01:07:46]  Bruce: Whereas we're saying it can. We're saying you can have positive evidence, you can have negative evidence. So our theory is better than yours. Okay, let's clarify this now. What's really going on? When we make a corroboration... So let's use an example of Sadia. She doesn't like the Many Worlds interpretation theory. She still won't say it's the best current theory we have, but she's starting to favor other more speculative theories in science based on certain problems that she perceives. And we had a whole series of podcasts where she got to explain her point of view on that. Let's say that we invent a time machine. So right now, we already have a best theory of quantum mechanics in its many worlds. So in a certain sense, those two theories are one and the same. And they are. I know a lot of physicists disagree with me, but they're wrong.

[01:08:29]  Bruce: It's actually not that hard to demonstrate that this is the case. Let's say though, but Sadia's point is not wrong either. There are problems, and it may be that one of those days, those problems force us to come up with a theory that does away with the Many Worlds aspect of quantum physics. That's not a completely logical impossibility. That's what Sadia is hoping for. Now, let's say though that we corroborate the theory in the following way. Let's say we invented the following. This is coming from David Deutsch's Fabric of Reality, his chapter on time travel. Let's say we invent a time machine, and when you do, it works exactly the way quantum mechanics claims it should, where you go back into a moment of time, but at that moment, that reality branches off from the original reality. You go back in time, you kill your grandfather, so you're never born in that reality, but it doesn't cause you to disappear back to the future, because you existed in a timeline where your grandfather continued to exist.

[01:09:27]  Bruce: This is the Many Worlds version. We had a whole podcast on this, on the Deutsch proposition, which doesn't really exist, but came up in Avengers. Let's say we do this. Let's say we do an actual experiment. Now, no one in their right mind would not see this as strengthening the theory of many worlds quantum physics.

[01:09:48]  Unidentified speaker: I agree.

[01:09:48]  Bruce: What do we mean in this case by strengthening? Okay. So let's use Sadia as the example. She sees this experiment. She's doing research into other alternative theories that explain everything in quantum physics, but don't have a many worlds aspect. They're single universe, and she's trying to do that. She's—up to this point… She hasn't come up with a good alternative theory. We don't know. The fact that she hasn't come up with one doesn't mean one doesn't exist. It could mean that there isn't one, and that's why she will never actually be able to find it because it just doesn't exist, because reality isn't that way. There actually is multiple universes out there. Then she'll never come up with a good alternative theory, but there's no way for us to know that for sure. We've now corroborated many worlds by actually entering an alternate universe that we can go observe.

[01:10:35]  Bruce: There's no way Sadia doesn't kill her research program at that point. This is what corroborations really do, is they make it harder to come up with alternatives. So there's still cases of refutations. It's just that the refutation may not be an actual existing theory, it may be in its nascent research program state. And as we do corroborations, and we talked about this with evolution, I said it was only until I reached the 100th problem in evolution that I really said, oh, now I really get why this theory is so important. And if you're a naïve creationist, it's not that you don't understand the theory, you just haven't dealt with enough problems at this point to really understand why your alternative theory is in fact not an alternative theory. It's the same thing for everything, right? As you start to get corroborations, as you start to have, I stress tested this theory and this happened, it starts to close down alternatives.

[01:11:30]  Bruce: And as those, even if you only have one problem, it's not that hard to come up with some sort of ad hoc [save?] that really sounds reasonable and that maybe you can see as a research program and you're going to go try to come up with this alternative. This is also why refutations in the sense of actually doing away with the theory never happened in science. What really happens is people start to give up on the alternatives until finally there's only one left. And that takes time. Every scientist has to have their chance to say, I think maybe it could be this, state it, have someone try to test it. But as each test happens and as the alternatives continue to die out, people will start to fail to come up with alternatives. This is what corroboration does, is it strengthens a theory by killing nascent alternatives, killing either real alternatives, if it's a crucial test, or alternatives that we're working on.

[01:12:26]  Bruce: Alternatives that at this point are maybe not even legitimate theories, they're just ad hoc theories. We think maybe we can turn into real theories, then some corroboration comes along and we go, oh, no, that one's dead. That's what's really going on. This is what an actual strengthening of a theory is. This explains what corroboration is, why it's an important part of Popper's epistemology, why you cannot do without it, and why testing is so dang important to science. The Popperians that I've talked to, I've quoted Brett Hall as downplaying corroboration. This is why I believe he's wrong. I think he has missed this connection, that corroborations are in fact a kind of refutation too. Therefore, it's refutation all the way down. That's true. Sometimes it's easier to use the language of support, corroboration. It makes more sense to people to say, the fact that we entered an alternative universe using time travel corroborates many worlds, rather than to say it kills these following nascent theories.

[01:13:24]  Bruce: It would be awkward to try to phrase it that way.

[01:13:28]  Unidentified speaker: Interesting.

[01:13:30]  Bruce: Okay. The language of support should not be downplayed by Popperians but needs to be understood in a refutational context because it always takes a refutational context in Popper's theory. That is the right way to think of it. That's why Bayesians are wrong. The fact that Bayesians haven't figured out that positive evidence is actually negative evidence for nascent theories, is one of the many things that Bayesian epistemologists have gotten wrong and why their theory can't produce good results and never will. They are wrong about their theory. I don't know what else to say.

[01:14:04]  Cameo: That's as clear as you can be.

[01:14:05]  Bruce: Okay, so ultimately, [corroborations] are still about refuting competitors. This is the Bigfoot example that I just gave. This is an example of theories that are required to verify. Yes, there are theories that you are required to verify. But ultimately, the empirical aspect of the theory is always about falsification, and that was Popper's point. So now we've resolved this problem.

[01:14:29]  Unidentified speaker: Do you agree?

[01:14:30]  Bruce: Can I move on past this one?

[01:14:32]  Unidentified speaker: Yeah, absolutely.

[01:14:33]  Bruce: Okay, Penrose's challenge to Popper. Now, I mentioned part of this already, that when Penrose uses his supersymmetry and superpartners example to disprove Popper's epistemology, first of all, he's mistaking that it's not the demarcation between science and non-science. It's actually demarcation between empirical and non-empirical. If he had understood that, he probably wouldn't have used that example. Let's now address Sean Carroll's question, which is, but what if we actually do find it? Okay, let's say we build a large Hadron collider, and we actually find superpartners at this point. Can you really say that that theory is non-empirical? Well, it depends on what you mean by empirical. It's still not empirical in the sense that we never had a test that was definitive.

[01:15:18]  Unidentified speaker: We just by chance happened to find it.

[01:15:21]  Bruce: Okay, what's the status of the theories there? So now consider the two competing theories are, on the one hand, regular physics, let's say the standard model, because that's the model that talks about particles, and its competitor is supersymmetry. Standard model without supersymmetry, and standard model with supersymmetry or string theory with supersymmetry. It would probably be seen as creating some evidence for string theory, because string theory, you don't have to have supersymmetry with string theory, the relationship is more complicated than that, but the two have some sort of theoretical tie. So let's just say that because theories and their inverse, theories inverse isn't usually its competitor, that we have a weird circumstance here, where we find a superpartner and we go, well, the standard model doesn't say there aren't superpartners, so this doesn't refute the standard model.

[01:16:14]  Bruce: You could make that argument. In fact, Andrew Croshaw does make that argument to me. He's a very good Popperian. I'm actually not completely sure I'm right about this. Let me give both points of view because both are productive. Andrew's point of view, if I understand it correctly, is that, and I may not, I'll actually bring him on to clarify himself at some point, if I can convince him to come on the show. He says, look, what you've done is you've produced a way to produce superpartners, and that's an important theoretical find. But you really haven't refuted the standard model, but that doesn't matter because any more than say you've refuted evolutionary theory. Evolutionary theory doesn't say if you have superpartners or not either. So if evolutionary theory doesn't say it and standard model doesn't say it, then clearly you haven't refuted it.

[01:17:03]  Bruce: So you can't really say you've strengthened supersymmetry. Well, I don't agree. Let me just admit he's making a good argument, so I'm not sure I agree with myself. I could probably be persuaded otherwise. Let me make a case though. His argument that the status of the standard model hasn't changed in the same sense of the status that the biological evolution hasn't changed, because neither says anything about super partners, that's actually a pretty good argument and gave me a little bit of pause. Here's what I would say about them. In real life, what would actually happen? We know what would happen is people would announce, scientists would announce, we have found evidence for supersymmetry, and it has confirmed the theory or supported the theory, confirmed in the absolute sense is always impossible. I don't even disagree, confirmed in the sense of supported, tentatively supported.

[01:17:58]  Bruce: It supports the theory of supersymmetry. It provides support for string theory because string theory was built on supersymmetry more strongly than [standard model?] doesn't have to have. They would have immediately announced that. Now, my point of view is that they're not wrong because I think there's actually a tacit refutation going on. I just said there's no such thing as support unless it's tacitly actually a [refutation] of a competitor, although the competitor might be nascent. Now, Andrew says, no, that's not true. You have not supported the theory because there is no theory out there that you refuted. Okay, I'm going to make the case that there is. And here is what I think it is. In real life, not only would they announce that they now have evidence for string theory and supersymmetry theory, they would also announce an updated version of the standard model that now includes super partners because the standard model is supposed to contain every single particle.

[01:18:53]  Bruce: So they have to update it all the time if they discover new particles. The standard model, by the way, is kind of a weird theory in that most of it just comes by observation. They just go out and they test, what's the weight of this? And then they put that in just based on empirical evidence. They don't derive it from first principles like they wish. So it's generally saw as a fairly ugly theory by scientists, even though it's like one of our main most important theories. Scientists hate it, right? Inherently, they know it's ugly and it's just the best we know how to do right now. So they update the standard model all the time. Anytime something, if the god particle is found, then they have to go update the theory. Now remember, when you update a theory, you're really refuting the old theory and you're creating a new one. Do you follow my logic?

[01:19:40]  Unidentified speaker: Yeah.

[01:19:40]  Cameo: Well, all right, because you're saying we were wrong and here's our new understanding.

[01:19:46]  Bruce: Yes.

[01:19:46]  Unidentified speaker: Right?

[01:19:46]  Cameo: That's what you're doing.

[01:19:49]  Bruce: Now notice that we won't be announcing a new evolutionary theory based on the discovery of super partners. So in my mind, the discovery of super partners supports the theory of supersymmetry and string theory, depending on how you want to look at that, in that it tacitly refutes the existing standard model, because even though the standard model didn't say there are no super partners, it was tacitly saying that based on the fact that the standard model was supposed to contain all the particles, which is why they have to then go update it. So they're going to treat it like a refutation because it is a refutation. And that's the sense in which it strengthens supersymmetry theory, is that now there won't be a non- supersymmetry version of standard model anymore. That's gone. They will have to come up with, here's how supersymmetry ties to the standard model, which by the way, I think they've worked a lot about that out already.

[01:20:46]  Bruce: So in a sense, we've already refuted the original standard model, which is why this is kind of a weird example. But I think the idea of a tacit refutation makes sense to me. And I can kind of see why Andrew pushes back on me on that. And really, I'm going to throw that out there to the audience to form their own opinions. Now they've got both opinions on that one. But I think something more is going on. That's why I have to look for the tacit refutations, because I think it is the case that finding super partners supports supersymmetry theory. Therefore, as per Popper, something got refuted. I want to know what it is. I think it's the original standard model got refuted. I think it was a tacit refutation. Now, let me say that I don't have the quote here, but David Deutsch in Logic of Scientific Discovery, or sorry, the Logic of Experimental Tests, he actually says this.

[01:21:37]  Bruce: He actually says you can have a theory that doesn't say this won't happen, and yet becomes a worse theory because it doesn't explain something that does happen. Now, he was using this in context of quantum mechanics, where it's like probabilistic theories. So let's say one theory says you can get A, B, or C outcomes, and one theory says you will only get C as the outcome. And then you do the test and you get C, then you do the test again and you get C, and you do the test again and you get C. From a certain point of view, if you're understanding Popper too strictly, stricter than it is intended, you might say both these theories are still viable because neither has been refuted. What Deutsch is saying is that the first theory now has a basic statement that it can't explain, and therefore we treat it as a sort of tacit refutation. We say the first theory, which predicted it would be either A, B, or C, is now the less explanatory theory, and therefore we're going to go with the theory that actually explains why it's all in C.

[01:22:43]  Bruce: And so even though there's no actual definitive refutation, as you would normally think of it as a contradiction, there's a sort of a tacit refutation. I'm doing the same thing, but I'm using it in different circumstance. And I don't know if Deutsch would allow me to use it in that circumstance or not, but I don't see any reason why it can't be stretched into that circumstance and therefore explain why Sean Carroll is wrong. Why Sean Carroll's example that we might find these particles and that would strengthen the theory is at once correct, but that is still consistent with Popper's theory. Does that make sense? Do you see where I'm going with this?

[01:23:19]  Cameo: Yeah, I absolutely do.

[01:23:21]  Bruce: Okay, so now based on what I just said, let's take the best theories without refuting the competition. So [refutations] are problems, but [refutations] are problems, but they're not definitive [refutations]. Popper does in Logic of Scientific Discovery, he points out that probabilistic theories are a problem for his theory, although he explains how to solve the problem at length, like for chapters, he explains how to solve the problem. He says, you can't really definitively refute a probabilistic theory, just exactly like Deutsch just said. So he says, how do you handle that? Well, he gives this giant technical discussion that I didn't understand, and I don't think is even necessary for the most part for a layman like myself. The bottom line is that let's say you had a theory that, you know, the example of the outcome could be A, B, or C, but instead it's always C.

[01:24:09]  Bruce: And yes, you can explain that as we happen to get C every single time. Well, do the experiment 100 times, do the experiment a thousand times. If the theory that it can be A, B, or C is correct, and just by chance you keep getting C, that's not going to be repeatable. It will never become a basic statement. The theory that it has to always be C can't ever have a true basic statement. It won't be repeatable by a different group. It won't be repeatable if you do the experiment again. It won't be repeatable. I mean, it just won't happen, right? Because that's the way probability works. So the bottom line is that it never really ends up posing a problem for his theory. While in principle, if you take it too strictly, yes, it's a problem. In practice, it just never does, even when it's probabilistic like this. So now based on that, let's take the two examples I came down to.

[01:25:01]  Bruce: A man is murdered and his wife has a motive. So why do the police go to the wife first and start to investigate her first? Well, it's because that's their best theory. They haven't really refuted an alternative theory at this point. What they've really done is they have, let's say a basic statement, something like a basic statement. I don't know if this would technically be a basic statement, that this wife has a motive and they don't know of anyone else that has a motive yet. Well, that's good enough. That is enough to make the other theories problematic. Remember, refutations aren't definitive. They're problems and we don't know how to solve them yet. The other theories have problems that we don't know how to solve. So it makes sense that the first thing we're going to go do is we're going to try to take the theory that the wife committed the murder and we're going to try to make that a stronger empirical theory.

[01:25:45]  Bruce: We're going to try to strengthen the theory by making it more empirical. So we're going to try to figure out, was she there? Does she have an alibi? And during that process, we may end up refuting the theory that she committed the murder, in which case we will be then forced to go to the alternative theories that somebody else did and try to figure out who it is and figure out, did we miss somebody else having a motive? Nothing about this is actually at odds with Popper's theory. It's really just being too strict about the idea of a refutation. You have to understand refutation in the idiosyncratic way that he meant that term. If you ever understand it in a different way, then yes, Popper's theory doesn't make sense anymore. And then the DNA example was that we're trying to figure out if so-and-so is a descendant. Thomas Jefferson, in real life, we have the case of Thomas Jefferson where we wanted to know if he had a child by his slave.

[01:26:41]  Bruce: Because, of course, in our day and age, we want to know if he's a hero or a villain, right? So they do a DNA test, and they discover that there's a certain marker that the descendants of the slave have that Thomas Jefferson also has, and it's rare. Okay, I'm making up that it's one in 10,000 chance of [unclear]. This doesn't really definitively refute the idea that... There's nothing definitive here, right? We haven't demonstrated it is Thomas Jefferson's child. We haven't even refuted the alternative, right? It could be that his brother lived there. In this case, I think his brother did live there. But let's say the brother didn't live there. That's too obvious a case. Maybe the brother came by and on a chance encounter had a child with his slave. Maybe somebody else that was unrelated to Thomas Jefferson. Obviously, we haven't really refuted the competitors.

[01:27:37]  Bruce: What we've done is we've made a problem for them. At this point in time, yes, our best theory is... Again, let's ignore the true real life case. Let's pretend like this is a fictional case. There's no other brother in this case to worry about. We've only got the one person that happened to live with the slave. Yeah, our best theory is that it's his child. That's our best theory. But it's just a best theory. The other competing theories, they're problematic. They're maybe even bigly problematic. But you could see how you might save them if you could demonstrate, well, actually, he did have a brother and they lived together. Or there's ways you could go about, therefore, resolving that problem, which would then make the competitors better again. Again, this fits perfectly into Popper's epistemology. There's not a slightest ounce of problem here.

[01:28:28]  Bruce: Insofar as it seemed like there was a problem, it was really just because we were understanding the word refutation as definitive, when really we should be talking about it as a counterexample or a problem. Because that's really what they are. It's the problem that's objective, not the definitive refutation that's objective. Okay, that's it. I've now resolved all the problems with Popper's epistemology that I originally raised. It turns out all of them boiled down to you simply misunderstood Popper's epistemology.

[01:28:59]  Cameo: So you, of course, presented a very much shorter version of this today.

[01:29:05]  Unidentified speaker: Yes.

[01:29:06]  Cameo: Do you think that that's going to start conversation with some of these hardcore Popperians? Or do you think, like with David Miller, there wasn't sufficient time for you to really talk through, essentially, their refutations of your refutation theory? And that now they're essentially going to dismiss your concepts because they either are just too unwilling to hear a contrary theory, which you have said about Popperians in the past.

[01:29:38]  Bruce: Yeah. Okay. So first of all, I've said that about Popperians, but I would say that about any group of people anywhere.

[01:29:45]  Cameo: That's fair. People get passionate about their beliefs.

[01:29:49]  Bruce: Yes. So let me make a few statements about that, because that's actually a good set of questions. First of all, yes, my intent was to start a conversation. Realistically, how far will that conversation go? Probably not very far. I've got neither the credentials to gather interest. I don't even know how to go about really getting published. There are people who will dismiss me, that really probably you should be listening to, like Miller. Even though ultimately I feel there wasn't necessarily a disagreement there. I don't know what to say about that other than to say that's the way the system works. I've raised something that I think is a legitimate issue. At heart, the issue isn't a problem with Popper's epistemology. It's a problem with the way we present Popper's epistemology. Now, will David Miller take that to heart and present it a little differently so that people understand what he's saying better?

[01:30:41]  Bruce: Probably not. My guess is that won't hinder him that much. What it will really do is it will hinder the people trying to read him. Because really, when it comes right down to it, David Miller understands this. He's got a certain way of speaking that may be a little misleading to people outside of his circle. That's my real point. But he's not wrong. He's got conceptually the right concepts. David Miller would agree. None of these problems that I raised are actual problems. He would maybe want to describe them differently than I did. But ultimately, what he would describe would be the same as what I'm saying. We're just using different wording. It's kind of up to David Miller if he wants to stop, take enough interest, understand what I'm saying, and then realize, oh, there was something there for me to learn. This wasn't what I originally thought it was.

[01:31:30]  Bruce: And we all have to make that judgment call, because honestly, the spam to ham ratio for everyone is high. I mean, we do the 15-minute sniff test to decide if we're going to listen to this person or not, because we just don't have the physical time to really stop and understand every single viewpoint. And I don't see an alternative to that, honestly. I think we to some degree rely on the fact that if this person's right, eventually it will succeed. You know, how long do bad theories stick around in place of good theories? Sometimes very long periods of time.

[01:32:05]  Cameo: Hundreds of years.

[01:32:07]  Bruce: Right. Amongst brilliant scientists, right? To use an example of, I can't remember which guy, this is mathematician, not scientist, but the guy who discovered that there are different magnitudes of infinity. Cantor, I think, maybe I'm wrong. He discovers that there's, so like, if I were to tell you, hey, Cameo, did you know that there's different magnitudes of infinity, that some infinities are larger than others? What would you think of that?

[01:32:35]  Cameo: Kind of baffled by it.

[01:32:38]  Bruce: It sounds wrong, right? It sounds like there's no way it could be right, because infinity is infinity. So there was a guy who figured out that that was the case, that there's actually an infinity of infinities, each one larger than the next. And he came up with a proof that demonstrates that that is the case. He then published the proof and got made fun of by absolutely every mathematician alive, and died upset that he was looked down upon. And, you know, honestly, some people claim he kind of went insane over it, that he was attacked and all his peers thought he was crazy and made fun of him. He was right. And there are no mathematicians alive today. I shouldn't say that. There's probably are mathematicians alive today that would dissent over this, but they would have some sort of really off theory, okay, that they would be backing, like intuitionism or something.

[01:33:30]  Bruce: There are no mainstream mathematicians today, let me put it that way, that really have any doubt that he was right. And the reason why is because this proof demonstrates that he's right. Now, the people back then had the same proof, but they just couldn't get past the way they were thinking of the word infinity. So they weren't able to see, look, you can't just make fun of the guy, you've got to actually deal with his proof. Later mathematicians who didn't have the same biases, and that's one of the great things about the fact that you've got constantly a new group of people coming in. As they come of age, they study, they become a mathematician or whatever, and they don't have the same biases as the old group. They haven't already had it drilled into their head. They're going to see that and they're going to go, look, why are you making fun of this guy?

[01:34:17]  Bruce: You have to deal with his proof, and you haven't.

[01:34:20]  Unidentified speaker: Right, right.

[01:34:22]  Cameo: You just mocked him into silence.

[01:34:24]  Unidentified speaker: Yeah, you mocked him into silence.

[01:34:29]  Bruce: Unfortunately, at any given moment in time, we're always starting at the mock into silence phase, right? And I put something out there, and then I'm hoping somebody catches something useful out of it. Assuming I'm even right, you know, what if I'm wrong, right? I mean, but I should still put it out there, and I should see if somebody else noticed what I noticed, or maybe they'll correct me. Maybe they'll say, hey, here's what you need to understand, which is what they did to me today. And then I have to decide, are they actually telling me something substantial, or are they really agreeing with me, and they don't like my wording? Well, I've got to figure that out, and they've got to figure that out, right? And this gets back to why is philosophy, why does it actually matter even though I hate it, you know? It's hard to know if you've got the right legitimate problem.

[01:35:14]  Bruce: But even Andrew listened to our podcast after we talked about him last time, and he said, what you said was good, but it wasn't really quite what I meant. And so he clarified, even in a case where you have, it's not just that lots of philosophy is bad, but some of it's good, and it's the good that matters. He said, even the bad stuff matters, because a lot of times bad philosophy is trying to deal with a real problem. And I thought, that's really good. That's a good point.

[01:35:47]  Cameo: It's very Popperian, right? It's applying the same rules to philosophy that we're applying to empirical science.

[01:35:58]  Bruce: Yes. And this is one of the things that's interesting is, as much as we concentrate on Popper's theory in terms of empirical science, it is possible to extend Popper's theory, and Popper knew this, into other realms where we're not dealing with empirical science. For instance, instead of having counter examples, which he called refutations, you can have criticisms. And so a non-empirical theory may have criticisms that then have to be dealt with. And then the Popperian theory still works almost the same way, but you're dealing with a criticism instead of an actual empirical basic statement. Okay. Obviously, that's going to be less, it's going to be fuzzier. It's going to be harder to know if you've got the problem right. When you're dealing with empirical theories, you've got a lot of advantages because ultimately you can ask the real world if this is correct or not.

[01:36:45]  Bruce: Right. Where with non-empirical theories, you can't. But Popper points out they still have problems. You can still try to resolve the problems. They may have problems that have relevance to real life, and you may be able to resolve them that way. Popper's whole theory was a reaction to inductivism, which was a bad theory. But Popper's theory wouldn't have existed if they hadn't been dealing with a real problem that needed to be solved. So induction existed to solve a problem. The problem was real. It was just the wrong solution. And its existence as a bad theory was what led to the creation of the correct theory. I don't think we would have Popper's theory, but for the existence of induction as a theory. And so I guess the bottom line is it's okay that people disagree with me. It's okay if they misunderstand me. I've already had at least one person write to me after reading my paper.

[01:37:36]  Bruce: It was a PhD. And he said, you know what? I think I've been misreading Popper this whole time.

[01:37:42]  Cameo: Seriously? Oh, interesting.

[01:37:46]  Bruce: As long as that clarifies something for him, right? Where he understands something a little better, that's the point. And you know what? Maybe I didn't phrase this all well. Maybe David Miller is completely correct in the way I'm currently phrasing things. I'm on to something that's a real problem. I may be even on to sort of the right solution, but I'm just doing it in my own, you know, halting way that I can. I understand a certain amount. I can get to the solution to a certain level of abstraction. And then maybe I'm just wrong, anything outside that level of abstraction. And someone smarter than me is going to have to come along, formalize this and really say, Bruce was right in this domain. Here is what you need to say for the rest of the domains. And here's now how you say it. And this is really the right answer. And that's fine. You know, if that's all I do is get somebody to that next step, who's smarter than me, understands this better.

[01:38:43]  Bruce: That'd be fine too, right? On the other hand, when it really comes right down to it, most people aren't David Miller. Most people you talk to on Twitter or Facebook about Popper, they don't have David Miller's understanding of Popper's epistemology. They've got a number of misunderstandings that they're carrying around as of today. I believe that everything I just said will clarify a number of those misunderstandings. Maybe it will create new misunderstandings, but hopefully smaller ones in a different domain, right? That I partially solved the problem. I'm admitting it's a linguistic problem. It's not a fundamental problem. It's ultimately a linguistic problem.

[01:39:21]  Unidentified speaker: Right.

[01:39:22]  Cameo: Well, and to your point about David Miller, he worked directly with Popper. Incidentally, when he wrote his important book on croquet, was right about the same time he had just started working directly with Popper.

[01:39:41]  Bruce: Interesting.

[01:39:41]  Cameo: I don't know if those are connected. Actually, it was about… looking at Wikipedia.

[01:39:48]  Bruce: That's amazing. All right.

[01:39:51]  Cameo: And on that note, I think we're done.

[01:39:55]  Bruce: All right. So we will revisit some of these topics. I feel like a lot of these topics deserve some more in-depth discussion. This was a little bit fast. This is the fastest I think I could go. But we won't revisit this immediately. But I know we have an upcoming podcast. We're going to talk about the topic of, was Popper open-minded or was he kind of non-Popperian himself, which there's a great deal of rumor that Popper was a very bad example of his own epistemology. And I've always been curious about that. And you actually get mixed opinions. People who will say, yeah, he was really bad, and people who knew him, and people who knew him will say, no, actually, that just wasn't true. And I've always wondered about that. And that example that I use, I'll bring up that example again, the example of the philosopher that asked me about, you know, does the Church–Turing–Deutsch thesis imply that the brain really is a computer?

[01:40:59]  Bruce: You know, I think that there's a big open question around what Popper intended on a lot of things and how he reacted. A lot of times when a person's right, they come across closed-minded. You know, it's sometimes hard to tell the difference between those two. And I've often wondered, did Popper come across closed-minded because he was closed-minded? Or did he come across because he was often right? Or is it some combination of the two?

[01:41:26]  Cameo: Well, and after seeing how much misunderstanding there was about the language that he was using, how much of that is just a misunderstanding of the way he was communicating.

[01:41:37]  Unidentified speaker: Yeah.

[01:41:38]  Bruce: And again, if he really was English as a second language, you know, I think there's a good case to be made that people were misunderstanding him, and then he got upset because they kept misunderstanding him.

[01:41:50]  Unidentified speaker: Right.

[01:41:51]  Bruce: So anyhow, we'll do a podcast about that. I think that will be a fun subject, and we'll tie that in with these episodes we just talked about, because I do believe there is likely to be a connection that a great deal of Popper's anger at other people came down to the fact that his theory was a little hard to get exactly to understand all the way, and that there were numerous slight misunderstandings that were just close enough that you could talk, but the misunderstanding would take place. And that's really what I was after here is, is there a way we can reword this for non-Popperians where we can get our point across easier? And that was really what I was after.

[01:42:28]  Cameo: Well, you won in my perception.

[01:42:32]  Bruce: Thank you.

[01:42:33]  Unidentified speaker: All right.

[01:42:34]  Cameo: Thanks so much, Bruce.

[01:42:36]  Bruce: All right. Thanks, Cameo.

[01:42:37]  Unidentified speaker: See you next time. Bye-bye.

[01:42:42]  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.

[01:43:28]  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. Thank you.