Episode 78: Are Animal Memes Knowledge In the Genes?
- Links to this episode: Spotify / Apple Podcasts
- This transcript was generated with AI using PodcastTranscriptor.
- Unofficial AI-generated transcripts. These may contain mistakes. Please check against the actual podcast.
- Speakers are denoted as color names.
Transcript
[00:00:07] Blue: Welcome to Theory of Anything podcast. Hey, Peter. Hello, Bruce. How you doing today? Good. Good. Good. Hey, last time we talked about the Constructor Theory of Knowledge again, and I talked about some interesting examples or counter examples that, in my mind, have fit the Constructor Theory of Knowledge or at least have some sort of very close relationship to the Constructor Theory of Knowledge, but that are generally understood to be not considered knowledge because they don’t come from the two sources, biological evolution or human minds. Anytime I say the two sources hypothesis, I mean the part of the Constructor Theory of Knowledge that there’s only two sources of knowledge, biological evolution and human minds. And in many ways, that’s the part of the Constructor Theory of Knowledge that I’m challenging. I actually don’t have any concerns with the rest of the Constructor Theory of Knowledge. Last time we talked about it, I mentioned animal memes as a counter example, but they didn’t go into it because I felt like it deserved its own podcast. So why don’t we talk about animal memes today and why I have, I think that’s a particularly interesting example of something that, in my mind, fits the Constructor Theory of Knowledge but is outside the two sources hypothesis.
[00:01:26] Red: So just before we’re clear on what an animal meme is, I mean, in the two sources hypothesis, you kind of have knowledge created by nature in the form of genes and knowledge created by human brains in the form of memes.
[00:01:41] Blue: That’s correct.
[00:01:42] Red: Make some sense. I mean, it’s not like it’s completely weird, I guess. But what you’re saying is that animals can actually pass memes.
[00:01:54] Blue: Yeah. Well, Deutsch admits that. In the beginning of infinity, he explicitly states that animals can pass memes.
[00:02:00] Red: Okay.
[00:02:00] Blue: Okay. So he doesn’t consider it to be knowledge though.
[00:02:05] Red: I see. Okay.
[00:02:07] Blue: So we’re going to talk about that today because that actually, the animal memes example raises some really interesting problems for the two sources hypothesis that Deutsch at length addresses in beginning of infinity. And I kind of want to raise the issues and talk about his attempts to solve the problems that come from it.
[00:02:27] Red: Okay.
[00:02:28] Blue: So in the last episode, we covered several important counter examples of the two sources hypothesis. In particular, we talked about astrology. So there’s counter examples going two ways. You’ve got things like astrology, which the two sources hypothesis would declare to be knowledge, but most people would consider to be not knowledge. Then you’ve got things like trade secrets that most people would consider to be knowledge, but the two sources hypothesis are sorry that that the constructor theory of knowledge because it’s really about replicators would maybe not consider to be knowledge. And then we talked about like the immune system in particular, which like completely fits like in every way is analogous to how knowledge is created by biological evolution, right down to being knowledge in the genes, but just in this case, hyper mutated within the life of a single animal and not passed down through sex cells. And then we also talked about animal learning where animals literally in their lifetime create adapted information that’s not in their genes and they can even creatively come up with entirely novel goals and solutions to problems. We gave the example that were never before faced in their ancestral history. We gave the example of an orangutan stealing a boat to cross a river to be able to scare away staff members so that it could have fun washing clothing. Okay. So the rest of this episode, we’re going to concentrate on animal memes. Now, up to this point, every podcast in the past we’ve done, I’ve been assuming that the adapted information that we’re talking about that was not created by the two sources was in fact, not knowledge at all.
[00:04:11] Blue: And there’s a good reason why I did this, because that is the most common way people who address this, including David Deutsch, the way they address his own theory, they’ll usually declare something to not what these examples to not be knowledge. Okay. So the walking robot algorithm is typically declared to not be knowledge. The immune system is declared to not create knowledge. Actually, in past podcasts quoted Deutsch saying it was not knowledge. So that I’ve stuck with that reading of Deutsch’s theory because it’s the most common one. And as we’ll see, I think it’s the best one. It’s the one that’s… Well, let me explain further. Then you’ll see why there is an alternative to it, but it’s in almost every conceivable sense, worse. So defenders of the theory will consistently tell me the walking robot algorithm is not knowledge. Deutsch has himself said the immune system does not create knowledge in his interview with Eli Teer. But occasionally, particularly when I raise animal memes, there’s another potential counter… As a potential counter example, I get a different argument. And the argument goes like this. Animal memes, sometimes even animal learning, is knowledge. It’s just that the genes created that knowledge, not the animal that created it. Now humorously, I usually get this argument after the person has spent considerable time first arguing that the animal memes are not knowledge. And then partway through the argument, they’ll suddenly realize, oh, this argument might work better if I claim animal memes are knowledge, but I just claim that it’s knowledge created by the genes. And this is usually justified on the grounds that the genes created the learning algorithm that created the meme. So the genes deserve credit for having created this knowledge.
[00:05:57] Blue: Now, I’ve rarely seen this argument used on, say, the walking robot example, but there’s no reason why you couldn’t use it on the walking robot example. So for example, you might say, well, actually that algorithm, the genetic programming algorithm, creates the algorithm that causes the robot to walk. And that algorithm is itself knowledge. But it wasn’t the genetic programming algorithm that created that knowledge. It was actually the human that wrote the genetic programming algorithm that created that knowledge.
[00:06:28] Red: And this is when you hit your buzzer that says ad hoc save?
[00:06:32] Blue: Yes.
[00:06:33] Red: Right by your desk, I can imagine. Okay.
[00:06:38] Blue: So animal memes are particularly problematic example. And I think that’s why people will, by the way, I call this argument the credit assignment argument. So to try to save the current understanding of the constructor theory of knowledge, including the two sources hypothesis, from these counter examples, you either have to declare these examples not knowledge or you can declare them knowledge, but declare that they came from the two, the two sources, biological evolution or human minds, by assigning credit to one of those two. Okay. So I call that the credit, credit assignment argument. So anytime I say the credit assignment argument, that’s what I’m referring to. Now, why is it that this particularly comes up with animal memes? I think it’s because animal memes are a particularly difficult example to try to deal with. And typically when I’m trying to discuss it with people, they almost immediately see that memes, by definition, are knowledge outside the genes. Like topologically, that’s what a meme is. So to try to declare an animal meme not knowledge is blatantly bizarre thing to try to do. Okay. Now, Deutsch knows this. And in beginning of infinity, he states that all memes are knowledge. So this is on page 375, like genes, all means, all memes contain knowledge, often inexplicit, of how to cause their own replication. In fact, that is ontologically the definition of a meme. A meme is something that keeps itself instantiated, right? Causes itself to be replicated, but that isn’t part of the genome. So a meme is an idea, a behavior that contains knowledge, how to replicate itself. And it often, but not always does this by being useful in some way.
[00:08:35] Blue: Now, one of the things that I should really mention here is that as they’ve started to study memes, they’ve realized this whole idea of the selfish gene has a version of the selfish meme. Memes are often really bad for the organism, just like genes can be bad for the organism, but that they exploit the organism to get themselves replicated. So to try to define memes solely in terms of usefulness to the animal is a mistake. Although that is often the case, that the reason why a meme exists is because it was useful to the animal, it was adaptive for the organism. But particularly with humans, that’s just not always true. Like astrology isn’t really adaptive to humans. It doesn’t give you some sort of useful piece of information about stars, right? It’s, it replicates itself by exploiting human psychology. So Deutsch continues now. He says that some memes can replicate themselves with great fidelity for many generations is a token of how much knowledge they contain. So Deutsch initially does claim that all memes are knowledge or at least contain knowledge. Okay. Then he explicitly states that memes are stored outside the genome. So here he said, continue on page 375. However, the logic of cop of the copying mechanism for a meme is very different for genes and memes. The situation faced by memes is utterly different from genes. Each meme has to be expressed as behavior every time it is replicated for, for it is that behavior and only that behavior, given the environment created by the other memes that affect the replication.
[00:10:11] Blue: And then on page three 76, he says the upshot of this is that memes necessarily become embodied in two different physical forms alternatively as memories in the brain and as behavior. Okay. By the way, notice I want to emphasize memes are memories in the brain. Like they are not encoded in the genes. Like by definition, I’m sorry to overemphasize this, but this is like really important to the point I’m trying to make. They are absolutely not adapted information in the genome. By definition, if it is adapted information in the genome, then by definition, it is not a meme. Okay. But this leads to a direct contradiction to the two sources hypothesis because he’s forced to admit that animals have memes just like humans do. So on page 405, he says if complex behavior is impossible to imitate, without prior knowledge of the theory causing the behavior, how is it that apes famously can ape? They have memes. That’s me quoting David Deutch. Okay. So Deutch is and then he goes on to talk about how animal memes work using Richard Bern’s theory. Okay. But understand that there is no doubt that animals have memes. Okay. This is a well -known thing and Deutch is not denying it. Okay. So up to this point, it seems like he’s arguing that animal memes, which by definition are knowledge outside the genome, are examples of knowledge outside the genome. Okay. So I think I can understand, given this argument, why some people might jump to the credit assignment argument, they somehow need to save the two sources hypothesis and they can’t really claim animal memes aren’t knowledge like they could to the walking robot algorithm.
[00:11:56] Blue: Because like by definition, animal memes are knowledge like that is what a meme is. And it’s a perfect example of the constructor theory of knowledge. Okay. So I think this is where the credit assignment argument comes from. However, Deutch never raises it himself. And I think for good reason. And I’m going to show why because it is really a pretty awful argument. It is a dangerously awful argument. Okay. It’s much better to just try to declare them to not be knowledge. So he goes on and he seems to argue that animal memes are not actually knowledge after all. So first he tells us that they are not explanatory knowledge. So on page 407, Deutch says, eight memes may seem to depend on explanation on understanding how and why each action within the complex behavior. But recent discoveries from Richard Byrne have revealed how apes are able to imitate such behaviors without ever creating any explanatory knowledge. And then at this point, it introduces the idea of behavior parsing, which is the theory of Richard Byrne of how apes do this. Now, I’ve argued in the past on my when I actually read Richard Byrne that I feel David Deutch has misunderstood a very important aspect of Richard Byrne’s theory. If you mean that animals have no explanatory knowledge in the sense of human level explanations, then surely that’s true. But as we just talked about in the previous podcast, Byrne does see them as having kind of a deep shrewd kind of understanding. And understanding is a kind of explanation. So he does see animals of having a sort of limited explanatory knowledge. Whether you want to call it explanatory knowledge or not, I don’t really care, right?
[00:13:41] Blue: I’m perfectly comfortable with saying no, animals do not have explanatory knowledge. And in this case, I would just simply understand explanatory knowledge to mean human level, explanatory knowledge. Okay. But I want to really kind of note there is a bit of a word shell game going on in that last quote. We went from discussing knowledge, possibly implicit knowledge, he explicitly said, implicit at one point, to a discussion of explanatory knowledge as if they are one and the same, and they really aren’t. Okay, explanatory knowledge is explicit knowledge. Implicit knowledge is non -explanatory knowledge. So when we’re talking about animal memes, they could be implicit knowledge. They don’t have to be explanatory knowledge. And in fact, I would assume that in the vast majority of cases, since most animals don’t have insight, but most animals have memes, that we are talking about implicit knowledge, not explanatory knowledge. But I do want to point out that that which isn’t entirely off base here either. Why I do think there’s a little bit of a misunderstanding of Burns theory. Burn is saying that each individual movement is learned as implicit knowledge, not as explanatory or explicit knowledge. And that is really important to Burns theory. Like he sees if he sees anything like explanatory knowledge, it’s at the animal’s ability to string the different movements together to figure out if I do this series of movements, I can solve a unique goal creatively. That’s where he kind of places an animal’s explicit knowledge if it’s an ape, if it’s an animal that has insight. But there is no doubt that animals have nothing like human level explanatory knowledge or understanding. And so I don’t deny the point which is trying to make here either.
[00:15:30] Blue: So I apologize that I’m being a little bit vague. But it’s like I’m agreeing with the way to a certain level, but I’m disagreeing with him at a different level. And I’m trying very hard to tease out the difference. Because of this limit, according to do it, that animals have according to do it. Animal means consist of learning simple correlations. That’s his term, not mine. And this is part of Burns theory. So I’m not denying that he’s correct here. First, they learn individual movements. Initially, Deutsch claims that all these movements are inborn. Here’s the actual quote on page 407. An ape parses a continuous stream of behavior that it witnesses into individual elements. That term individual elements is going to come up again. So I want you to pay attention to it. That it witnesses into individual elements, each of which is already known genetically. It already knows genetically how to imitate. So this part of the sentence, he basically says that the individual elements, the behaviors that it does that make up the overall behavior, that they are genetically known. Now, this isn’t actually correct. Okay, think about the example of the orangutan that rocks the boat to then gets the water out and then goes across the river to be able to try to scare away the staff to wash clothes. Clearly, it doesn’t have an inborn behavior to rock a boat to get water out. Okay, so if Deutsch had stopped here, he would have been wrong. However, he goes on to clarify.
[00:17:01] Blue: And on page 407, crossing over to 408, he says the following right after the sentence I just quoted, the individual elements can be inborn behaviors such as biting or behaviors learned by trial and error, such as grasping a nettle without being stung or previously learned memes. So while he initially says they’re all genetically inborn, he then quickly corrects himself and he admits, well, actually, that’s only one place where the individual elements movements could come from. They might also come from it just learning by trial and error learning, or it might come from being a meme. Okay, and in the case of the orangutan that crosses the river to wash clothes, it did at some point see a human rock a boat to get water out. And it learned to do that by watching that human do it. So it’s mimetic. It’s clearly not knowledge within the animal’s genes that you can rock a boat to get water out. It learned that mimetically and Deutsch isn’t denying that. Okay, the reason why I’m being very careful here is when you read over what he says and you read it quickly, it really does at first seem like he’s saying all the individual elements are genetic. And because he actually does say that at one point and then corrects it, it leaves you with this impression that it’s all genetic. But really, if you read it carefully, he backs off that very quickly. And he admits that a lot of these maneuvers can come as just being a meme. So checking Burns research, we’re able to teach entirely new movements to apes. And I think this is more this is an episode 40. I talk about this. So apes in the wild, rarely actually learn movements.
[00:18:48] Blue: The vast majority of hand gestures that apes learn are inborn. And I think that’s what Dwight is really talking about is that in Burns research, nearly all their gestures are just inborn and part of their genes. But we know we can teach them new ones, but we haven’t really seen apes in general do that in the wild. But in captivity, we can actually teach the obvious example here is sign language, like the different signs that apes do, those aren’t gestures that are inborn genetically to the ape. We were able to actually teach these new gestures to these apes, and they were able to learn to communicate using these gestures that aren’t part of its genetic inheritance. Okay. So some movements are not genetically stored in DNA. And even Dwight is admitting this. So what apes do is they learn to string these movements together in complex ways that are that are nothing like how just an automaton would work. And again, the example here is the orangutan that rocks the boat to get the water out, crosses the river, and then washes clothes. That is not a string of events that it could have learned through imitation. But the individual movements, it did learn through imitation. So Deutsch claims, because animal memes are simple statistical patterns, rather than explanations, quote, it takes them years to learn a repertoire of memes that involve combinations of actions. Now, fact check. I don’t actually think that this example of the orangutan crossing the river, that it saw hundreds of examples of rocking a boat to get water out. I think when I, when I actually checked Bern’s research, he does talk about how they have hundreds of examples, and that they can learn from these hundreds of examples.
[00:20:39] Blue: But he never actually makes the claim they need hundreds of examples. And I think Bern leaves that quite open, how many examples they need to be able to learn. Deutsch seems to have taken that to mean they require hundreds of examples, but I can’t see anywhere where Bern ever makes this claim. So I think it went slightly beyond what Bern actually originally said. Having said that, I guess I don’t know, right? Like, that’s actually an open question. Clearly, Deutsch is right that animals learn memes much more slowly than humans do precisely because animals learn them through statistical correlations instead of through explanations. So Deutsch’s point is like valid, but they may be learned faster than he thinks they do. Now, on page 408, he says, as for connecting these elements together, the elements of the individual movements in the right way without knowing why, the necessary information, I emphasize the word information here because we’re talking about adapted information, can be attained merely by watching the behavior many times and looking out for simple statistical patterns. Now, recall that Deutsch defines knowledge as information that causes itself to remain so. That is exactly what an animal mean is. And even the most careful reading of Deutsch in beginning of infinity, he is calling it information and he is showing it is information that causes itself to remain so. And I think this is why I find animal means such an interesting example is Deutsch is working very hard to show that it’s not knowledge and that all knowledge comes from the two sources, but he’s basically admitting yes, it’s adapted information and yes, it causes itself to remain so. So how does one look out for simple statistical patterns?
[00:22:29] Blue: Deutsch is saying, let me read it again, as for connecting these elements together in the right way without knowing why the necessary information can be obtained merely by watching the behavior many times and looking out for simple statistical patterns. What does that mean? What is looking up, looking out for simple statistical patterns? What’s the term for that? It’s induction. Now, Deutsch doesn’t use the word induction here, but that’s what he’s talking about. OK, he’s saying animals through induction learn their means. This is something we’ll probably have to come back to at some other point. OK, Deutsch is admitting here that these animals can generalize from observations via statistical patterns. And he’s insisting that animals do this without any sort of explanatory ability, which, by the way, at least if you’re talking about individual movements, I believe he’s correct about that. OK, in other words, the information for for how to do this is not in the DNA and which is admitting this. He’s even admitting it is information adapted for a useful purpose. And the example that that Bern uses is how they actually use they eat by taking food, you know, seeds with nettles and they’ll go through a series of patterns to clean the nettles off so they could eat the food. This is where the whole theory of behavior parsing comes in. OK, so he’s admitting and these animals cannot survive without learning these memes. These animals literally can’t survive in the wild without these memes because they have to learn it from each other to be able to eat.
[00:24:04] Blue: And the current theory is that animal is that apes, the theory from Burm, Burm’s theory, I’m trying to say, is that these apes evolved insight so that they could outcompete other animals that didn’t have this level of intelligence to be able to pass these memes along so that they could learn to eat these types of seeds that allow them to live but allow, say, monkeys can’t or intelligent enough to be able to pass these memes along. OK, and this is. Burns theory of animal parsing. OK, so there’s no doubt we’re talking about a meme that exists because it’s useful. It’s outside the DNA. It’s not in the genome and it exists solely because it’s useful. And in fact, these animals can’t survive in the wild without these memes.
[00:24:49] Red: It seems to me that it kind of fits well with Deutsch’s ideas to say that animals might learn these memes through induction, whereas humans create memes through explanations.
[00:25:06] Blue: Actually, I think that that is getting closer to the real truth, right? That what we have here are two different kinds of knowledge. One is heretical knowledge and one is explanatory knowledge.
[00:25:18] Red: And do you think Deutsch would disagree with that?
[00:25:20] Blue: Well, that’s not what he actually says, right? Let me finish reading what he says. OK, then you can see why the problem exists.
[00:25:29] Red: OK,
[00:25:29] Blue: so he seems to argue that even though the memes are information outside the genome, he seems to argue that they do not count as knowledge because they are just simple statistical patterns. OK, and I think this is where I’m struggling, right? If we were to say it is knowledge, but it’s created by something closer to machine learning, OK, something that we would call induction. And there’s actually a question whether induction actually exists as its own kind of epistemology. I don’t think it does. I kind of agree with, although I’ve got concerns with Campbell’s overall theory, I think Campbell and Popper were on the right path where they were trying to subsume induction into their epistemology. OK, but what we would call induction. I do think that that is correct, that these animals are learning through simple statistical patterns. I do need to point out what an incredibly difficult problem that is to solve, though, like we can’t do anything like this with our machine learning algorithms today. Certainly not with the level of efficiency that animals do it, even if it’s hundreds of examples, that’s still very, very, very efficient compared to our existing algorithms. OK, so I think what’s going on here is that he wants to say they don’t count as knowledge because they’re just simple statistical patterns. Thus he concludes, and this is a quote, that is how eight memes can replicate without the impossible step of literally copying knowledge from another ape. This is on page 409. Now, I feel this last quote eliminates the credit assignment interpretation, and I don’t think Deutsch ever intended the credit assignment interpretation.
[00:27:13] Blue: But if you disagree with me, I have an argument for you that I think will show why I don’t think that the credit assignment interpretation is ever going to work. Before I go on to explain that, though, I do want to note the word game here. Knowledge explicitly means explanatory knowledge in this last statement. Statistical knowledge about patterns apparently does not count as knowledge. There’s a bit of a contradiction worked in here, and I’m trying my best to tease it out. But let me actually quote this again. Memes are knowledge, contain knowledge that keep themselves instantiated by definition. But he doesn’t want to count animal memes as knowledge, not because they aren’t adapted information that keeps himself instantiated, because clearly they are. But instead, he wants to declare them as not knowledge because they’re just simple statistical patterns. Well, I guess I see no particular reason. Like, is this a new criteria that he means to add to the constructor theory of knowledge that to be counted as knowledge, it has to be explanatory knowledge? Well, if so, then you’ve just eliminated biological evolution as a source of knowledge. So it doesn’t make sense that he’s trying to now say it’s literally that they do not copy knowledge. Let me read the quote again. This is how memes can be replicated without the impossible step of literally copying knowledge from another ape. But what are they copying? They’re copying statistical patterns from each other. And the reason why they’re doing this is because it’s useful and it’s adapted information that exists outside their genes. I guess my question would be, is knowledge adapted information that exists outside the genes or not? Okay, is that a necessary but not sufficient set of criteria?
[00:29:08] Blue: Is that what Deutsch is trying to say? He never explicitly states it. Like he kind of just leaves the implication here is that it has to be explanatory knowledge to count. But if that’s true, we got the problem with biological evolution again. So I feel like there’s a little bit of a shell game going on here with words, where statistical knowledge about patterns does not count as knowledge, but really only in the case of animal memes. If you’re talking about biological evolution, then they don’t do explanatory knowledge either. All of that is just statistical implicit knowledge. So it’s okay in that case, but not in the case of animal memes. And this is where I really kind of have a problem with with Dwight’s presentation. I need him to kind of work this out for me. What is it he’s really trying to say knowledge is, right? Is it adapted information that keeps itself instantiated? If so, animal memes should count. Does it have to be explanatory? If so, biological evolution shouldn’t count. And yet he’s clearly here making some sort of distinction between knowledge and adapted information, which is why I think he intended to say animal memes are adapted information, but they’re not knowledge. Now, I need to kind of compare this now to Campbell’s theory, which I haven’t talked about very much in a little while. But Campbell and Popper developed a theory of evolutionary epistemology, which I previously called the universal Darwinism. That may not be the best term for it, but that is a term for it. And I cover this in past podcasts. It’s more or less identical to Dwight’s theory of knowledge, but without the two sources hypothesis tacked on.
[00:30:45] Blue: Basically, Campbell claims, and Popper seems to strongly endorse this, that all inductive achievement, and that is the term they use, inductive achievement and knowledge creation comes from variation selection algorithms. Now, I don’t know if that’s a correct theory. I actually gave some counter examples to it. So I’ve challenged whether that’s even a correct theory, but Campbell’s basically saying that there’s a unique ubiquitous variation selection, blind variation selective retention algorithms in nature. It’s not just the two sources. And he says that nature consists of this giant hierarchy of variation and selection algorithms. Now, given that, and let’s be honest, that that’s true, right? There is even if we don’t want to call it knowledge, animals learn things in their lifetime, and it creates adapted information that is stored in their nervous system, not inside their genes. And animals pass this information in some places, it’s quite complicated information to each other via means. Again, none of this is in doubt. And it does this through variation selection algorithms that aren’t biological evolution or human ideas. This is where I have a problem with the credit assignment argument. So let’s say that the walking robot algorithm is knowledge, but it is adapted information with because it’s adapted information with causal power to causes itself to remain so. But since the algorithm was created, the algorithm was created by that created this knowledge was itself created by humans. We can now claim the humans created this knowledge. And actually, it’s at least partially true due to Campbell’s hierarchy, because that knowledge that is created by the genetic programming algorithm, the genetic programming algorithm was created by a human idea.
[00:32:35] Blue: Therefore, if we’re allowed to assign credit, you’re always going to be able to trace back to the two sources. It makes the two sources hypothesis completely irrefutable. In fact, you don’t even need the two sources hypothesis. You only need the one source hypothesis because literally all human ideas are traceable back to knowledge created by Darwinian evolution that created human minds to begin with. So we could using the credit assignment argument, we could literally say that all knowledge is created by Darwinian evolution. And we can always prove it to be true by just simply showing that whatever did create the knowledge was created by something that was created by something that was created by Darwinian evolution. Okay, so this argument is 100 % impossible to test. You cannot refute it. You can’t even criticize it other than by pointing it out that it is a bad explanation due to this fact. It is absolutely guaranteed we will always have some source to assign credit to that’s in the two sources. At a minimum, all existing knowledge started somewhere in the hierarchy with biological evolution. So you’re guaranteed to be able to traverse that tree of back to biological evolution at some point. And if you’re comfortable with the credit assignment argument, it’s guaranteed to always be correct. So let me use an analogy that is I’m choosing intentionally to make you as uncomfortable as possible. It’s like claiming that all wealth is labor. Okay, rather than wealth being knowledge, which is what I think it really is. When you try to argue that wealth is actually knowledge, the person just points to the fact that all knowledge requires some sort of effort, which is labor. And it’s impossible to refute argument, right?
[00:34:19] Blue: It’s intentionally been set up so that it’s guaranteed to always be true and therefore tells you nothing useful or about the world at all. Now, I’ve often pointed out to fans of Dwight’s theory of knowledge that the credit assignment argument is a bad explanation because you get to arbitrarily assign credit to wherever you want in Campbell’s hierarchy. And if the theory was wrong, you’d still be able to do this. And we want to avoid arguments that always work even if wrong. And if we want, we can now claim biological evolution is the source source of soul source of knowledge using the same argument to say that I’ve that the defenders of the theory but unfazed by my argument would be an understatement. Typically, they will point out how obvious it is that humans are creating knowledge. So it’d be silly to assign credit to biological evolution. Of course, I didn’t really mean we should assign credit to biological evolution. I was trying to show that there was a problem with their argument. And but I get this answer a lot. It’s just obvious seems to be the underlying counter argument. Now, let’s take this idea that all an animal’s knowledge is in its genes. This follows directly from the two sources hypothesis. If there is only two sources of knowledge that obviously animals can’t create knowledge during their lifetime is pretty much a tautology. It’s also why the Twitter crit rats think animals are automatons. Because if there are only two options, creative beings, humans and robots that don’t create knowledge, everything else, that obviously animals have to be in the later category. But let’s take animal memes as our example. Let’s take an example.
[00:35:58] Blue: And this came up in our past podcast of monkeys stealing tourists hats so they can get fed or apes learning to eat nettles through complex non -automatic actions or the example that we that we used recently of birds learning to create a mating song based on their creativity. OK. The Twitter crit rats have argued to me while the genes created the learning algorithm that they used to learn with or while the genes have give gave the monkey knowledge to seek food. It’s an instinct or well, the genes gave the birds knowledge of beauty. But an animal mean itself is clearly not stored in the genes because by definition, a meme is not stored in the genes. The monkeys learn to steal from tourists for food by watching other monkeys do it and then get fed and then they figure out they can do it, too. It will not pass to the next generation via DNA, of course, because the adapted information behind this behavior is stored in the monkey’s brains and nervous systems, not in its DNA. Further, this behavior clearly fits all three of the required criteria for the constructive theory of knowledge. It’s capable of enabling its own preservation. It’s adapted information that is capable of enabling its own preservation. The monkeys behavior persists by passing for monkey to monkey outside DNA because it’s useful. And number two, it can be copied from one embodiment to another without changing its properties. The monkeys copy each other and they are able to do the same behaviors but adapted to themselves. It can enable transformations and retain the ability to cause them again. The behavior is a transformation. OK, so this fits the constructive theory of knowledge.
[00:37:42] Blue: Now, here’s the questions I would have for those that are trying to use the answer is in the nervous system, not in the DNA. What created this adapted information? Even the Twitter quit raps will admit that the animal has something like a learning algorithm that is the proximate cause of causing this adapted information into existence. They just simply are trying to assign credit to biological evolution instead. So I’m really unclear why they insist on saying animal memes are knowledge in the animal’s genes. I don’t even see how this argument qualifies as anything but intellectual dishonesty. Plus, aren’t memes, by definition, knowledge not stored in the genes? Isn’t that what makes them memes in the first place? Now, recall that Deutsch admits that animals have memes and thus adapted information outside the genome. Let’s pretend that Deutsch says the following. Animal memes are all examples of knowledge in the genes. What could he possibly mean by that if he were to say that? Deutsch has never said this, by the way. I’m just imagining this so that I can show how ridiculous the credit assignment argument is. He might mean animal memes are not knowledge, which is what I just from the book of Big Game Infinity I showed. I think that’s exactly what he’s trying to say. It’s not explanatory knowledge, and therefore it’s not knowledge. Let’s imagine instead he actually intended the credit assignment argument. Then this is what he must mean. He must mean when he says animal memes are examples of knowledge in the genes. He must mean evolution gave animals the ability to learn from others, and this knowledge created by this algorithm is not stored in DNA, nor does it pass on to the next generation via reproduction.
[00:39:30] Blue: But I’m going to still call this knowledge in the genes, because I feel this biological evolution deserves the credit for how that knowledge got into their brains, because the learning algorithm that was created biological evolution, and that is stored in their genes, plus the genes gave the animal a desire to eat, and that is why they had to learn to steal hats for food. That is what I actually mean by all the animal’s knowledge is in its genes. Well, of course, this is absurd at this point. I mean, this is why I say it borders on just outright intellectual dishonesty. To say animal memes are examples of knowledge in the genes, it’s just an outright contradiction. It does not make sense. No one would really, really reasonably say that. This is really why I wanted to tease out animal memes as an example, because I wanted to put down the credit assignment argument once and for all, that animal memes make this impossible, that you cannot just go around assigning credit to something else in the hierarchy, unless you’re prepared to really claim that animal memes are an example of knowledge in the genes, and then admit that what you really mean is that you’re assigning credit to the genes. But really, it was created by the animal’s ability to learn. It’s not stored in the genes. It’s just to the point of just being meaningless at this point.
[00:40:56] Red: Can I ask you a question about this? Yeah, hopefully this will be somewhat related. But these crit rats, it sounds funny to say out loud, say that animals are just automatons, which, I don’t know. I mean, you’ve made the case it’s really complicated, which I agree with, I guess. But even based on their assertion, they’ll kind of, or they’ll say that animals don’t create, all the knowledge comes from the genes. Okay, yeah, that’s debatable, as you’ve said. But then they’ll say, well, that means the animals don’t have feelings. Now, when I hear that, I say, well, my dog, when I come home from work and she’s excited and licking me and happy, I mean, I think people interpret that as meaning animals are not happy. Where’s this connection between creating knowledge and having feelings? I’m not really quite sure.
[00:42:07] Blue: That’s a very good question. You know, I need to probably do a podcast on that specifically. Luckily, Dennis Hackathol has actually explained that on his blog.
[00:42:18] Red: Okay.
[00:42:19] Blue: And we probably need to actually read through it together and then talk through, frankly, the critical rations mistakes being made all throughout the blog post.
[00:42:28] Red: Okay.
[00:42:28] Blue: I actually admire the fact that Dennis made a very sincere attempt, though, to explain his reasoning. I will try to do it justice for memory as best I can here. But it makes more sense to actually go over his actual argument. It really boils down to what I said earlier in this episode, that if there’s only two sources of knowledge, then by definition, animals must not be creating knowledge. All the knowledge has to be stored in the genes. So it must be that animals are just basically robots, just like a Boston Dynamics robot. Okay. I can program a Boston’s Dynamics robot to act like it’s happy when you come home. The fact that that Boston Dynamics robot acts like it’s happy when you come home doesn’t mean it feels anything, right? So their argument would be you can’t use the acts outside behaviors of the animal to judge if it has feelings or not. And we know that all the animals’ knowledge is created in the genes, animals’ knowledge is in its genes. That’s their argument due to the constructor theory of knowledge. Therefore, we know they must just be mindless robots and they can’t possibly have feelings. I think that the argument goes no further than that, if I’m being honest, is it’s this kind of inductive leap that’s taking place. I know that Boston Dynamics robots can act in ways like they have feelings. So that doesn’t mean anything to me, that animals act like they have feelings. And I know, due to the constructor theory of knowledge or how they currently interpret it, that there’s only two sources of knowledge, biological evolution and human minds.
[00:44:18] Blue: And therefore, it’s impossible that an animal’s anything except a clever set of non -feeling algorithms that create behavior. And I think that is the whole argument in a nutshell.
[00:44:31] Red: I mean, it’s interesting. I mean, I don’t think that animals are really like self -reflecting in the same way humans are not by a long shot. But I mean, to say they don’t have feelings, it just seems like there’s a jump that I’m just not getting. There’s a jump there.
[00:44:49] Blue: So you really have to go back to what is critical rationalism, right? You’re putting up a conjecture and then you’re trying to refute the competing conjecture. Or you’re trying to create the conjecture in such a way that it tells you something testable about the world, right? So this is something that I don’t think the Twitter crit rats know. But it used to be, not even just a few decades ago, that scientists overwhelmingly believed animals had no feelings. And that was almost the settled science on the subject. I’m getting this from one of Franz de Waal’s books, which I don’t have handy, so I can’t quote him. But it’s one that I listened to recently. And there was this group of scientists interested in animal behavior who just didn’t really believe that. So they came up with testable theories of animal feelings and emotions. And they went out and they performed tests. And these tests got corroborated over time. In some cases, not all animals probably do have feelings. Some don’t seem to. Insects, they don’t really think have feelings. And I would have to then cover each of the experiments they came up with, how they came up with them. The one that’s probably the most impressive is animal grief. And I think I’ve mentioned this in a past podcast, that it’s difficult to explain grief unless you explain it as a feeling that has adaptive behavior because it causes you to pay attention to your children and to take care of them, because you’ve got fear of grief. By itself, it’s unadaptive, right? To be able to explain how the feeling of grief came to evolve requires you to first hypothesize that there’s a feeling of grief.
[00:46:50] Blue: Now, in the case of humans, we’ve all felt it, so we know it’s a real thing, okay? But we don’t know how else to explain grief because grief by itself is a non -adaptive behavior. You stop taking care of yourself, you might die, which is not adaptive to reproduction. So one of the things that you could then ask is, could we go test for animal grief? And that’s exactly what they did. And of course, famously, there are certain animals that do stop taking care of themselves when they’re in grief, dogs being one of the most famous examples, but like great apes, elephants, and they have lots of documented examples of animals literally dying from grief, okay? Like grief for a dog is incredibly dangerous, and you have to actually give them human antidepressants to get them out of their grief so that they won’t, they’ll start eating again and they won’t die, okay? Now, why is that? You have to somehow explain this grief -like behavior as not being a feeling, and you have to come up with some testable alternative to it. And like, I don’t know how you would do that. I mean, the theory that the reason why a dog dies when the dog, its playmate dies or its owner dies because it won’t eat, the most obvious hard to vary explanation is that they feel grief, okay? And then you can actually go further than that. You can test for that by saying, well, we know that humans feel grief is related to various hormones. So we can test to see if the animals have the same hormonal changes.
[00:48:40] Blue: They actually did an experiment where they tracked down this troop of apes in the wild, and they would wait to see which ones were experiencing what seemed to be grief, and then they would check their feces for hormone levels, and they also tracked what the apes did to try to survive the grief that they were feeling. And what they found is the hormones change exactly like a human does, the same hormones, because they’re physiologically so similar to us, and they even solve their problem of grief the same way we do. The apes that were like, let’s say it was the death, it was a mother, it was the death of the child or something like that, okay? Or it could be the child or the death of the mother, that happened too, okay? Or it even could just be that they were close friends. They don’t even have to be genetically related. What they found is that an ape going through grief will actually seek out additional grooming from its friends. So basically to survive the death of a loved one, an ape will do exactly what a human does. They will go seek out friends that they have and try to interact with them more and be around them and experience touch more to resolve their feelings of grief that is so painful for them, okay? Now does this prove beyond doubt that animals experience grief? No, of course not, but you can’t ever expect proof. That’s justificationism. What we have here though is a group of very clever scientists that figured out if animals feel grief, that should have testable consequences.
[00:50:17] Blue: And they figured out what those testable consequences should be and they figured out how to try to refute those consequences and they were unable to refute it. Now let’s compare this to the the twit rat version of animals don’t feel things because of this analogy with Boston Dynamic Robots. There’s not an ounce of testability there, okay? If they really want to get serious about their theory as an actual scientific theory, they need to actually come up with, okay, if animals aren’t feeling things, then what does that say about the world? What are the testable consequences of the world? And then show the critical experiments that show that the animals behave in a way different than if they had feelings. And by doing this, we actually can test whether animals have feelings or not. We can actually determine which of the two theories is the better theory, okay? And these scientists, like Franz DeWall, they went out and they performed these experiments over and over again until finally the scientific field changed its mind about whether animals feel things or not because they just, there is no theory of animal behavior that doesn’t include feelings at this point, like that I’m aware of. Okay, there may be one, I’m not an expert. So you know what, I’m not even discounting the possibility that animals don’t feel things, right? I’m simply trying to do a critical rational assessment of the current state of the discussion. And that is, it looks to me like the reason why the idea that animals have feelings is winning the day scientifically is because the group that cares about that managed to come up with testable theories and then test them.
[00:51:57] Blue: And the other group either has lost interest or maybe they’re out there and I just don’t know about them, they’re just not as well known yet and eventually it’ll come forward. I don’t know, right? I mean, you never draw conclusions definitively. The reason why it’s winning the day though is because of testability. They’ve made their theories testable and they’ve tested them. I don’t know what else to say. This is just how science is done, right? It’s of course it’s a conjecture. Of course we don’t know for sure. But right at the moment, if you want to explain animal behavior, you need to reference emotions and feelings. And because you have to do that right now, the vast majority of animal researchers believe animals have feelings. I don’t think it’s 100 percent. I think there are some, there still are a lot of scientists who aren’t animal researchers who believe animals don’t have feelings. Friends DeWall and Byrne both call this out, right? That if you’re talking about scientists not in the study of animals that it’s actually still quite popular to believe animals don’t have feelings or that animals have much lower intelligence than experiments now show they have, things like that, right? Byrne in particular points out that he feels like he’s stuck between a rock and a hard place because there’s this whole group of sociologists that don’t study animals that are convinced that animals have no feelings, that they have no special insight or understanding like Byrne believes they do. And he’s afraid of offending that whole group because they’re so convinced of it, even though they’re not in studying animals at all. They’re studying humans. Is it sociologists?
[00:53:39] Red: Yeah, I think he said
[00:53:40] Blue: it was sociologists that there’s a whole field where it’s very psychologists, sociologists, where it’s very popular that study humans, that it’s very popular for them to claim that animals have lack X and that it’s, they don’t have insight, they don’t have understanding, they don’t have feelings, and that’s what makes us human. And they’ve kind of latched on to some of these things. And he knows that that Byrne and his fellow animal researchers are going to offend them, right? And then he turns around, he says, but on the other hand, if you talk to layman, layman just assume that whatever’s going on inside an animal’s head is exactly analogous to what humans do and that animals have to understand things and they have feelings exactly like us and there’s no difference at all. And he says, the problem is, is that I don’t think that’s true either. I’m afraid of offending layman. So basically he feels like you got the group of animal researchers that are starting to settle upon this idea. Animals have feelings, animals have learning algorithms, animals, some animals, a very rare group have a sense of self. He thinks most animals have no sense of self based on the mirror test, right? But I know Deutsche’s criticized that. I’ve then shown that Deutsche’s criticisms don’t fully take into consideration what the mirror test was supposed to be. But the idea that your cat has no sense of self is going to offend the average layman. But then on the other hand, the idea that an ape does have a sense of self because it’s an animal with insight and can pass the mirror test, that’s going to offend the sociologist.
[00:55:16] Blue: So the animal researchers feel a little caught between a rock and a hard place, but they’ve slowly made progress and they’ve really been winning the day because they’ve made all their theories testable. And it turns out that the, and you really stop and think about it. Of course, the reason why feelings exist in humans is because it had some sort of real life consequence that had survival value, right? I mean, like, there’s no, there should be no doubt about that. I mean, we can imagine philosophical zombies where it’s a person just like you and me. They act like they have feelings, but in fact, they feel nothing. But I think more realistically, you know, if philosophical zombies could physically exist, I think that we would be them, right?
[00:56:01] Red: But
[00:56:01] Blue: we’re not. So there’s some reason why you need feelings, you need qualia, and it has survival value. Now, whatever that reason is, even if we don’t understand it, it would apply to animals too. So it’s not really unthinkable that animals evolved it all first and we get it through our animal ancestry. The other possibility is, and this seems to be Deutsche’s theory, that it all evolved all at once, that animals have no feelings, and then you have this kind of leap to universality that takes place. And suddenly all at once, animals get universal knowledge, get knowledge creation, animals as humans, get knowledge creation, get feelings, and that all these things kind of came as one giant leap. Okay. Now, if they’re really serious about that as a theory, that’s great. Like I’m not against people going out. And if they could actually come up with consequences for that and go test it and show, look, if animals have, under the theory, the animals have feelings, this would be true. And under this other theory, this other consequence should exist instead. Then that would be pretty impressive, right? That would actually bring the animals don’t have feelings theory back into the fore and cause it to become a true competitor again. And I think that’s what Dennis thought he was doing, right? He’s trying to say, well, if animals did create knowledge and have feelings, then they wouldn’t do these automatic movements. But nobody believes that. Like that’s literally a straw man argument of what Bern’s theories are, right? I mean, like, nowhere does Bern say, in my theory, animals should never have automatic movements. And never mind the fact that humans have automatic movements, too, right? And that’s true. And humans clearly do
[00:57:47] Blue: have feelings, right? So I think the issue is that you really have to get serious about making this other theory that animals don’t have feelings into a true testable scientific theory. And the people who are defending it today, it’s just a philosophical theory to them. It’s interesting. They’re not really serious about trying to turn it into a true scientific theory. And there’s just no particular reason why animals couldn’t have feelings, even though they don’t have explanatory knowledge. I see no explanation being offered for why those two have to go together. And therefore, I think tentatively we go with what the current best theory is, which is the animals have feelings.
[00:58:32] Red: Well, I have no problem believing they have vastly different feelings than us. But I mean, when I leave in the morning and my dog is, like, sulky and doesn’t want to say goodbye to me, because she’s mad that I’m leaving and then I come home in the afternoon and she’s licking me and happy to see me. I can’t think of a simpler explanation than she’s actually having an emotional experience. The alternative is that she’s some kind of master manipulator that’s just trying to. I don’t know.
[00:59:08] Blue: It’s interesting on the podcast when we had David Doichon for that special session. He actually said, you know, I can’t I don’t have the exact quote, but it was something in the effect of I’m not sure dogs could show so much feeling if they didn’t have feelings.
[00:59:28] Red: So he almost he said, well, I don’t really know that for sure.
[00:59:31] Blue: Right. And he kind of backed off after that. But I kind of get the feeling that that David Doichon accepts that dogs have feelings.
[00:59:39] Red: Well, yeah, it seems like maybe the assertion there is that that out of domesticated animals in general, we may have somehow given.
[00:59:49] Blue: Yeah, we we created so we created the argument might go something like this. We created evolutionary pressures for dogs to evolve feelings.
[00:59:57] Red: Yeah. And presumably maybe other domesticated animals to
[01:00:01] Blue: maybe other domesticated animals to with dogs being the one that has been the longest domesticated by humans.
[01:00:07] Unknown: So
[01:00:07] Blue: is
[01:00:08] Unknown: that
[01:00:08] Blue: right?
[01:00:08] Unknown: OK.
[01:00:09] Blue: Yeah, by far, like dogs. In fact, as an interesting theory, dogs are obviously domesticated wolves, right?
[01:00:19] Red: Yeah.
[01:00:20] Blue: And domesticated animals have certain things that show up about them. They tend to keep behavior similar to when they were children. So like a wolf is kind of naturally playful and curious, like a dog, right? And as it grows up into a wolf, it loses those traits as it becomes an adult. And that’s really common amongst various animals that as a child, it’s kind of curious and playful. And they usually don’t have as much hair, you know? And so dogs into adulthood keep a lot of the traits of baby wolf.
[01:00:59] Red: So
[01:01:00] Blue: one of the theories is, is that domestication is the process of breeding the animal to keep certain baby traits into adulthood. OK. And so basically, dogs are perpetual baby wolves.
[01:01:20] Red: That’s interesting. I’d never thought about that.
[01:01:23] Blue: That’s a really common theory. They don’t just mean that for dogs. They claim that all domesticated animals have similar traits. Well, here’s where things get interesting. Humans have similar traits. We are, in essence, perpetually baby apes, right? With the low amount of hair and things like that. And we have a lot of the same traits. One of the traits of a domesticated animal is that the frontal lobe grows larger and humans have these giant frontal lobes. So one of the theories that’s out there that Temple Grandin mentions in one of her books is that humans and dogs domesticated each other. And so humans are, in fact, a domesticated animal. And it was dogs that domesticated us while we were domesticating the dogs. And the theory goes something like this, that once humans figured out that they could have wolves in their kind of family unit, that the wolves allowed these humans that would be, at this point, maybe not humans, they’re prehumans, right? That it allowed these prehuman apes to have more time to be able to concentrate on other things. And that, in essence, that was the beginning of a sort of domestication process. Now, I don’t know how much I buy this. For one thing, to the best of my knowledge, I think that the relationship between dogs and humans only dates back like 10 or 15,000 years, which is a lot compared to any other domesticated animal. But I’m pretty sure humans prior to dogs, that there’s good signs that human paintings exist 30,000 years ago or something. So I don’t know if this theory would survive testing or not, but it is kind of an interesting theory, this idea that maybe humans and dogs domesticated each other.
[01:03:14] Red: Yeah, it’s interesting. I know when I got my dog and it was just the whole cliche completely played out where I’m the dad that doesn’t want the dog and then you get this dog and it just changed my life. It felt like a new hole in my heart just filling in. I didn’t know it was there and it really does feel like something really special in nature, this connection. I’d never experienced that.
[01:03:47] Unknown: But
[01:03:48] Blue: like you say,
[01:03:49] Red: our species has been around for like what, 100,000 years or something? It’s relatively recent. 200,000 years. 200,000.
[01:03:58] Blue: You sometimes hear two million years. I think it sort of depends on when.
[01:04:03] Red: How you define humans.
[01:04:06] Blue: But I don’t think it’s more than two million years, the hominids at least. So actually, I’m going to Google that now. So dogs, it says, has been around for 11,000 years. 300,000 years, it says. So probably hominids is like 200,000 or sorry, 2 million years old. And humans is two or 300,000 years, which by the way, really isn’t very long. But have you ever seen the movie The Cave of Forgotten Dreams?
[01:04:39] Red: I know it’s been a while, but yes, the Werner Herzog, right?
[01:04:42] Unknown: Yes.
[01:04:43] Blue: That is such a fantastic movie. Like it was life changing to watch it. And I own a 3D TV and I watched it in 3D. You really have to see it in 3D because the caveman artists use the 3D surface of the cave to create their art. And you don’t get the full effect without the 3D version of it. So they filmed it in 3D. And you look at what those humans were doing 30,000 years ago, which would be before the existence of dogs. And they look like humans to me, right? It’s hard to see them as a separate species because they’re so obviously like us at this point. So I don’t know. You mean judging by the artwork they were creating? Yeah, judging by the artwork. They seem very human to me,
[01:05:32] Red: right?
[01:05:33] Blue: So I think some of it’s quite beautiful. I mean, obviously they’re not as good as artists as modern artists. Like we’ve gained a lot of knowledge. But their level of artwork was so much more sophisticated than I thought existed 30,000 years ago. Orders of magnitude more phenomenally beautiful than anything I had thought they could possibly do at that stage. And the fact that they painted these pictures in the cave over the course of thousands of years, right? That some of these paintings in the cave are thousands of years older than some of the other paintings in the cave. And one of the artists intentionally left his hand prints to leave his mark behind. And he was like missing a finger or something. So we know it’s the same guy. Each time he leaves the hand print. And it’s weird to think about that the amount of time that art had been put into that cave by different humans is longer than than since the existence of the Roman Empire, you know, longer than the system, the existence of the the earliest civilizations, the earliest beginning of history, where history is defined as actually being able to write stuff down, right? They were doing cave artwork in there for whatever reason they were doing it for a longer period of time than that. And it’s kind of just stunning to think about it.
[01:06:59] Red: Yeah, well, I’ll have to have to rewatch that one now.
[01:07:02] Blue: Okay, let me just do a quick summary. And then we can wrap it up.
[01:07:05] Red: Okay.
[01:07:06] Blue: Animal memes are an example example of knowledge that exists outside the genome. And they’re an example of animals creating knowledge that exists outside the genome. I do agree, though, with Deutsch that it’s primarily rooted in statistical correlations. I don’t see why that isn’t a kind of knowledge, though. I don’t see why he’s trying to say it shouldn’t count as knowledge just because it’s not explanatory knowledge. I will admit that animals don’t have explanatory knowledge in the human sense. I think they might have a limited version of that based on Burns theories. That kind of makes sense, though, that maybe animals like apes got introduced to a sort of pre -explanation. And that’s why they have greater insight than other animals. And then the leap between us and animals is to a full universal kind of explanation. Now, if I knew what that meant, I would probably be able to create AGI. So obviously, I’m intentionally being a bit vague here because we just don’t understand more than to be able to explain it at a very vague level. But I think animal memes in particular, because they do pose a problem, it has led to what I call the credit assignment argument. And I hope I’ve gotten across why I feel the credit assignment argument is so bad, right, that it really needs to be done away with it to claim all an animal’s knowledge is in its genes. When what you really mean is that I’m assigning credit to what the animal learned or what its memes are because the biological evolution created the learning algorithm. And while at the same time admitting, this is adaptive information that exists outside the genome. It never existed in the genome at all in any form.
[01:08:59] Blue: What actually existed in the genome was a learning algorithm. Um, to try to call an animal meme an animal’s not that it’s a form of knowledge in the genes. It really does border on the absurd, right? And so misleading that it’s just kind of silly. Because of that, I, I hope I can, I’ve explained why I feel like it really is something that needs to be contended with that the constructor theory of knowledge needs to say something more meaningful about animal memes than what is currently being said in the beginning of infinity. I don’t think we can dismiss it as not being knowledge merely because it happens to be about statistical correlations. And then there’s the fact that it isn’t just about statistical correlations. According to Bern, there is the level above that you learn the individual movements through statistical correlations, but that you then the animal then has to creatively string those movements together, the individual elements as Deutsch calls them. So I think this is something that the constructor theory of knowledge needs to come with, come to grips with better. And we needed just a description and an understanding, a theory of knowledge that does a better job of explaining what animal memes are and how they relate to other kinds of knowledge.
[01:10:20] Red: Well, thank you, Bruce. And looking forward to editing this and re -listening and continuing on our journey next week.
[01:10:28] Blue: All right, sounds good.
[01:10:30] Red: Thank you very much.
[01:10:35] Blue: 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. 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 four strands.org. There is a donation button there that uses PayPal. Thank you.
Links to this episode: Spotify / Apple Podcasts
Generated with AI using PodcastTranscriptor. Unofficial AI-generated transcripts. These may contain mistakes; please verify against the actual podcast.