Episode 86: Fuzzy Categories, Essentialism, and Epistemology (Hofstadter Part 2)

• 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: Hello out there. This time on the theory of anything podcast, picking up where we left off last week, Bruce gets deeper into Douglas Hofstadter’s ideas on language and the mind and his assertion that analogy -making lies at the heart of intelligence. Bruce considers how Hofstadter’s theories may be interwoven with ideas on language and cognition promoted by Steven Pinker in how the mind works along with, as always, the epistemology of Carl Popper and David Deutsch. We again consider if Hofstadter’s ideas are based on induction and how should critical rationalists view ideas that seem at least superficially to be inductive. I hope you enjoy this conversation as much as I did.

[00:00:56]  Red: Welcome back to the theory of anything podcast. Hey Peter. Hello, Bruce. How are you doing? Good. We are going to continue the discussion about Hofstadter’s theories. I’m intermixing a bunch of things this time, though. I’m going to include quite a bit from Steven Pinker’s How the Mind Works, which I thought covered very similar subjects. There’s some really interesting things that they bring up that make sense to me, and I think they match real life really well, but I’m curious what other people think about these as far as telling us something about human intelligence.

[00:01:34]  Blue: Okay. Do you think first, before we delve into that, we could maybe just for my own benefit, and I imagine some portion of the audience really summarize what we talked about last week, what are Hofstadter’s contributions to human knowledge or ideas on consciousness, which I don’t believe. I think we kind of established or not really that much of an influence on David Deutsch, except in that his description maybe of the Deutsch -Turing thesis. Yeah, Deutsch -Turing church thesis. Church -Turing Deutsch thesis. I knew that didn’t sound right. Church -Turing Deutsch thesis stating that reality can be simulated or the argument for universality, but past that, truthfully, I’m still a little fuzzy on what you’re asserting about his ideas last week.

[00:02:40]  Red: Yeah. First of all, I rarely assert much of anything. Fair enough. What I do is I explore interesting ideas. In terms of last week, I did my best to explain Hofstadter’s views. I did point out that at a minimum, human beings actually do act like this. If I’m going to assert anything, it would be that, that he is talking about something that has some level of verisimilitude. Because of that, I took the argument that he can be ignored because it’s inductive. I basically flipped that on its head. That is something I am asserting also. I am asserting you do not get to claim a theory that tells us something about reality can be ignored because it’s inductive. On the contrary, if it’s actually telling us something about reality, either you need to consider your theory refuted, or because it’s an observation now that goes against your theory, or you need to explain the observation in terms of your theory, which would be what I’ve been calling subsumption. You need to show that the inductive elements of Hofstadter’s theory can actually be explained in terms of Popper’s epistemology. I think that was, if anything, that was the main thing I was asserting. It was not really a strong agreement or disagreement with Hofstadter’s overall views. It was more an assertion that you are thinking of it wrong if you are trying to dismiss it on the grounds that it’s inductive, that you have misunderstood the nature of Popper’s relationship to induction at that point. Or even just Popper’s epistemology in general, the fact that if it’s a real -life observation, then it really has to be considered a refutation.

[00:04:31]  Blue: And just to summarize two 500 -plus page books in one sentence, he’s saying that analogies and self -reflective thought are core to human consciousness.

[00:04:50]  Red: Yes, I got the quote.

[00:04:51]  Blue: It’s a special.

[00:04:53]  Red: Yeah, I got the quote. Let me actually read the actual quote. I’ll repeat that for people who didn’t listen to the last episode. I actually want to introduce this subject today with some thoughts on feedback from our six -part series on Deutsch’s constructor theory of knowledge. I actually feel like all of these things are related. That was why I did the stuff on knowledge before I dug into the epistemology and then before I dug into what Hofstadter and, as we’re going to see Steven Pinker say about how humans think in terms of words and concepts. And the studies that they’ve done to back it up. It’s interesting what they’ve actually come up with in terms of studies. We’re going to talk about that today. Even just the idea, can you dismiss something on the grounds that it’s inductive or does that disprove your epistemology if human beings actually do induce things? Even that we’re going to see is a type of exactly the problem that Hofstadter and Pinker are talking about. Because the word inductive itself is just a fuzzy category. It doesn’t have any necessary and sufficient conditions that define what it is. So we’re going to talk about that today. So even just the very fact that people will try to dismiss Hofstadter’s theories as no that’s inductive is itself an example of how Hofstadter is right.

[00:06:19]  Blue: That

[00:06:20]  Red: human beings think in terms of these fuzzy categories. And really maybe even sometimes just outrightly struggle to think precisely into precise categories. Clearly we can do it. And we’re going to talk about that too. That human beings can think in either precise categories or fuzzy categories. But the two kind of mingle themselves in funny ways that are kind of surprising. The main feedback that I’ve received from that six part series. First of all, we were joking about why would anybody even listen to that six part series. And I still don’t know the answer to that question. It was Bruce getting stuff off his chest and talking through a bunch of stuff. And in fact, we’re actually going to see that that six part series was completely unneeded because this one episode’s going to give a much, much better answer than everything I did across that whole six part series.

[00:07:09]  Blue: Okay.

[00:07:10]  Red: So thereby making that whole six part series completely irrelevant, basically. All right. So the main feedback I receive, not surprisingly, is something along these lines. You know, try as I might. It just doesn’t seem to me that adapted information created by the immune system or by a genetic programming algorithm, such as the one that makes the robot walk, should really count as knowledge. It doesn’t feel very knowledge like to me. I can wrap my mind around the two sources hypothesis. This idea that biological evolution in human minds create knowledge. Now, right there, that is what I think of when I’m thinking of knowledge. That’s what I mean by knowledge. And then sometimes they’ll throw in something like this. We’ve talked about in the six part series, all sorts of different caveats that get thrown in, all of which turn out to be problematic, but probably the most common one is something like this. I can see that the immune system and the genetic algorithm create adapted information mechanically through perspiration. And if that’s all it’s doing, then it doesn’t really feel very knowledge like to me. Okay. So in this episode, I went through that six part series to try to address it in a certain way, knowing that I’m probably going to be unsuccessful. And we’re going to talk about why that six part series is basically going to be unsuccessful. There’s really a correct answer to the question that’s being raised and to the objections that are being raised. But you have to be familiar with Hofstadter’s theories or Stephen Pinker’s theories or they’re not really their theories. There’s like a whole field of study in this area where they’ve done numerous experiments about it.

[00:08:49]  Red: And once you’re familiar with it, there’s a much more obvious answer to this whole question. Okay. In the previous episode, we went over mostly kind of the first chapter of Hofstadter’s book, Services and Essences. And I explained briefly Hofstadter’s theory that human creativity is ultimately about analogy making. So here’s the actual quote you were thinking of, I think. Once all of this has been laid out so explicitly, it seems a trivial thing to assert that analogy making lies at the heart of pattern perception and extrapolation. What could be more obvious? And when this banality is put together with my earlier claim that pattern making is the core of intelligence, the implication is clear. Analogy making lies at the heart of intelligence. We then talked about in the previous episode, but isn’t this an inductive theory? And, yes, it turns out that Hofstadter himself calls it an inductive theory. Okay. And then I argued, but so what? Sure, maybe as a paparian, if you’re a critical rationalist, you have this allergic reaction to the word induction. Okay. But the very fact that you have an allergic reaction to the word induction and try to get away from anything that’s inductive feeling is in and of itself a misunderstanding of Popper’s critical rationalism. Okay. First of all, this is something that human beings actually do. We talked about the story of the man who met a woman that he was attracted to that was a journalist for his university. And then when he gets another opportunity to talk to a woman writer who’s a writer for his university, he is immediately interested in meeting her because he just can’t help himself.

[00:10:29]  Red: It’s so analogous to the previous circumstance that he immediately just sort of, even though he knows it doesn’t make that much sense, he wants to meet this woman because, hey, last time it was, it worked out and I was attracted to this woman. So he can’t help but feel like maybe this is going to be the same way. Okay. It’s not even a set of repetitions. It only happened once. It wasn’t like this has happened a whole bunch of times. And Hofstadter made the claim that these analogies, these fuzzy categories, whatever we want to call them, they kind of force themselves on us. Right. There’s so much a part of the way we think and the way we act that we just can’t even, we can’t help but feel certain ways. And maybe a better example of this was the translation example where the guy who was disappointed with a translation in a European language. So he just can’t bring himself to allow his book to be translated into Korean, even though he’s got no explanation for why he should think that the translators in Korea are going to be bad, just like the ones that was in some European language that he was at least somewhat familiar with. And yet it’s such a human thing for us to fill this way. Right. We really do. These analogies, these inductions, if you will, they really do, they’re a huge part of the way we think and they kind of just force themselves on us.

[00:11:47]  Red: We also talked about Hans Rawlings example where scientists for forever thought you should lay children on their stomachs because they had this category in their head that unconscious people like wounded soldiers will choke on their vomit and so a baby should too. Turns out that’s completely wrong. But it was really hard for science to change their mind about that because the analogy was so strong. Sorry, go ahead.

[00:12:12]  Blue: Well, I was just going to say I think if as the two sources hypothesis sort of feels right, and I know we’re not just talking about what feels right on this show, but it does kind of seem like it feels has the ring of truth, I guess, that being away from induction as humans might not be something that would be very possible. Maybe not desirable either. I don’t know. It’s just how we make sense of the world, I think.

[00:12:44]  Red: And there’s several things here. Like the fact that people latch on to the two sources hypothesis so strongly when it’s in fact an inductive theory, right? And they’re critical rationalists latching on to this theory that’s exactly equivalent to what they call induction.

[00:12:59]  Blue: I think we just lost about five listeners there.

[00:13:04]  Red: So yeah, probably did. That would be five of our 10 listeners. There’s this interesting relationship between all these things, right? The thing I was trying to get at though in the previous episode, I don’t know if I got to it as well as I should have, is that you don’t get to dismiss the theory about something that’s correct about the world on the grounds that it’s inductive. The burden explanations on you to fit this observation to your epistemology without referencing induction. So induction is itself a fuzzy category without a precise set of necessary and sufficient criteria. So it wouldn’t be that surprising that Hofstetter would call it inductive, even if ultimately we find that the creativity algorithm is not inductive. It’s actually based on conjecture and refutation and variation and selection, okay? Presumably that’s exactly the truth, right? And which is why you don’t get to dismiss it just because Hofstetter calls it inductive or even if you feel it’s inductive. Because induction, when thought of that way is just a fuzzy category. It doesn’t have a precise definition, okay? And ultimately when we do get to whatever the precise thing is, that’s going to be the creativity algorithm, if Popper’s epistemology is correct, it won’t be based on inductive logic. It will be based on conjecture and refutation or variation and selection, okay? And if that’s the case, then you were wrong to claim that we should ignore the theory because it was inductive because that was just a word that Hofstetter was using. It had nothing to do with how it actually worked, okay? It was his way of throwing it into some rough, fuzzy category that people are familiar with, okay?

[00:14:52]  Red: Or the theory’s wrong, in which case it’s not based on induction and so that was never a basis for refuting it. And I don’t know any other way around this, right? Coming up and saying, I can ignore it because it’s inductive, you’re automatically wrong the moment you say that. There’s no scenario where that turns out to be a good argument. Even Hofstetter’s theory being wrong, that’s still not a good argument. And the reason why it’s so confusing is because induction consists of this fuzzy concept that Popper’s epistemology kind of inhabits the same area, right? Only in a much more precise way, in a much more straightforward, testable way, okay? And so, and this is why Popper, actually, I don’t have the quote handy, but Popper will actually say, well, you know, if you want to call my theory induction, I’m fine with that, right? Because it doesn’t have it the same space. This makes sense. Popper’s epistemology was a refutation of theory of induction and replacing and subsuming it with a better theory. So, of course, they have very similar feelings around them at times, okay? And that’s not too surprising. In any case, you cannot dismiss Hofstetter’s theory on the grounds that it’s quote inductive. You would have to first try to make sense of what specifically is wrong with it. And there’s nowhere in the theory that it has the problems that typically Popper would cite as problems with induction. It’s got nothing to do with repetition. Okay, that was one of the main things he used to cite. We’ll have to do a separate podcast on Popper talking about what the problems of induction are, right? It’s got nothing to do with a claim to some sort of justificationism.

[00:16:43]  Red: Okay, when the guy who can’t help but feel like he’s going to like this new woman he hasn’t met yet, when he inductively decides that, so he wants to meet the woman, he knows he’s not certain, he knows it’s not a justified claim. So, the main things that Popper said, this is what’s wrong with induction. They don’t apply to the way this theory is using induction. This is something different now, okay? And it has to now be thought of in its own terms. I don’t know, does that make sense where I’m trying, what I’m trying to say here on why I don’t feel you can dismiss it on the grounds that it’s quote inductive?

[00:17:19]  Blue: I, yeah, I think it makes sense both on the level of just ringing true and it makes sense rationally and I get it. It just seems, I don’t know, I mean, I guess I’m open to the idea that many or most critical rationalists just have this deeper understanding of epistemology than I do and they’re able to just look out at the world and just reject a large part of human ideas as being inductive and that helps them in some way, but I don’t know. I guess that’ll be as detector, it’s just a little trigger there, right? It just doesn’t seem quite right for the reasons you’ve stated.

[00:18:06]  Red: Yeah, okay. All right. Now, having said that, it’s a totally different question of whether Hofstadter’s theory is actually correct, right? And it also depends on what we mean by correct. I mean, he could be correct that humans do this, but that may not actually turn out to be the core of human intelligence, in which case we’d probably want to say, well, his theory had some verisimilitude, but it wasn’t ultimately correct, right? And so there’s this whole shades of gray and we don’t even know what the answer is. Since we don’t know what the creativity algorithm is, you can’t really at this point dismisses theory, nor can you really say it’s correct. And so I’m approaching it from that standpoint, that it’s an interesting idea worthy of some discussion. And really not too much further than that, okay? There’s certainly some aspects I’m going to advocate and I’m going to say, isn’t it obvious this really is what human beings do? But that doesn’t mean I’m necessarily saying, and therefore that proves that this is the creativity algorithm, right? Like I’m not even suggesting that, really. On the other hand, it could be related. I don’t know, right? And if you’re interested in AGI, since we don’t know what the answer is, you have to kind of be open to lots of different ideas, most of which will probably turn out to be false, you know, or maybe partially true, but ultimately false. And this is really where I’m advocating for the study of some of these ideas, okay, is look, let’s just look at what they say. Maybe they’ll turn out to be a stepping stone for us towards AGI. Maybe they won’t. And let’s just discuss them.

[00:19:41]  Red: Let’s not dismiss them out of hand on grounds like, oh, it’s inductive, you know, which to me is just the wrong way to go about it, right? We got to look at each theory individually, figure out how testable is this? Can it be tested? How far can it be tested? Things like that, right? And let’s just look at them individually and let’s just criticize them individually. I guess I’m with you on that from what I know about Hofstadter. I mean, you know, from reading at least one of his books and looking at many lectures.

[00:20:10]  Blue: You know, it seems like he’s got some ideas out there that are worth exploring, that might be one piece of a puzzle in terms of understanding human consciousness and AGI. Again, as I said before, he seemed to take a lot of pages to express some ideas that could have been expressed much more succinctly in my view, but it is what it is.

[00:20:38]  Red: I feel the same way about a lot of his books, actually. Okay, by the way, I don’t really feel that way about Gödel Ascherbach. I highly, highly, highly recommend Gödel Ascherbach.

[00:20:50]  Blue: Well, you kind of made me want to read that one when you said it was such a good description of the church -turing -deutsch, or I guess this is the church -turing thesis.

[00:20:58]  Unknown: At

[00:20:58]  Blue: the time, church -turing -deutsch.

[00:20:59]  Unknown: At the time, yeah.

[00:21:01]  Blue: Okay.

[00:21:02]  Red: Since the Deutsch version didn’t come until later. Yeah. Okay. So now, what has Popper actually said about this? Now, we’ve kind of hinted at Popper’s views on essentialism and how this is clearly related to essentialism in some way. In fact, the name of the book is surfaces and essences, right? That’s essentialism that we’re talking about. So let’s actually take a look at what Popper actually said about essentialism. Okay. This is from Open Society, Volume 1, Page 31. I use the name methodological essentialism to characterize the view held by Plato and many of his followers that it is the task of pure knowledge or science to discover and to describe the true nature of things, i.e., their hidden reality or essence, and a description of the essence of a thing they call a definition. And then he goes on to say on Page 32, methodological essentialism, i.e., the theory that it is the aim of science to reveal essences and to describe them by means of definitions, can be better understood when contrasted with its opposite methodological nominalism. Instead of aiming at finding out what a thing really is and at defining its true nature, methodological nominalism aims at describing how a thing behaves in various circumstances and especially whether there are any regularities in its behavior. In other words, methodological nominalism sees the aim of science in the description of the things and events of our experience and in our explanation of these events, i.e., their description with the help of universal laws.

[00:22:40]  Red: The methodological nominalist will never think that a question like what is energy or what is movement or what is an atom is an important question for physics, but he will attach importance to a question like how can the energy of the sun be made useful or how does a planet move? That’s also Page 32, I believe. So that was kind of Popper’s famous statement about essentialism versus nominalism. And I feel like that’s a really good starting point for the rest of our discussion. And I’m actually going to show that there’s some nuances here that I don’t think Popper quite addresses. Kind of he does. He’s got some interesting ideas around this. And I’m going to show that some of these can kind of be melded together somewhat, okay, that there’s some interesting parallels between Hofstadter’s view that you might claim as essentialism and Popper’s view that you might claim as non -essentialism, that in some sense they meld together into a single theory. When we introduced the idea of fuzzy categories using Hofstadter’s theory of intelligence, we talked about this idea that they were changing categories constantly being expanded by analogy, as well as sometimes contracting as a term goes out of use, as some term goes out of use. Now, I claim that this is the real reason why the philosophy of essentialism is bad philosophy. This is the part where I think Hofstadter’s theories and Popper, what we just read from Popper, actually have a really heavy overlap that in essence, they’re both saying, there’s me using the word essence, they’re both saying the same thing, that the philosophy of methodological essentialism is wrong, that that is true both of Hofstadter, who explains why it’s wrong, okay.

[00:24:38]  Red: Popper kind of says it’s wrong, and instead you should be doing nominalism. And Hofstadter says, well, there is something to the fact that humans think this way, but the reason why you can’t ever find that essence that essentialists were trying to find is because these fuzzy categories never stay still, and they’re constantly being changed by analogy. And there isn’t a single set of necessary and sufficient conditions or criteria that define the individual category, nor could there ever be, nor would that even be as useful. And this is where Hofstadter definitely is going beyond Popper now, that there’s something useful about the fact that our fuzzy categories have no essences. And I don’t know if Popper would agree with that or not, Popper maybe just wasn’t interested in that subject, but that is where I think we are starting to get now into where Hofstadter’s theory maybe does go beyond anything that Popper actually said. The key point here is the early Greeks were trying very, very hard to work out questions like, what is the category known as beauty? What is the category known as knowledge? What is the category known as creativity? And they were trying to find it, right? They were trying to come up with these definitive set of necessary and sufficient criteria that was the category knowledge, that was the category creativity or beauty, okay?

[00:26:05]  Blue: It seems to me he kind of goes beyond like you have one school of philosophy that says that words don’t really mean anything. These are just approximations that kind of you can go down this path, knowledge isn’t real or whatever. And then there’s another school of philosophy that wants to sort of pin down exactly what words mean and approach things from that direction. And so as I’m thinking about it, he’s kind of saying, well, it’s actually a good thing that everyone’s got their own definition of words and that that’s actually helpful in a way that words mean different things to different people and you can’t, you kind of get at the truth more from the space between words or something. Is that fair?

[00:26:54]  Red: That’s a fair understanding of what Hofstadter is getting at. He’s, we’ll get into some of the specifics of what he’s suggesting here. And we kind of did in the last episode too. But I think he’s really kind of important. You’re always looking at what’s the context of what a person says. You can’t just grab something somebody says outside a context of where they said it. And then it’s going to be some sort of golden truth that’s always true, right? When Popper is talking about essentialism, he’s talking about in terms of what is science, basically. I mean, there’s a what is question right there, right? And he’s talking about physics. He actually uses the term physics, you know, that physics isn’t interested in what is energy. Okay. By the way, I’m not even sure if that’s entirely true. Like if I were to go ask a physicist what is energy, they could almost assuredly give me a pretty good definition of it today. It took a long time for them to get there, a first working out what you could do with energy. But I don’t think it’s actually true that a physicist would be disinterested in the question, what is energy? But Popper is looking at how does science actually use terms? And Hofstadter is looking at how people think and how do we think in terms of words and categories? And he’s pointing out that it’s really fuzzy categories, not precise categories that we for the most part use. Okay. And he’s saying, yes, that’s a useful thing. It’s the fact that we can constantly grab a completely new circumstance and reapply an old one because of that. Okay. So sorry, I’m getting ahead of myself here.

[00:28:26]  Red: Let me actually finish what I was trying to say here. No matter how much you try to define a category like knowledge, there’s always going to be an exception case. Now, if that sounds familiar, it should. That was the point of the six episodes on Deutsche’s theory of knowledge. Okay. Is that there is literally all I did for six episodes was I played that game knowing that was what I was doing. Okay. Is that there is no set of criteria you’re ever going to come up with that is going to define the fuzzy category knowledge that you have in your head. It’s impossible for the exact same reason that that’s true of every fuzzy category. Okay. Let me put this a different way. So long as you see the constructor theory of knowledge as specifically trying to answer the question, what is knowledge? You’re basically doomed. There will never, ever, never, ever be a satisfying answer to that question because what you were really doing is you’re making the essentialist mistake. So it is actually not even slightly surprising that I can spend six episodes and easily show that there is no matter what criteria you come up with, there’s going to be some counter example to your definition of knowledge. Okay. And that’s just a fact. Okay. And I really don’t think there’s any way around that fact. Okay. As long as you’re trying to get to what do I mean by knowledge? What does knowledge feel like? How do I understand it in my head? You’re just making the essentialist mistake. You’re already off track. There’s just no way to do it. Okay.

[00:29:58]  Red: Going back to Popper, page 32, and to those philosophers who tell him, the word nominalist, a methodological nominalist, that before having answered the what is question, he cannot hope to give exact answers to any of the how questions, he will reply, if at all, by pointing out that he much prefers the modest degree of exactness, which he can achieve by his methods, then to the pretentious model, which they have achieved by theirs. And I really strongly agree with what Popper is saying here. If you understand it as a response to methodological essentialism, which was the idea that science is about trying to define terms, try to define what knowledge is, try to define what creativity is. Okay. As long as you’re trying to do that, it really will only ever be a pretentious model. It will never be anything else. Okay. Now, let’s talk though that this is different. If we’re talking about science, that’s one thing. But how do human beings actually think? That’s like, in some sense, a different question. Okay. Let’s actually now take a look at how human beings actually do use fuzzy categories. Now, I want to use an example that I’ve mentioned in past podcasts, but I’m going to give the word for word. This is from Stephen Pinkers, How the Mind Works. So Peter, give me a definition of the word bachelor.

[00:31:27]  Blue: Bachelor, an unmarried man.

[00:31:31]  Red: Yeah. Okay. That is that is the quintessential definition of a bachelor.

[00:31:37]  Blue: Okay.

[00:31:37]  Red: So Stephen Pinker, taking that definition, that very one, he says, now imagine that a friend asks you to invite some bachelors to her party. What would happen if you use that definition that you just came up with to decide which of the following people to invite? So now he gives a list of people. Arthur has been living happily with Alice for the last five years. They have a two -year -old daughter and have never officially been married. Bruce was going to be drafted, so he arranged with his friend Barbara to have a justice of the peace marry them so he would be exempt. They have never lived together. He dates a number of women and plans to have the marriage gnolled as soon as he finds somebody wants to marry. Charlie is 17 years old. He lives at home with his parents and is in high school. David is 17 years old. He left home at 13, started a small business and is now a successful young entrepreneur leading a Playboy’s lifestyle in his penthouse apartment. Eli and Edgar are homosexual lovers that have been living together for many years. No, this was before there was such a thing as homosexual marriage when this book was written. Faisal is allowed by the law of his native tribe to have three wives. He currently has two and is interested in meeting another potential fiance. Brother Gregory is a bishop of the Catholic cathedral at the Grontin -upon -Thames. Knowing who is a bachelor is just common sense, but there’s nothing common about common sense, Stephen Picker says. Somehow it must find its way into a human mind or a robot brain.

[00:33:09]  Red: Common sense is not simply an almanac about life that can be dictated by a teacher or downloaded like an enormous database. No database could list all the facts we tacitly know and no one has ever taught them to us. So when I give that list, obviously he’s picking examples of people that would technically be bachelors or technically not be bachelors, but you would actually know we’re bachelors. And what he’s trying to say is you probably can figure out who to invite to this party, right? Even though you’re not even probably going to even worry about the fact that the word bachelor has this technical definition of an unmarried man, right? You’re going to have to figure it out based on social cues and consequences and the context of the question. And yet you will come up with an answer using the fuzzy category bachelor, which really doesn’t have that much to do with that definition of bachelor, an unmarried man. And it’s actually got a whole bunch of other properties attached to it. That you would have a hard time even specifying what they are. Like if I were to say, well, what are you basing this on? It would be difficult to actually list out how you’re determining whether someone counts as a bachelor or not.

[00:34:33]  Blue: So saying it depends on context is kind of similar to saying it depends on a web of analogies.

[00:34:42]  Red: That’s what Hofsteder would say, yes.

[00:34:46]  Blue: I don’t want to actually endorse his theories too strongly.

[00:34:52]  Red: I endorse them as interesting for discussion, but I don’t endorse them as necessarily correct.

[00:34:57]  Blue: You’re not ready to add them as another strand of the fifth strand of reality. Okay.

[00:35:05]  Red: Also, I would endorse them as having some truth to them. Like I would go that far, right?

[00:35:09]  Blue: Okay.

[00:35:10]  Red: But

[00:35:10]  Blue: yeah,

[00:35:10]  Red: that’s about as far as I would feel comfortable going.

[00:35:13]  Blue: Okay.

[00:35:14]  Red: Now, in services and essences, Hofsteder on page 33, he says, at every moment, we are faced with a new situation. Now, you never really think about that, but every single moment of your life is a new situation that is completely novel. Okay. And you have to figure out what to do with every single completely new novel situation that has happened to you, is happening to you at any given moment. And then he says, actually, the truth is much more complicated than that. The truth is that at every moment, we are simultaneously faced with an indefinite number of overlapping and intermingling situations. It’s not just a single situation. Then again, on page 33, far from being faced with one situation, we are faced with a seething multitude of ill -defined situations. Our poor besieged brain is constantly grappling with this unpredictable chaos, always trying to make sense of what surrounds it and swarms into it willy -nilly. Also on page 33, and what does to make sense of even mean? It means the automatic triggering or unconscious evocation of certain familiar categories, which once retrieved from dormancy help us find some order in this chaos. To a large extent, this means the spontaneous coming to mind of all sorts of words. So that is actually Hofstetter in his own words, why he thinks it’s actually a good thing that we use fuzzy categories for the most part, rather than precise categories. If it was a precise category, you couldn’t keep applying it to every single completely new novel circumstance. It would have some sort of boundary you couldn’t go past. The fact that the fuzzy categories are constantly expanding by analogy allows us to keep saying, oh, this is actually similar to this circumstance.

[00:37:07]  Red: And then suddenly you feel like you know what to do. You might be right. You might be wrong. I would argue this is a conjecture, not an inference, but it gives you an idea of what to do with this circumstance by labeling it in your head as the same as some other circumstance that you’ve faced some time in your past. You can see how this would have survival value, this ability of humans have to be able to do this. So he gives an example of how this happens in the human mind. Obviously we’re making conjectures here. So this is his conjecture, but I feel like this one’s fairly compelling. So I’m going to give it in full here. So consider this is from page 34 of Surfaces and Essences. Consider the word mommy. So imagine a child and mommy initially refers to a specific female caregiver for that child. And that’s it. Mommy with a capital M, okay? And the word has no other meaning except that parent, right? Then one day the child, Tim, is what Hofstadter names him, makes a thought something like this once he gets old enough. That person is taking care of her just like mommy takes care of me. That key moment marks the birth of the concept mommy with a small M. Okay. So at that point, through that human ability to make an analogy, Hofstadter is arguing that suddenly mommy with a capital M becomes the category mommy with a small case M. And suddenly this idea of different people have different mommies and there are many mommies in the world comes into existence where once it used to be a very specific person and that was it. Okay.

[00:39:01]  Red: So Hofstadter then says, you might think the concept mother is very precise. Perhaps as precise as that of prime numbers. We talked about how prime numbers is precise. It has an actual essence. Okay. That would imply that to every question of the form is X a mother or not, there would always be a correct objective black and white answer because that’s true of prime numbers. But let’s consider this for a moment. If a little girl is playing with two dolls, one is bigger and one is smaller. And she says that the big one is the small one’s mother. Is it an example of motherhood? Does the large doll belong to the category mother? This is page 37 of surfaces and essences. So he’s trying to make you think about this now for a second. It’s easy enough to say, well, yeah, mother is a very precise concept. It’s the concept that it’s you have a female parent and they give birth to a child and that is the mother now. Okay. And that may at first come across as precise as prime number. But in fact, that’s just not what the word mother actually means to human beings. Right. You may somewhere in your head, no, that’s one way of one concept that’s covered within that fuzzy category of motherhood, but it’s a much, much broader category than that. So he goes on to give examples of how the concept expands for Tim or child over time. Tim eventually realizes that the category mommy doesn’t even require that the woman gave birth to a baby, say in the case of an adopted mother. So from page 50, mother think of mother versus surrogate mother or adoptive mother or single mother, each

[00:40:42]  Red: modified think of how each of those modifies the concept of mother, such as the concept of mother no longer requires giving birth, raising the child, being married, or even being female. So there is no essence of mother. Okay. Precisely because it’s really just this rich fuzzy category that’s grown through analogies over time based on your experiences, your culture, you know, how people happen to use words, all sorts of different things. Let’s go back to page 38. Instead, motherhood as a category holds together several different properties, such as that of female biological parent, female nurture, female protector, and these properties do not need to be present simultaneously. And this all happens without anyone directly teaching this to Tim. Tim just nobody teaches you this is how you understand the concept of motherhood. You’re able to just figure out how people are using the term or you’re even just expanding it just through its use. At some point, Tim is going to reach a point where that the expanding concept of motherhood becomes increasingly expansive until he can understand that motherhood includes things like Madam Curie is the mother of radioactivity or what mother nature is. Okay. And those will just be seamless. No one even needs to teach you what those mean. Okay. You just the very fact that you keep expanding through analogy what the concept of motherhood is eventually leads to you realizing what those terms mean and how they’re being used. So on page 39, each category, I thought this was interesting, but Hofstadter says each category is synonymous with the term concept, the way he’s using them.

[00:42:37]  Red: Now, you’ve probably heard me say this a few times in past podcasts, where I’ll say we’re not interested in the word, we’re interested in the concepts. And a word typically actually encompasses the fuzzy category that is the word often encompasses many, many, many different concepts. Okay. But then it also has kind of this fuzzy concept that’s kind of like this umbrella concept that attaches itself to, to all of them at the same time. You still understand what motherhood is in terms of a biological female biological parent, but you would also understand that there’s many other concepts of motherhood that are analogies to it. Okay. And then Hofstadter says that to the adult mind, the category mother is a highly developed mental category to which thousands of analogies have already contributed. Page 39, he asks if it’s really possible that this very same mechanism that causes a two year old to call St. Bernard a sheep is the same one a scientist used to discover a new scientific idea. Now, this is kind of an interesting idea. He gives some even better examples later on, which I’m not going to get into into this podcast, but I’ll maybe I’ll do in a separate podcast, but he gives one that I thought was interesting. He says consider it as a scientific example Galileo. At one time, the moon was a capital M and it referred to the moon and there was only one moon. Like there were no other moons. Okay. Galileo goes and he looks through a telescope. He discovers there’s these satellites orbiting around Jupiter and he decides to call them moons, moon with the lower case M. And at that point he had invented a new concept based on an old one.

[00:44:21]  Red: The original one was very precise. It referred to a specific body, the moon. And now the concept of moons, little M, has been invented because he’s decided to call the little satellites around Jupiter moons because he sees them as analogous to our moon and our planet. Okay. So in fact, when a child calls a St. Bernard a sheep, they are doing the very same thing that Galileo is doing when he discovered moons around Jupiter using analogy and that Hofstadter gives several really kind of compelling examples of this from children. One of the things he kind of points out is that we often laugh at children when they say things that sound funny to us, but the child is exposing how humans think. Okay. And how they learn concepts. So here’s some, a child that says, I undressed the banana or when their mother has their eyes closed, turn your eyes on mommy or a child that says dentists patch people’s teeth or when it stops raining, they said they turned off the rain. Okay. Now we kind of think this is really cute when a child says this, but what the child is doing is they’re taking some concept they already have and they’re by analogy, extending it. And then later as they hit adulthood, they’ll realize, okay, that’s not the way people speak and they’ll adjust the way they speak. But you know exactly what they’re talking about in each of these cases when they say undressed the banana, peel the banana, right? They don’t have the concept of peel yet. So they use the concept of undress. And nobody taught the child this.

[00:46:07]  Red: And you’re not confused when the child says it, you know exactly what they mean, even though nobody speaks this way. Okay. And so he’s making a pretty compelling case. This is how human beings think. They use analogies to expand categories and slowly come up with these fuzzy categories over time that are very rich.

[00:46:31]  Blue: Just the way humans think, maybe a little bit more compelling than something like Universal Grammar or the Superior Wharf Hypothesis as we compared it to last week.

[00:46:46]  Red: You know, it’s interesting. This does have a relationship to the Superior Wharf Hypothesis. It’s not the same as it, but clearly there’s some overlap there. Okay. So this one we’re not going to get into in this podcast. Some categories that we use directly map to a word, but some don’t. In the case where it maps to a word, I think that’s where there is some truth to the Superior Wharf Hypothesis, right? That we do think of some concepts in terms of a word. And Hofstadter gives some examples of this, that every moment of your life, you’re constantly applying things into a category, probably in most cases, but not all cases, using words. And you’re thinking cute little girl or something like that, right? You see a little cute little girl and you think cute little girl. And there is some truth to the Superior Wharf Hypothesis precisely because there is a relationship between words and fuzzy categories.

[00:47:41]  Blue: Yeah. And it just occurs to me that if you think about, well, okay, maybe our brains are really less controlled by words per se and more controlled by analogies. That’s kind of a bit like saying our brains are controlled by culture.

[00:47:57]  Red: Yeah. If

[00:47:57]  Blue: you think of culture as an interwoven web of analogies, which kind of rings true for me, I think.

[00:48:05]  Red: Yes.

[00:48:05]  Blue: Yeah. The whole thing seems to make a lot of sense. And

[00:48:08]  Red: you can also see why it kind of does fit Deutch’s theories and Hopper’s theories. It’s not. It certainly isn’t completely at odds with it. And there’s some really heavy overlaps between them. Okay. Because you’re right. We are talking about memes at this point where this is kind of partially a description of how memes work or how memes interact with the ideas in a person’s head or ideas in person’s head interacts with memes. It’s hard to make a separation between those, by the way. Human ideas aren’t exactly the same as memes, but human ideas don’t exist without memes. So there is a very strong relationship between them. Yeah. Okay. Now, we, I think in the last episode, we talked about the idea that multiple uses of a word or concept like play. Okay. So we talked about playing a sport versus playing a board game versus playing a musical instrument and how they have conceptual distances. Okay. And I argued that probably to most people playing the instrument is the longest conceptual distance. Interestingly, that wasn’t true for you. And it does actually matter what you’re currently thinking about and what your current context is depends on how you’re going to tell what the how you’re going to determine what is the strongest conceptual distance. And we use the example of how in other languages, you do not play a musical instrument. There’s like a totally different concept or word for that. And so in French or some other language, they’re going to have no idea why we say we play an instrument because it’s nothing like playing a sport or playing a board game. There are also conceptual distances to related concepts, which is another interesting thing.

[00:49:59]  Red: So Hofstra gives examples of golfers versus golf courses or holes or fairways or teas or irons. Any of those terms immediately kind of evokes the idea of golfer and has some sort of relationship distance between those two concepts. Okay. Or to put it another way, a fuzzy category or concept exists in such a way that the concept has a center and a halo around the center that extends and gets fuzzier as it goes out to edge concepts and eventually related concepts. Okay. So on page 50, Hofstra says, thus we come to the idea of a multidimensional space in which concepts exist somewhere like separate points. However, around each such point, there is a halo that accounts for the vagueness, blurry and vague, blurry and flexible quality of the concept. And this halo becomes ever more tenuous as one moves further out from the core. Words have halos of meaning. So on page 62, he says the halo of a word are rather the blurry boundaries of the concept named by the word gradually engulf what were once metaphorical. This works by analogy slowly expanding which sub concepts are included in the fuzzy category. So on page 62, he says, the most recent fresh novel creative uses of the word still strike us as metaphorical. And yet over time, these usages, if they resonate with native speakers will become so widespread and bland that after a while, no one will hear them as metaphors anymore. Now, maybe that’s hard to believe, but he gives examples of this. We talk about legs of a table, spine of a book, head of a lettuce, marginal ideas, roaring wind, a belly button, the high point of a melody.

[00:51:50]  Red: Some of those may have seemed metaphorical to you, but many of them probably didn’t, right? You probably never really thought of legs of the table as a being a metaphor for human legs. And yet that is how it started, right? It’s just that the usage became so common that now it doesn’t feel like a metaphor anymore. That’s just what the word leg means now as part of its core central part of the of the word, right? The core meaning of the word. And the metaphorical stuff was the stuff that now we can currently think of as being inside the halo of the word instead. Okay. So on page 66, he says, all these concentric layers making up categories in its full glory are the result of a spectrum of analogies of different types collectively made by millions of people over a period ranging from dozens to thousands of years. These analogies form a seamless continuum. They range from the simplest and easiest to make, giving rise to the core concept core. So simple and natural that they are not even seen as analogies by an untrained observer to more interesting and lively ones and finishing up with extremely farfetched and unconvincing analogies. I want to talk now again about why this shows why word essentialism is wrong. And I specifically now I’m saying word essentialism instead of essentialism. And I’m going to explain why. Okay. This is why you can’t apply essentialism to words and their related fuzzy concepts. On page 54, Hofstadter says, until quite recently, philosophers believe that the physical world was divided into natural categories. Under essentialism, there’s supposed to be a precise set of criteria that is necessary and sufficient to define an object as being in a category or not.

[00:53:38]  Red: So for example, what is a bird? If the philosophers have been correct, that there were specific attributes that we could objectively point to and say the category, category bird is defined by having a beak laying eggs is covered with feathers, et cetera. Okay. So on page 55, he says, the set of membership criteria, the defining properties is said to be the intention of the category. While the set of actual entities that meet the criteria, the members is said to be the extension of the category. The notions of intention and extension borrowed from mathematical logic are thought of as being just as precise and rigorous by these philosophers, the essentialist philosophers, precise and rigorous as the discipline, as that mathematical discipline itself. And the use of these terms reveals the ardent desire to render crystal clear the abstract essences of all the highly variegated objects that surround us. So that was page 55. But there literally is no such set of necessary and sufficient criteria for human fuzzy categories. Membership is not determined by any specific set of criteria. And the words slash concepts used to define the category feet, beaks are themselves also quite fuzzy. So even if we could somehow say, a bird must have feet, you know, we would then have to decide what counts as a foot, right? Like as a, as a Dalper’s Flynn, Finn, is that a foot? You know, I mean, they’re themselves fuzzy. Okay. And this doesn’t even get us into things like saying something like, if you say a bird has feathers, does a chicken stop being a bird? If you pluck it, right? And so, I mean, like,

[00:55:20]  Red: any set of criteria you try to come up with to define bird, it’s just isn’t going to work. So even if you could identify a necessary set of sufficient criteria for membership, the fact is, is that by analogy, it’s going to eventually change. The fuzzy category is going to stretch as people start to use the word in interesting metaphorical ways. And as those metaphorical ways become more common in use, they’ll stop being considered metaphors and it will just simply be part of what that word means. So this is now going back to Stephen Pinker from how the mind works, page 126. He says, the members of a fuzzy category lack a single defining feature. They overlap in many features, many features, much like the members of a family or the strands of a rope, none of which runs the entire length. In the comic book comic strip Bloom County, Opus the penguin, temporarily amnesiac, objects when he is told he is a bird. Birds are svelte and aerodynamic, he points out, he is not, birds can fly, he cannot, birds can sing. His performance of yesterday left his listeners gagging. Opus suspects that he is really bullwinkle the moose. That’s the difference between a fuzzy category and why it’s just completely impossible to apply to find a set of essences for each of these categories. And it’s why the methodological essentialists, Plato, Aristotle, it’s why that they were just basically on a doomed path with

[00:56:51]  Blue: their

[00:56:51]  Red: approach to trying to

[00:56:52]  Blue: figure out things in terms of defining words.

[00:56:55]  Red: Now, here’s the thing though, it’s a good thing that human concepts or fuzzy categories are not like math. Now, we’ve already talked some about why Hofstadter argues that, but he tries to prove this by asking a simple question. Is Tweety from Looney Tunes cartoons, is it a bird? Is he a bird? Or actually he asks this, when I imitate Tweety, am I a bird? Now, if the human mind dealt with precise categories instead of fuzzy categories, it’d be impossible to make sense of that question. And yet you know what I mean when I ask that question. And whichever way you want to answer that question, you’re able to interact with it precisely because your mind’s able to stretch the concept of bird to include someone that is imitating Tweety, even without really necessarily believing that person’s actually literally a bird.

[00:57:46]  Blue: Okay.

[00:57:47]  Red: Other examples, he asks people to give their opinions on, and he finds that they often have strong opinions when he asks them. And then a few days later, they change their mind and have an equally strong opinion going the other direction. So some examples he uses are our sandals shoes. Is Big Ben a clock? Is a stereo a piece of furniture? Is a calendar hanging on a wall? A book? Is a wig an article of clothing? These examples all play at the edge of the fuzzy categories, neither fully in or out of those categories. So this is why Platonism is incorrect. Platonism is incorrect. There is no platonic concept of bird or really any of the human concepts that we deal with normally outside of very specific scientific and mathematical ones like prime number that are used within a limited scope. Now I want to note here, though, that prime numbers do have a set of necessary and sufficient criteria to be a member of that set. So prime numbers do have essences. And if we’re talking about prime numbers, the essentialists are actually correct. Okay. But the fuzzy category bird does not have an essence. So I once had a paparion on the Critical Rationals page on Facebook challenged me when I would refer to word essentialism and angrily said, that’s redundant to say it that way. It’s just essentialism. He was wrong. Word essentialism is false, but essentialism is sometimes true, like for prime numbers. And there’s actually a distinction there worth making. It’s really when you try to apply essentialism to words that it becomes a problem. And I’m going to take in the side here just for a second.

[00:59:34]  Red: And I’m going to talk about various debates that you’ve probably seen or participated in, talking to the listeners at home. And absolutely all of them are silly, because they’re actually debates over essentialism of a fuzzy category.

[00:59:50]  Blue: Well, sorry, I was just going to say, so are there examples outside of math of words having essentialist definitions? I mean, I can think of things, I don’t know, coming back to the hypotenuse of a triangle. I guess that’s a mathematical concept. And has an essentialist definition. Is this the difference between something that’s more mathematical and not, would you say?

[01:00:18]  Red: Good question. And we’re about to actually see what Stephen Pinker says about that.

[01:00:22]  Blue: Okay.

[01:00:24]  Red: So here are some debates that you may have participated in, listeners at home, that are actually debates over a fuzzy category. And therefore, they are the essentialist mistake. Is there free will? What is creativity? What is knowledge? What is novelty? Okay, I have seen and have participated in extensive discussions about this. And I always start the discussion with, you’re making the essentialist mistake to the point where people now make fun of me for saying, that’s your answer to everything. But these are, in fact, examples of the essentialist mistake. Okay. If you actually want to have a serious discussion about these subjects, you can’t start with the fuzzy category free will. You can’t start with the fuzzy category knowledge. You can’t start with the fuzzy category creativity or novelty or something like that. It’s you’re literally starting with something that is impossible to define and should be impossible to define, needs to be impossible to define. And you’re wasting your time trying to argue over it. Of course, what’s really going on is that something like say free will, it contains within it many different concepts, some of which are real and some of which aren’t. And it depends on quite an interesting

[01:01:42]  Blue: conversation though, don’t you think? I mean, I wouldn’t, you can have a debate on free will and, you know, I’ve participated in hundreds, I would say.

[01:01:54]  Red: So if what you’re trying to say is Bruce, you’re a hypocrite because you apparently think they’re interesting enough that you ended up participating in the discussion, you’re right. I’m being a hypocrite.

[01:02:07]  Unknown: Okay.

[01:02:08]  Red: On the other hand. Or maybe

[01:02:09]  Blue: your actions speak more to what you really think than your words in a sense. I mean, you can’t convince me that I haven’t got something valuable from many of these discussions.

[01:02:22]  Red: So you’re right. And there is something valuable from these discussions. And yet I would suggest that I’m still saying something correct here, which is you actually can start with a discussion of what type of free will are you talking about?

[01:02:37]  Unknown: And

[01:02:37]  Red: you can very quickly narrow down the discussion and you can quickly say, well, I guess I have in mind the following. And it’s like, oh, no, that doesn’t exist. We can all agree that doesn’t exist. And I guess I had in mind. So an example, recently there was, and the reason why I have free will in my mind is because I just recently listened to Sam Harris was talking with the guy who wrote the book Determined, which I have not read yet, but I’m planning to.

[01:03:03]  Blue: Is that Robert Sapolsky? Yeah,

[01:03:06]  Red: that’s right. Robert Sapolsky. And they were attacking Daniel Dennett’s belief in free will. Well, the whole thing just strikes me as silly. OK, because Sam Harris, when he talks about free will, he rather explicitly defines it as a belief that you could roll back time and that physics would allow you to change your opinion on something, even though actually all the inputs are exactly the same. OK, well, I agree. If that’s what you mean by free will, then I agree with Sam Harris, free will does not exist. And really, we can be done with the discussion because now I can agree with him on that. Daniel Dennett, on the other hand, understands, and he agrees that kind of free will doesn’t exist. And he says so. OK, what he wants to do is he wants to say, but the word free will, the concept, the fuzzy concept free will doesn’t only point to that. It also points to the idea that no one is forcing you into this decision. And that that is a useful way of thinking of the concept of free will and that kind of free will does exist. And Sam Harris in the discussion says, well, I’m not saying that’s not true. Of course, that exists. OK, so there is no disagreement. And then and then after we realize that there’s no disagreement, Sam Harris then spends the whole time. Well, Daniel, then it’s just changing the subject. Well, that is the essentialist mistake. OK, because the fuzzy concept free will that we use when we say free will does not equate to exactly the boundary that Sam Harris wants it to. And that just isn’t the way the word actually gets used. So

[01:04:49]  Red: when he claims Daniel Dennett changes the subject and he uses the example of Atlantis, where Atlantis is real, it’s the city in Sicily that people thought was Atlantis. So you’re really just changing the subject. That’s a terrible analogy. OK, because the word when we say, did you sign this contract of your own free will and choice? That is an example of how the concept and the uses use of the word free will include something that’s very, very real and that has nothing to do with the kind of free will that doesn’t exist. That Sam Harris is harping on. OK, so it’s a really unfair thing that he then tries to paint Daniel Dennett as changing the subject. When really that concept included both subjects from the outset, that’s part of the whole idea of a fuzzy concept is that you can’t pin down the fuzzy concept. You have to almost decide. I’m going to now explain that when I say free will, I really mean something more specific than what you probably mean.

[01:05:46]  Blue: I really liked that Twitter thread recently where Brett Hall sort of defined the free will debate as being oftentimes at least about just an argument about whether humans are special. Whereas so much of the people who are anti free will really the subtext of what they’re really saying is that humans are just another animal. I think that we are not not special in nature.

[01:06:15]  Red: I don’t doubt that that’s sometimes true, but I guess I would question whether that’s true of Sam Harris. I don’t think Sam Harris would fall into that category.

[01:06:22]  Blue: He kind of straddles the line maybe a little bit.

[01:06:25]  Red: So I don’t like Vedin and Ben are really staunchly quote against free will, but they think humans are special. Yeah,

[01:06:32]  Blue: fair enough. Maybe it’s a little bit of a strawman, but I think oftentimes there is that element of having participated in many of these conversations. I think the same people who will on one hand make it a big part of their worldview to emphasize how humans have paleolithic minds and all this that are oftentimes the people who are most adamantly anti free will.

[01:07:04]  Red: You’re probably right. And because the fuzzy category encompasses so many other subcategories concepts, it’s as long as we’re trying to stay at the level of arguing over free will, you really are kind of wasting your time, right? Yeah. It is a very fuzzy category, isn’t it? Yeah. Okay. Now, how do we actually deal with all this? Now, we’re going to get into Stephen Pinker’s answer to this. Okay. If there are no essences, at least when dealing with human fuzzy categories, because there are with say prime number, that how does the mind actually work? Okay. Well, that is in some sense the question we’re trying to answer. And we do know a little about it. Okay. So science has come up with an interesting and testable theory around the idea of a prototype or a prototypical example of a category. So now this is from page 126 of how the mind works, Stephen Pinker. So even concepts like bird seem to be organized not around necessary and sufficient conditions, but around prototypical members. If you look up a bird in the dictionary, it will be illustrated not with a penguin, but with typically a sparrow. The idea is that some members of a fuzzy category are more prototypical of the category than others. A pigeon and an ostrich are clearly both birds, but they are not equally prototypically birds. This may be what we really, what is really going on when we talk about a conceptual distance. Pigeons are near the center of the fuzzy category of birds or sparrows or near the center of the fuzzy category of bird. And ostriches are outside the center and are closer to the edge of the fuzzy category. Now, I said it was testable.

[01:08:51]  Red: This is a testable theory. And in fact, it is, and it has been tested. Okay. What they do is they’ve actually measured, this comes from surfaces and essences, is you’ll read a sentence like the pigeon was approaching or the ostrich was approaching. And then the next sentence is the bird was now just a few yards away. And then they measure how long it takes you to comprehend the sentence, the bird was now just a few yards away. And it will actually, it’s actually harder for you to make sense of the sentence. The bird was now just a few yards away. If the first sentence was the ostrich was approaching because it’s harder to process that an ostrich is a bird, even though you know it is because it’s not a prototypical member of the category bird. So Pinker and how the mind works talks about the same experiment, page 126. Experiments in cognitive psychology have shown that people are bigots about birds, other animals, vegetables and tools. People share a stereotype, project it to all the members of a category, recognize the stereotypes more quickly than the non conformists and even claim to have seen the stereotype when all they really saw were examples similar to it. Their responses can be predicted by tallying up the properties that a member shares with other members of the category, i.e. the more birdie properties, the better the bird. This also explains why people often call a spider an insect, even when they intellectually know it is not part of the zoological definition of an insect. It still fits within the fuzzy category of insectness, nevertheless. This also shows why context matters so much. A

[01:10:36]  Red: spider is an insect, according to Hofstetter, in some contexts, contexts, such as there’s a crawly thing in your bedroom, right? You don’t really care if it’s a spider or a rachnid or an insect. But in others, it’s not. When you’re taking a test, a spider is not an insect and we all know that. So on page 58 of services and essences, he says, context thus changes categorization and can modify how we perceive even the most familiar of items. For example, an object can slip in the blink of an eye from the category chair to that of stool when the light bulb has just burned out. In fact, objects actually have a huge multitude of categories at once. How many categories do you personally fit into at this very moment, right? Father, son of somebody else, teacher, Peter. You fit into numerous categories simultaneously. Okay. And there is no single category that you fit into. And that’s a useful thing. Now, here’s the thing, though. We are not always fuzzy. And there actually would be a problem. If we could only be fuzzy, that would be a problem. And if we were never fuzzy, that would be a problem. But this is now from page 126 of that. People are not always fuzzy. We laugh at Opus from Bloom County because a part of us knows that he really is a bird. We may agree on the prototype of a grandmother, the kindly gray -haired septuinary and dispensing blueberry muffins or chicken soup, depending on whose stereotype we’re talking about. But at the same time, we have no trouble understanding that Tina Turner is also a grandmother. So page 127 of how the mind works.

[01:12:30]  Red: Examples of unfuzzy thinking are everywhere, a judge may free an obviously guilty suspect on a technicality. Bartenders deny beer to a responsible adult the day before his 21st birthday. We joke that you can’t be a little bit pregnant or a little bit married. So that’s actually the answer to your question, Peter. So in math, obviously, there’s a need, and in science, there’s obviously a need to define some categories as what we might now call precise categories that do not have fuzzy boundaries. But in fact, humans do it all that time. And we were going back to the free will discussion. What we’re really doing, as I was explaining the stupidness of arguing over the term free will, when what you really should do and when what in fact Daniel Dennett and Sam Harris have done is they’ve taken the fuzzy concept of free will and they’ve split it into two far more precise categories, one being you roll back time and the laws of physics allow you to make a different decision. And the other one being that you’re not currently being compelled by something outside of yourself. These are nowhere near as fuzzy. I don’t know if things are as clear as you’re either fuzzy or you’re precise, but clearly now we’re talking about something way more precise, where you can figure out membership of the category and it doesn’t have a central core and the same fuzzy boundaries anymore. And if people were willing to do that, in a case like this, you would quickly find there’s just nothing to discuss between Sam Harris’s opinion of free will and Daniel Dennett’s because they never disagreed to begin with across the actual subcategories that were more precise.

[01:14:15]  Red: So I’m going to argue that yes, we do in fact use precise categories and that they live in our minds alongside the fuzzy categories and that we can in most cases tell the difference between the two. So Hofstadter argues that while we use analogies to expand a category or concept, the original concept does not disappear. Tim as an adult has a very rich mental category of motherhood, yet never feels any confusion as to the original concept of his mother. Pinker now page 127 of how the mind works says fuzzy and crisp versions of the psychologist. I think this is fascinating what he is about to explain. Some psychologists mysteriously gave the standard test of fuzzy categories to university students, but asked them about knife edged categories like odd number and female. This would have been before the whole transgender thing where female is now far more a fuzzy category. The subjects happily agreed to daft statements such as that 13 is a better example of an odd number than 23 is and that a mother is a better example of a female than a comedian is. Moments later the subjects also claimed that a number either is odd or even and that a person either is female or male with no gray areas. People think in two modes they can form fuzzy stereotypes by uninsightfully soaking up correlations among properties, taking advantage of the fact that things in the world tend to fall into clusters, things that bark also bite, lift their legs at hydrants, etc. But people can also create systems of rules, intuitive theories that define categories in terms of the rules that apply to them and that treat all the members of the categories equally.

[01:16:07]  Red: Now, does this have any relationship to AGI or at least AI? It does and on page 126 Pinker explains why neural nets easily implement a fuzzy logic in which everything is kind of something to some degree. To be sure many common sense concepts really are fuzzy at their edges and have no clear definitions. The philosopher Ludwig Wittgenstein offered the example of a game whose exemplars jigsaw puzzles, roller derbies, curling, Dungeons and Dragons, cock fighting, and so on have nothing in common. Let’s go back to now Carl Popper and his discussion about essentialism. Okay, so Carl Popper tries to deal with this fact that human beings have both these fuzzy categories. No, Carl Popper never uses the term fuzzy category and precise category. Okay, so let me first say that, but I feel like he’s still trying to deal with some of the same problems that exist, right? So he says while the essentialist reads a definition normally that is to say from left to right, we can say that a definition as it is normally used in modern science must be read back to front or from right to left. For it starts with the defining formula and asks for a short label to it. So he says the scientific views is from page 14 of volume two of Open Society and Its Enemies. Also on page 14, the scientific view of the definition of puppy as a young dog would be that it answers, it’s an answer to the question what shall we call a young dog rather than an answer to the question what is a puppy? We can at once see from this that definitions do not play any very important role in science.

[01:17:58]  Red: Science does not ask how we got an idea. It is only interested in arguments that can be tested by everybody. I love that part because it goes along exactly along with what I was talking about in our epistemology poppers two axes episode that poppers epistemology really was originally based entirely around this idea that science is defined by the fact that we can actually

[01:18:25]  Blue: make arguments that can be tested by everybody.

[01:18:27]  Red: Now with this in mind, this idea of the way you would normally read a definition from to back versus how does he put it? He says from left to right instead of from right to left. So I think that this kind of gets at something that’s kind of interesting that I want to kind of discuss now. So let’s go back to those that can’t accept the immune system creates knowledge and it’s just they’re overwhelmed by this idea that it just doesn’t feel very knowledge like. Okay. There are actually two things we’re talking about here and I made so many attempts to try to explain this and it’s very difficult to explain this to people. I’m going to explain why I think that is. So the first is the fuzzy category knowledge. There is no essence or criteria that defines this category because words just point to fuzzy categories and they never have such necessary and sufficient criteria. But they do have prototypical examples and this is what the two sources hypothesis is. The most knowledge like examples of knowledge are the kind of knowledge created by biological evolution and human minds. This is the same as reading from right to left. Okay. What is knowledge? Knowledge is defined by the prototypical examples, biological evolution and human ideas. And when you’re thinking of knowledge in this way and you’re trying to answer the question what is knowledge? What you’re going to end up with is a fuzzy category that has no defining characteristics.

[01:20:05]  Blue: Well maybe that’s not so bad from this perspective.

[01:20:07]  Red: It is not so bad. Okay. The second thing we’re talking about is a precise category and the precise category here is adapted information that causes itself to remain so. This non -fuzzy or at least less fuzzy category needs a convenient label and David Deutch happened to slap the label on it knowledge. Okay. In his book, the beginning of infinity. This is an example of reading from right to left. So we’re starting with the concept adapted information that causes itself to remain so and we’re slapping a label on it and the label mean is not intended to be the same as the fuzzy category knowledge. It’s just a convenient way of speaking of this precise category, which is a subset of the fuzzy category. Okay. Now people are constantly because David Deutch happened to slap the word knowledge on it when that word already can already pointed to a fuzzy category. People are constantly confusing these two and Deutch himself confuses the two at times and that was really what I was getting at with our six episodes on knowledge is that Deutch accidentally slips between the fuzzy category knowledge and the precise category adapted information that causes itself to remain so that he happens to label as knowledge and he sometimes doesn’t notice he slipped between these two. So to avoid confusion, I’ve suggested in the final episode of that series that we call this precise category adapted information that causes itself to remain so that we call it simul knowledge instead of knowledge to try to avoid this confusion that keeps taking place. So the immune system does create adapted information and it does so via a blind variation and selective retention algorithm.

[01:21:58]  Red: Go back and listen to the episodes and we actually go through exactly how it does it. It is very clearly doing it as through a variation selection algorithm. So it is a case of adapted information that causes itself to remain so. So let me just very quickly remind people how this works. The system notices that there’s an antigen and it puts out antibodies which contain DNA and it hyper mutates the DNA and the successful antibodies, they are allowed to reproduce more and the unsuccessful ones are not allowed to reproduce as much. So the successful ones outproduce and eventually and so basically because this recipe contained in DNA, now mind you, so it’s genes but not genes that are in the sex genes, they’re in the sex DNA because they were hyper mutated in the life of the animal, the human being or the dog or whatever. And they still follow exactly the evolutionary algorithm and they even use DNA. But they do so in a far more constrained way, not following the they’re not open -ended. It wouldn’t want it to be open -ended. You don’t want your immune system to create an airplane when all you really want is to be cured of the antigen that you’ve got. So it makes sense that it’s now constraining it to just certain genes that it’s hyper mutating. Now, when you do this, this is absolutely an example of adapted information that causes self -dermain. So it even does it for exactly the reasons Doge is laying out that it was useful.

[01:23:41]  Red: It’s a recipe for how to build an antibody and the useful ones stuck around and continued to replicate and it’s replicators, by the way, using DNA, using genes, literally using genes and the ones that weren’t useful, they disappear, they die out. So the immune system is a mini biological evolution algorithm, but in a constrained setting. So it is absolutely within the precise category adapted information that causes itself to remain so, which we’re now calling simul knowledge. But it’s also clearly an ostrich, not a sparrow, when we’re talking about the fuzzy category of knowledge. Now there’s a fairly obvious relationship between the two categories, but they can never be made to be equivalent because one is an ever -changing fuzzy category based on prototypical examples and the other is a scientific definition, i.e. a convenient label for a precise category. When we want to speak of adapted information that causes itself to remain so, we use the string of letters K -N -O -W -L -E -B -G -E to refer to it as a simple convenience, but it is not the same as the fuzzy category knowledge, nor can it ever be, nor can there ever be a definition that would be the same as that because it is a fuzzy category. It has no precise definition. Now I don’t feel like this is necessarily that hard to understand. In some very real sense, we are not that bad at understanding that there’s a literal, precise motherhood, and then there’s all sorts of metaphorical, more or less metaphorical versions of motherhood. We do have this ability to understand that an ostrich is a bird, while also understanding that it’s not a prototypical example of a bird.

[01:25:45]  Red: Why are people having such a struggle to understand that I’m talking about the precise category, adapted information that causes itself to remain so, and they’re talking about the prototypical category of knowledge, the fuzzy category of knowledge? Here is a nearly true account of a real conversation I had with someone that we previously called James. I would start with, Deutsche is confusing two things. He tries to define a precise category of adapted information that keeps itself instantiated, which he labels knowledge, but he also claims that an algorithm created by a genetic programming algorithm is not knowledge, even though it is an example of that precise category, adapted information that causes itself to remain so. James says, no, you’re wrong, it’s not knowledge. I’m not saying it is. I’m pointing out that under Deutsche’s definition, that it would count, because it’s a precise definition when he puts it that way. I’m not saying I agree or disagree. No, it doesn’t fit his definition. Okay, but it is James adapted information that keeps itself instantiated. It’s not knowledge because it’s created mechanically. Okay, sure, but I’m not talking about if it was or wasn’t created mechanically, because that wasn’t part of Deutsche’s definition in the beginning of infinity. I was talking about the precise category, adapted information that causes itself to remain so. Note the lack of the additional criteria of that was created non -mechanically. By the way, I’m not even sure what that means since all algorithms are mechanical. James, no, Deutsche makes it clear it isn’t knowledge if it was created mechanically via perspiration. Yes, that’s what I’m saying.

[01:27:26]  Red: I’m saying there is a contradiction between the two ways he goes about trying to define knowledge, that there is something missing with the first definition he gives, which was adapted information that causes itself to remain so. Perhaps the thing that’s missing or the mistake is exactly what you’re saying, James. It’s that we need to add an additional criteria must be created, not be created mechanically. No, there’s no contradiction. The immune system does not create any knowledge. Look, I’ll call the original definition, adapted information that causes itself to remain so, simul knowledge, so that we can conveniently talk about it just like Popper says. As you pointed out, you don’t consider this to be the same as knowledge, but it is something. Isn’t it, James? No, it is not knowledge. I refuse to call it simul knowledge or allow any terms for it. I’m not saying it is I’m saying that it is not knowledge, James. I’m agreeing with you now, but whatever it is, it’s real enough. We need a name for it. No, it’s not knowledge. You need to prove to me that it is of any interest. If you can’t prove it, then I refuse to name it. That’s what I just said. I said it was not knowledge, but it is adapted information that causes itself to remain so. You basically admitted it was a real thing by claiming it was uninteresting. No, it’s not knowledge. I’m agreeing. Please stop acting as if I’m saying otherwise. No, it’s not knowledge. This is actually very close to a real conversation I had with James.

[01:28:54]  Red: I never could break past this point where I could get him to see that there’s a difference between the fuzzy category knowledge and the precise category, adapted information that causes itself to remain so, and that there is some sort of play between them that’s worth talking about. Now, what is going on here? I actually think there’s two things going on here. Why is James struggling to even make sense of what I’m saying? It’s not that we’re even really disagreeing. He at times even agrees with me in that conversation, but he keeps wanting to emphasize we’re not going to use the word knowledge for it, and that’s kind of like the main thing he wants to bring up. Even me trying to rename it something else, he’s not willing to go there because somewhere in his mind he knows I’m secretly saying it’s a kind of knowledge, and he wants to make it clear it’s not a kind of knowledge. But why can’t he see that there is a precise category, adapted information that causes itself to remain so, and that the fuzzy category based on prototypical examples, the two sources hypothesis, is different from that. The immune system does create recipes. He would have to admit that. They are adapted information. Genetic programming algorithm does create an algorithm. That is adapted information, and it’s trivial to see that these are cases of adapted information that was useful, so it got selected based on a blind variation and selective retention process, while its competitors are now gone. So here, think of the algorithms that failed to make the robot walk. They’re gone now, right? They’ve completely erased in favor of the one that did cause the robot to walk, or the recipes that create the antibodies.

[01:30:33]  Red: The ones that failed, they’re all gone. Those hypermutations, those genes disappeared. They’re no longer even in your body anymore, okay? But the ones that stuck around and created the antigen, sorry, created the antibody to fight the antigen, those at least for a while stick around your body, possibly forever for some diseases, and your body holds onto those because it wants to remember how to fight off that disease when it comes back again. So why is James incapable of having discussion about whatever it is we want to call this type of adapted information that he wants to call not knowledge? Now, this is where I think things get interesting, okay? Because this is actually, in some sense, what James is doing is he’s proving a certain aspect of Hofstadter’s theory correct, okay? We talked about how analogies force themselves on us. So they come upon us and we don’t experience them as conjectures. We don’t experience them as fuzzy categories. We experience them as inferences. And so page 31 of surfaces and essences, unsatisfied with merely being an agent that enriches our comprehension of a situation we are facing, the analogy rushes in and structures our entire view of the situation, trying to make us align the new encountered situation with the familiar old one. And notice also that this is consistent with Popper’s own view, okay? Popper says, page 60 of conjecture and refutation, instead of explaining our propensity to expect regularities as a result of repetition existing in the world, I propose to explain repetition for us as the result of our propensity to expect regularities and to search for them. And this is actually Popper’s version of the same theory that Hofstadter is talking about,

[01:32:29]  Red: which is why the two theories do have some really important commonalities and shouldn’t just be dismissed, okay? Popper is saying literally that human beings are pre -programmed to expect repetition, to expect regularities, and we theorize about what those are, and we’re sometimes wrong about it. We will enforce it. We will try to imagine a regularity that isn’t there, says Popper, because we’re expecting a regularity, okay? This is the explanation for why that gentleman with after one woman who was a journalist, who was a writer, and he found her attractive that he was rushing to meet another woman that was a writer for his university because he was trying to impose a regularity that in fact didn’t exist upon reality, and that is how our intelligence works, both according to Hofstadter and Popper, okay? Now to James, the fuzzy category knowledge based on prototypical examples, the two sources hypothesis overwhelm his ability to even think about the more precise category of adapted information that causes itself to remain so that doesn’t by itself feel knowledge like enough, okay? There is also, I think, a bit of dogmatism going on here. Deutsch’s theories are not always treated by people who are fanaticist theories as ideas that are to be error corrected. They’re okay, and that’s not necessarily bad. Dogmatism can be a good thing. We’ve had several episodes about that, okay? But you can see how if you had just a bit of, this is my meaning meme, so I’m going to be dogmatic about it, mixed with the fact that the fuzzy category enforces itself on you.

[01:34:21]  Red: It just comes upon you that adapted information causes itself to remain so by itself includes things that are ostriches when it comes to knowledge and just don’t even feel knowledge like, okay? And that since James is really trying to argue over specifically the fuzzy category knowledge, he has blinded himself from being able to see the subcategory adapted information that causes itself to remain so, and he’s not able to reason about it because it’s getting blown away by that fuzzy category that’s kind of coming upon him, okay, and forcing him to order his world around it. And this is what Hofsetter’s arguing happens to us. Now, so strange as this sounds, this whole thing, the constructor theory of knowledge, and the six episodes we did on it, and my debates with Deutschians around it, they’re actually an accidental example of how Hofsetter and Pinker are correct about how humans think, at least in this particular instance. We do not, for the most part, think in terms of precise categories. We’re capable of it, but most of the time we think in terms of fuzzy categories, and fuzzy categories are based not on necessary and sufficient criteria, but on prototypical examples, such as, and that’s why the two sources hypothesis is so overwhelmingly intuitively seems right, because those are the prototypical examples of knowledge. They are the prototypical examples of knowledge, at least within our culture, okay, maybe it would be different if we lived in the middle medieval ages or something, right, but we live in a post Darwin world, and to us, those are the prototypical examples of knowledge,

[01:36:07]  Red: and it’s really hard to see past those prototypical examples, especially if we’re feeling a little dogmatic, or we’re defending something that feels like a meaning mean to us, and they overwhelm us and they force themselves on us, and that feels like knowledge to us, right, the knowledge of what knowledge is, is the prototypical examples, the two sources hypothesis, and this is why the ancient plain plainists thought knowledge was obtained by defining things, and that’s why when I’m arguing with Deutschians about this, I say you’re actually just being an essentialist, okay, you’re trying to argue with me over a fuzzy category, and I’m trying to talk about a precise category, and the two aren’t the same, and I admit that, I admit they’re not the same, so you have to actually, just as you have to actually stop and think and realize that an ostrich is a bird, like we know that we tested this, human beings struggle to accept an ostrich as a bird, because it’s not a prototypical example of a bird, okay, just as you have to stop and think about to realize an ostrich as a bird, but you know it’s correct, but you literally need more processing power to be able to see that fact, so if a person’s feeling a little dogmatic, it can become impossible to see past the fuzzy category, and to see the underlying concept that I’m pointing to, adapted information that causes itself to remain so, so that would be kind of my summary of today’s episode, before we leave though, I do want to ask one question as food for thought that I’m not going to give an answer to, and

[01:37:42]  Red: I don’t even know if I know the answer to it, but I want to actually bring it out, because I’ve been kind of dancing around it, ignoring it, but it’s kind of like the elephant in the room, here is the question I would like to ask, that I would like people’s opinions on, how strongly should we take Popper’s claim that from page 14, we can see, we can at once see that this, that from this, the definitions do not play any very important role in science, and continuing the quote from him, the methodological novelist will never think that a question like, what is energy, what is movement, what is an atom, is an important question for physics, okay, here’s the thing, and I really want people to stop and think about this, and maybe write to us and say, you’re on opinions on this, because I’d really like other people’s opinions on this, isn’t Deutsche’s constructor theory of knowledge, in some sense, trying to answer the question, what is knowledge, okay, and isn’t that exactly, and it is physics, right, constructor theory is physics, so it does, who’s right here, okay, and I say who’s right, but there may be ways to reconcile the two views, and I’m actually hoping that there is ways to reconcile the two views, Deutsche is asking questions like, what is knowledge, and he’s specifically trying to come up with a definition of knowledge that makes it so that it isn’t based on,

[01:39:07]  Red: isn’t based on subjective knowledge, doesn’t require a knowing subject, okay, now Popper, by the way, did the same thing, he tried to come up with a definition of knowledge using his three worlds hypothesis, trying to define knowledge in such a way that it wasn’t dependent upon a knowing subject, is, was Popper wrong to do that, because that’s an inappropriate question for us to be asking, and only a word essentialist would ask such a question, okay, should he, should Deutsche, instead of asking the question, what is knowledge, should he just skip definitions altogether, and start talking about what can we do with knowledge, okay, and by the way, would that even make sense, in terms of the constructor theory of knowledge, so I’m not going to answer that question today, if you think you’ve got an answer to that, I’d be actually very happy to hear out your opinion on this subject, this is one I feel like I need more to give more thought to, and try to come up with my own answer to this.

[01:40:04]  Blue: Well, I think it’s an extremely compelling way to put a potential, potentially logical inconsistency in the thought of Hopper and Deutsche, I mean, the most intuitive answer, I guess, is kind of what you’ve said, basically, is that we, that word essentialism, or looking deeper into the definition of words, does have its place in the world, but perhaps we shouldn’t become too sentimentally attached to these theories either, I mean, in some ways that’s the spirit of critical rationalism, right?

[01:40:39]  Red: Yeah,

[01:40:39]  Blue: yeah, interesting. Anyway, that’s my initial thought, and I hope that we can explore this further, and as usual, I’ve gotten a lot out of what you’ve said today, Bruce, and I hope our audience does as well, and it seems to me that you are truly embodying the spirit of critical rationalism by criticizing the best theories that are out there, and you’ve made some great connections between Deutsch and, or I guess more from Pinker and Hofstadter and Hopper today, but I thank you for that.

[01:41:16]  Red: Oh, thank you. Talk to you later, Peter.

[01:41:19]  Blue: Okay, bye, Bruce.

[01:41:21]  Red: Bye -bye. 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 -4 -strands. 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.