 Hello and welcome everyone to active inference Institute. It is October 20, 2023. And we're here in active guest stream 61.1 with Shagor ramen. So today we'll be having a presentation myth of objectivity and the origin of symbols, followed by a discussion. So thank you very much for joining. Looking forward to the talk. Thank you very much for coordinating this. And thanks for all the work that you and the active inference institute does. I also wanted to briefly thank Carl Froston and Paul Badcock for helping me with this paper and making it a reality. So this is the title of the paper here. So it's called myth of objectivity and the origin of symbols. And it's in frontiers and sociology. And so scared into it as Daniel said, do hopefully a brief overview, and we can do a discussion at the tail end of this. And so I guess briefly the thesis that I'm trying to argue for is that some of these words will need clarification, particularly in the context that we use them. But it is that the belief in universal and kind of a codified egalitarian social norms and other words morality. This is essentially underwrites broadly our use of symbols. So what is that kind of mean functionally in terms of what other disciplines use or how they would think of our use of symbols. Well, there's cultural niche. And this you can think of as, you know, our ability to engage in cultural evolution from Joseph Enrique and others who have highlighted that as an area of an important area of humanities distinction. Another one is social reality. And this is more in the psychological realm, but you know how we use things like money, you know, how we obey someone with a crown on their head. So just things that are socially constructed. Realities are socially constructed as opposed to, you know, existing in more of a physical sense on that other non human animals do. And of course, this gets to our symbolic capacity. So this is your symbolic consciousness, our ability to kind of think deeply in the past to playing deeply in the future. Then to engage with any number of concepts, as well as people. And so the claim, we can divide this claim into three areas. The first is our modern social dynamics and social bias. And so, you know, how this is the sociology side. So, you know, how our economic systems look, how our civic and political institutions look, and kind of building on that. So I think once we have the structure was hopefully I convince you of a way we can look at humanities structure of their social reality and other kind of symbolic reality. Then, you know, the question is, well, how did we get there? You know, how did we become the kind of, you know, primate that points fingers and feels guilt. As you can kind of see that picture. And for this we can look to, you know, two areas. So one is, you know, non human social animals. So primates, you know, crustaceans and other animals that engage in social behavior of a similar nature. So this is distinct right from you social insects from may know ants and bees. But rather this is, you know, chimps, you know, bonobos. Remember some monkeys and some others that have been studied specifically, you know, by researchers seeking to examine humanity social norms and how that differs from primates and other animals. And lastly is the end of anthropology. So whatever we're able to, you know, determine about our, you know, the social environments of our ancestors, you know, what can that tell us about how we got here. And I'll make some kind of specific claims in regards. Yeah. And so it's going to set out my goal and essentially my goal is to kind of tackle that first part and to, you know, persuade folks, you know, the call here that our social system takes on a particular structure. And that structure is essentially a symbolic self. And I use that in the sense of Carl Jung, meaning that we inhabit, in addition to our own identities, we conceive of ourselves, our self-consumption is attached to a number of other broader cultural identities. And these could be your family, this could be in your country, this could be your company, this could be your community, kind of thinking those five C's of country, company, clan, etc. And so that's essentially where that star is that that's where I'd like to get to by the end of today and also get into some other areas as well. And so to explain briefly what I mean by the myth of objectivity. So first of all, by myth, I mean to say that we haven't, it's not that we've obviously identified some kind of underlying structure with reality or with our social systems that our symbols represent. So think here of Plato's forms that there's a number of chairs out there, but there's a kind of an ideal form of chair. That's not the claim I'm making, one way or the other. So instead of what I'm saying is, well, during the we asked the scene, we were just like another primate, trying to solve primate problems. And the way we did that, or one way we did that effectively was through this was through what constituted essentially a myth with essentially establishing a common reference frame for behavior. And I'll get into that term in a bit in a second. And the second piece of this, the objectivity. So that's to mean that it's not simply, it's not simply a claim that your ancestors listen to stories and believed in myth and believed in religion. But that our myths that were powerful were practical, were meant to solve social problems, and do so in an objective way. So be essentially more scientific. So one of the things that I'm making is that our ancestors kind of similar to how we are today. Or sorry, similar to how some of us are, are interested, we're interested in identifying actual real world consequences in a symbolic matter. So having some sort of objective truth, and kind of believing that an objective truth exists. So kind of right and wrong exist in an objective way, couldn't bad exist in the sense of, you know, this can kill me, this can't kill me. And objective in the sense of, you know, this is part of my group and this is outside of it. All right, so roughly how do we get there? So a little more specific on the anthropological side. I'm going to look at things like Shibboleth. So this is our most innate social bias, which is important because for a few reasons, but one is that societies, social, social units even outside of the human spectrum. Some researchers have used things such as social bias to analyze the structure of non-human animal social units. So essentially in group preferences, social biases in that respect are ways that we can understand how other animals, how chimps and bonobos interact with one another, how they establish friends and distinguish between those and essentially how they exist in these normative social units. And hence our use of Shibboleth speaks to kind of a unique way that humanity expanded upon our social biases. And so although it's a crux now, you know, it's often seen as problematic. There's a certain contingency of folk that see Shibboleth as a necessary step in our evolution. So I adopt that view and see that as an important kind of factor in our road to morality. And so within that kind of Shibboleth environment, which is essentially we can think of as still non-symbolic because Shibboleth, as I should probably explain what Shibboleth is. So this is essentially accent bias. So we're kind of taken from the Old Testament, but it represents our kind of innate tendency to associate and to establish an instinctive kind of common ground with folks who sound like us. And so we automatically trust folks who grew up near us, who have a similar accent. And the further out we get, oftentimes, the more foreign someone can see. Not always the case, but kind of instinctually and prior to the acquisition of symbolic identity, we'll typically practice Shibboleth as a way to identify. And so assuming that this practice of Shibboleth goes kind of far back into our past, then how do we transition out of that to something more symbolic? How do we start using morality? How do we start using language? Well, here are my suggestions that we use one of, or one way to think of it is using what's called a speech act. And so this comes out of the field of the philosophy of language, John Surley and Austin a bit earlier than that. But there's speech acts called performative and constitutive utterances. And these are essentially acts that immediately change the environment in some way. And so if I suddenly tell you that you're fired or that there's a war across the seas right now, that has immediate consequences, even though it was just a speech act. So before any violence happens, before someone is now kind of crowned a king and starts implementing physical policies just by speaking, you've changed something. And so in this kind of example, some of my kind of stylized example is that, well, someone committed some kind of act of deviation. They violated some norm. And so some vocalization was admitted that labeled that person. So this person should have been an in-group member, this person kind of should have been aligned with our in-groups norms, but they violated it in some way. They should have been, they violated food sharing, they violated some kind of unwritten contract about cooperative breeding, like how we should protect infants. And it could have been a number of different norm violations, but labeling that person as deviant would constitute to kind of a minimal level of a speech act that could be read as symbolic. And so the starting off point here actually is, or one of the important starting off points here actually is what is a self? And so symbols are one thing, and we can probably identify that and have some discussion, but we have some kind of clear idea of what the self is. Sorry, what a symbol is. It turns out however that define the self is a bit, it's not as confirmed and locked in. As at least I would have expected kind of when I started this project. And so typically agents are defined and we'll get to that in a second, they're defined as along their person, but it's not clear where say a cell starts, where an organ starts, where a person starts to where society is. And so what we'll get at here is using some frameworks, we'll see how we can understand bias perspective shared goals, reference frames, and the next topic, which is the free energy principle, see how we can use those to start understanding, you know, how self and how identity are how we can think of the self identity in the context of, you know, our creation of simple. And so, yeah, roughly, I mentioned some of these already, but we're going to look at as non human sociality. We'll look at your religion with a narrative that this is, you know, of course a very instinctual very ancient practices that defines humanity. We'll get a social bias, as I mentioned, and then ancient shibble and most importantly, you know, this is the active inference Institute. And so the framework that I'll be positioning this within is the free energy principle, and more specifically active inference. Just one more. So again, free energy principle. We'll go over the ideas of identity and how we can start to think about the subject and how that is evolving in the light of these new frameworks of the free energy principle active inference and some other researchers were sympathetic to these views. And then the the weird journey at all explain what that's an acronym so explain what that acronym is in a bit but essentially for now it's just, you know, Western society. So Western society is known for being particularly analytical. And as a result of that, one of the consequences of that is most recently and this is, you know, around the past century so around 20th century a lot of disciplines emerged that did something that had a common approach, which was an attempt to, you know, reduce uncertainty, identify agents and, you know, provide kind of explanatory frameworks in the field of that first one post-structural this is in the field of linguistics. The second one of course is economics. And the third one is behaviorism on the field of psychology. And so we'll just kind of go through those and see what we can learn from these paradigms when we try, you know, when I try to essentially, you know, convince and discuss humanity as, you know, an agent that occupies a symbolic self. Yeah, so going the route that I mentioned so far, we'll also discuss what this means for moral philosophy. And, you know, of course, this will, we'll just kind of dip our toes in. We won't go too far into moral philosophy, but I think that the implications are kind of immediately clear in some areas that bring some interesting ideas from the past of, you know, what is a bit more relevant and what's a bit less relevant. And whether, you know, effective altruism and things like that are kind of consistent with this view or kind of whether they're at odds. In the last piece, if we have time just because I know there was a similar stream on cooperative communication, so I'll kind of give my view on how this kind of aligns with cooperative communication and communication and viewed as kind of an ecological framework, which, you know, both of those were reportedly done within the Active Inference Framework. So, yeah, and so to start things off, the free Andrew Principle also just kind of give a brief explanation. So this is the Active Inference Institute. So there's obviously, you know, a great kind of wealth of information that it explains the paradigm and how it's useful and some of the terms. But essentially it's a framework that essentially attempts to identify a number of formalisms that describe the science of self-organization or autopiosis. And it represents agents, things, objects, everything as essentially statistical models of their environment. And so, more specifically, agents need to possess that minimum. They need to possess a generative model of their environment in which they can understand, in a sense, or infer the processes of the world. So, you know, how does a, you know, a creature navigate his environment? How does an ant, you know, find its way back home? They have the, they themselves are, can be represented as models of their environment. So we can see that as these agents come encoding Bayesian beliefs. And so one way to look at it, and this isn't part of the formalism, that's why it's italicized. But I heard Chris Fields say it, so I'll say it here, which is that one way to think of it is, you know, agents are essentially all scientists. So they have beliefs about the world, they act on the world in a certain sense, and those beliefs get updated. So everyone has a running hypothesis of what they should next do, who they are, and how they should maintain themselves, their environment. And so at a basic level, you know, everything that exists in the world has to engage in some sort of science. And this occurs at various scales. So this is at cells that organs, you know, agents, people, or animals, a few of societies and culture in that kind of last area, of course, is where our focus is here today. And this is italicized as well, but one of the takeaways from the 300 principle is that a foundational element is the search for one cell. And so it's not a given that we understand precisely who we are in the world, and it kind of extends to who your friends are, and who your enemies are. But this surprisingization can be thought of as this Socratic search for self, you know, meaning Socratic and Socrates' dictum to know thyself. There's something ancient there and not just, you know, for moral philosophy. Yeah, and one other point to make is that although belief in the fringe principle context is a technical term, it's refers to Bayesian statistics. It does actually come out of a use at a higher level. So Thomas Bayes was a preacher and he was specifically talking about religious faith. And it was a formal and he made a formalization for how we he could think about his religious beliefs. And so that's ultimately what led to Bayesian statistics. And so one attempt or one takeaway I have is that we can think of, due to fringe principle scale and very nature, we can think of beliefs that we typically use the term on religious context and another context. We can, you know, try to formalize these within the fringe principle, not as a given, you know, we could get things wrong. But it's at least conceivable that they're contained within that. That's the next area. So in addition to the fringe principle, there's, of course, other works that's been done to understand identity, to understand itself, and whether that exists kind of at the bounds of the skin, say, or whether that exists, you know, within social units. And so one interesting, one interesting piece that's been done is by Michael Levin, who focuses on bio electricity at the cellular level and how cells communicate. And there, he actually used game theory, which is a paradigm typically used in economics, but he demonstrated how cells, if we can think if we can model them, not just as kind of individuated. Little agents try to maximize their own interests. If we can set model them as potentially inhabiting a wider self, we can see how cooperation would emerge in a way that more competitive paradigms, more reductive paradigms would not. And so immediately, even before we get to how humanity acts with other humans, whether we do that cooperatively, whether we use morality, whether we use other norms, this same problem exists at the level of the cell. Because in the second piece is once we get to chimpanzees, if we're over in the upper right hand corner, what we understand there is, from Mark Moffitt, chimpanzees determine their group size by social intimacy. And so that means that their group size essentially cannot exceed, generally does not exceed what any individual chimp is able to recognize. So they basically have to recognize, so each person, each member of the group has to recognize everyone within that, as opposed to, you know, as opposed to us and humans, which he argues has existed for, you know, far into our past, we've been an anonymous social units. Right. And so here, what I did was, before I get to further down the line, what I did was take what I took another concept of Michael Evans, which is this light cone. So the light cone essentially is one way of representing identity. And so the scope of each agent's goals can be kind of captured in this idea of a light cone. And so in the area of a chimpanzee, their goals are associated with others in there that cohabitate in their social unit. And they're necessarily subjective. And so they're necessarily based on their own perspective. And so another piece of social identity about our way to look at it. And so there's a debate whether or not we can model animals as being in societies. And so this isn't a given. And so there's sort of an argument between, you know, dockings and Wilson, for instance, that, you know, claim that each side kind of claiming that you can, you either can think of animals as existing in social units and those operate similar to cells or the can. One way around that, on this point, our three researchers found that you can model them actually as just possessing in group favoritism. So if instead of kind of worrying about specifically where the boundary is from our point of view from, you know, an outsider's point of view, if you said model it as well, each individual agent. So like I said, each one maintains an intimate recognition of others within their in group. Following that they show preferences to those inside. So this could be as simple as when there's a fight when there's an attack from an outside group, you will prefer those within your anger versus out group. It could be food sharing could be hunting and things like that. But it doesn't have to be extensive cooperation. It could be simple preferences could be avoiding spending more time around your anger versus outside your outside group. And so this is one of the ways in which social identity, you know, exists and at this, at this level. And so another important piece is which Judith Burkhart in colleagues measured which was implicit norms and subjective norms that are exhibited by these chimpanzees. And so, as an example, if chimps saw an in group infant being harmed, they would experience a visceral reaction, they would have high end of razzles, they would pay attention. And so this kind of suggests that there are norms that chimpanzees and other animals and other animals exhibit and some animals even exhibit a more extensive, more extensive norms. There's more misalmonkeys. But chimps are on the lower, they still, they're on the lower end of that spectrum, but they still exhibit norms. And so one way that Judith Burkhart distinguished between primate morality and human morality was to say that we have a top down third party perspective. And so this is where I get the goofy graphic where there's a triangle on top of each of these, each of one of our ancestors heads is because our ability to engage in objective morality can be represented by the shift between a perpendicular perspective where I'm looking at those around me and I'm imagining them looking back at me versus I'm taking a top down perspective. And evaluating my actions and the actions of others in the context of some more transcendental perspective. And so this can that object, that view of objective morality versus subjective social norms will say I was also brought up by back off in Pierce, they did claim that morality exists in other animals but they were using morality in a broader sense of a non symbolic way. And they make a lot of good points with this is and they go across rules they go across crustaceans, you know, lobsters, etc. How these animals engage in essentially behaviors that suggest norms but also exhibit behaviors that suggest emotions that are under a lot that we typically think as underlying norms and morality. And so the last piece, I'll mention is that morality for for humans is typically is typically symbolic is pretty strongly symbolic. And the work of Shane Gray and others will typically, or have found that in order to identify an action or behavior as being immoral constantly more and immoral behavior will typically need to violate some principle or some code. So, continuing on the idea of what establishes a subject, you know, how do we first identify what an individual is, and what can this tell us about how we manage something like a symbolic self, and how does symbolic self then become to be constituted by, you know, by a belief more morality or what is good for the overall first as the individual. The work of Mark sums who used active inference and the free angel principle to understand the function of emotions. And his one conclusion you had is that emotions is the spot of consciousness. And that should be the direction of consciousness research. So I won't go that far. But one point that he makes that's relevant here is that the phenomenon of emotion is relevant specifically in helping manage categorically distinct needs. So distinguishing between, you know, pain and outside pain and hunger, which one's more important. Do I go to sleep or do I have food on managing these various different subsystems is the function of emotion and can start to help start to help us understand the boundaries of the cell. And so we go back to thinking about how Mike 11 looked at an agent how these things coalesce together, we can see how emotions fits nicely into that picture, because it helps essentially harmonize these various disparate systems by into a person. This gets at hedonic valence. This is kind of pleasure versus unpleasure. So this is an important dimension. There's a few dimensions that researchers have identified with regards to emotions. So how do we study emotions in a rigorous fashion? Valence, of course, is an important measure there. So kind of pleasure versus unpleasure. You can think of it as pain, but typically it's understood specifically as pleasure versus unpleasure. And it's in that sense that complex organisms start to move and help navigate the world and social creatures probably navigate the world. You can see too how it could help us conceive of ourselves as within the symbolic cell. And so some research on the idea as a symbolic self. So what I'm hoping here is that my idea here is that although this was brought up by previous psychologists at earlier times. I think it's clear that, you know, thinking of subjectivity and thinking of symbols in the context of active brain friends and, you know, minimizing uncertainty and a more expansive view of the cell. The frameworks that I list here can become more operationalized. So Carl Jung found that, you know, the self as symbol or the positive band was a motif that existed in numerous cultures from Hinduism to the ancient Greeks to ancient China. There is this constant theme of the symbolic man. And so each person's part from that view, you know, from a religious perspective, from a mythological perspective, everyone was conceived as part of a common person. So this helped establish essentially a common reference frame for behavior. So this is an insimilarly, Victor Frankel home's localist therapy is a Holocaust survivor and also a psychologist who emphasize indeed for humans to find meaning in groups and ultimately find a meeting and lives and our well being is largely determined by that. Yeah, this is kind of broadly our sociological background. So in a sense, we do a number of things that we kind of take for granted, but that are all essentially symbolic, and that all represent our ability to inhabit not just our own identities, but we attach to ourselves with culture roles with cultures that are in some cases extinct in the past, and with features that we can have you have to understand. But I would like to kind of briefly go over the weird journey that I have here. And so, I think this demonstrates how a couple things. And, but analyzing Western sciences so Western science paradigms in the field of economics and language and behaviorism can help bring some of the points of making tonight. And so we had the acronym that I mentioned earlier, this is Joseph Henry, it's Western educated and industrialized rich and democratic. As you can see, so essentially what he found was a lot of scientific studies relied on college students. They're typically college students in wealthy countries and they're of course well educated because they're in college. And this caused a lot of research to have certain biases on that if you start including other other societies that are less weird hunter gatherers for instance, etc. Even different results. So he came up with the acronym, but then he doubled down on it and in a more recent book he explained that weird psychology was the reason for the industrial revolution, the reason for the scientific mindedness that we exhibit. And it's her and so this and we differ. And the consequences of that is we possess an analytical mind. So, you know, think of, you know, a scientific oriented mind. You think of someone who's bookish versus more gestalt thinking which is present and more futile and cast and claim based societies and institutions. So, this has been a lot of thinking had, you know, has led more recently and it was a discussion a couple days ago on using narrative theory and economics. So that's firmly in this message I'm trying to try to overview here. But of course in neoclassical economics is an example of a more analytically oriented but something that is essentially reductive and structuralism and behaviorism fall along the same kind of heels. And so what these three, what these three paradigms have in common is that they attempt to explain a complex phenomenon by reducing uncertainty, although minimizing uncertainty is important within the range of principle. That's also important to understand that it's that it's minimized and that minimization happens, you know, dynamically within systems, and is part of the, you know, adaptive process. And essentially how we get the systems that we do and how they are dynamic how they change over time, how we have preferences, etc. And so one question is, well, if these, you know, if these paradigms are overly reductive and don't do a great job of, you know, explaining the systems that they purport to, then why do we use them? Well, one reason is that they are great null hypotheses. So, you know, as we'll see a lot of economists like to bash on neoclassical economics, a lot of post structuralists like to bash on structuralists. And that kind of goes on and, you know, there's good reasons for that because sometimes when things are bad, they're good at help elucidating important concepts that are true. And the second reason that I bring it up today is that they reveal moral conception at the heart of how humanity deviates, how our social systems deviate from these predictions. And the third reason is that the hunter-gatherers were also analytical. So although they differed from us in important ways, one thing they shared from us was an analytically oriented mind. So that means that the clan-based and the feudal institutions that Henry was talking about, this happened only past the agricultural evolution. So this was 10,000 years ago when most societies essentially shifted their normative structure, became more stationary, and as a result of that, adopted these more gestalt views. So one way to look at our trajectory to symbols was, well, our ancestral past was filled with people trying to do science using symbols or trying to categorize plants and animals, hunting methods or anything that they used for cultural evolution. And most importantly, they tried to classify one another and they tried to type one another. I'll do one example from that, and then I'll move over to discussion. So there's a few areas that I wasn't able to get to, but I'm happy to do that as a follow-up. And then I think most importantly, it would be great to get some questions in. But yes, there's the idea of transcendental signification. So this is one example where the post-structuralist, the philosopher Derrida and Piaget had a similar critique of what was known as structuralism. So there's a background cloudly by Strauss around 1960s, maybe a little bit before that, essentially an ideology known as structuralism, which attempts to view cultures, view language and some other institutions that he examined as corresponding to an underlying and stationary structure. And so it's removing the variability from different cultures, from different myths, and believing that there's a certain structure to this. And Derrida and others' critique of this will essentially withdrawing arbitrary lines and cultures so easily reduced to a specific culture. There's time, there's history, there's kind of cultural variation. And one example is Piaget-Strauss tried to draw a hard line between nature and culture. And really those things are continuous. In a different cultural, in a different natural environment, you might say you'll have a different culture, and the cultural will feed off of that natural environment. And so as Derrida said, there's nothing outside the text, so everything is a matter of perspective in that sense. So the question then becomes, well, if that static structure that cloudly by Strauss purported existed, then do cultures have any structure? And of course Derrida and others still did believe that there is room for structural science. And so the structural science we're looking at today, of course, is the free and principle and active inference. And so what is the kind of core components? What is that? I didn't go over the concept today, but word is that Markov blanket or boundary that we're minimizing uncertainty within. And that's where the transcendental signification idea comes in. And this is where Derrida said, well, there's typically, especially in Western societies, there's typically a concept that you just don't, that you can't argue against. That's just something accepted. And from there that the constant daisy chain of concepts, the constant flow of concepts can stop essentially. And so you think of any, at any given time, it took me a minute to think about what our transcendental signification is today, you know, because there's a lot of fluid concept and there's a lot of arguments. But one example is freedom. So basically, so if someone brings up that in a political context or in a certain other context on that this would fringe upon freedom, but it's very hard to argue against that. And that's because it goes without saying that, you know, although there could be in some context, infringements on freedom, it's generally understood that, you know, freedom is a good thing. It's a positive thing. And the kind of disagreements from it will have to begin with that. So although it's critical of Western society's use of transnational significations and, you know, that's one modern example of course and others are, you know, the belief in God, democracy, other things that are just taken for granted that the culture should abide by. And although he was critical of these are our use of them without transcendental signification is difficult to imagine how a culture would be maintained. I think it's also difficult to imagine how, you know, language would remain stable. That's because there needs to be some agreement on there needs to be essentially some common ground in order for communication to happen. That's because of course language involves adopting a whole bunch of norms, but also our daily lives that are kind of empowered by language and symbols need some sort of stop and point need some sort of around to say, you know, this, this isn't a good thing or this is a good thing. And we're shocking orient ourselves to that. And so this is one representation of why moral conceptions are necessary for our systems. And you can see that in language. And you can see that I think in economics in the field of social trust as well that I can go into that kind of later time. And the last thing is, this relates as somewhere things happen to the motion but I'll stop there actually and you know folks want to do you kind of a follow up we can, we can do that but that's going to turn on see if there's any questions at the moment. Wow, well awesome. A lot of pieces are rising. So for those who are watching live they can write questions in live chat and I wrote down a few things as well so just while we get started. I mean what brought you to bring in active inference. And we'll explore the role in the coming questions but what brought you to this were you studying theories of cognition and looking for what or how did they come to be this way. So I guess technically I took the high road to the difference. So I was studying economics and I thought to myself well we have economics has a behavioral framework. And that might not be great but it's it's related to utility it's related to welfare. It should actually be a pretty good starting point for understanding morality. So essentially, it could be utilitarian framework, except that we have prices so we should be able to clearly at least propose you know a set of policies and set of behaviors or a set of, or an essentially normative theory that you can score behavior against and say whether it's moral or immoral. So that became very hard because it turns out that the neoclassical paradigm doesn't allow that because it it fixes individuals and their and their preferences. And so I then kind of turn to some other moral paradigms that that used actually prisoners dilemma and others that brought in morality. And yeah, then I saw the scientific American or the white article about Carl Friston and him talking about the free Andrew principle. Yeah, and it seemed like okay there's there's an option here there's, you know, a way to, you know, use math like economists love to do, but kind of in a way that still allows for, you know, true subjectivity and kind of true, you know, variation without ever reducing that all the way. I really like this never never even heard it that way like math with a true subjectivity it's truly about the subject and what is subjective to them. A lot of ways to go from there. Okay, you brought up shibboleth and syntax and shibboleth at least in its original biblical style it had to do with how the word was said. So it was truly a question of the morphology of the word it's not like it was an unknown word. It was how it was said, not the underlying term. And then you seem to suggest that we were in a new phase or mode of semantic communication that there would be something passed or different than the shibboleth. So what what was that pointing to or what do you see there. I didn't get to the to the anthropological side. Yeah, so essentially that what I'm referring to here is, I guess he had two things so one is Emily Cohen as a paper called the case for accents. Mark Moffitt has a book called human swarm and he kind of echoes that that same argument there. And the last piece that I mentioned with Catherine Kinzler has a book called how you say it. And they're all about shibboleth. The first two were about making the argument that what we call shibboleth and what I mean, but I'm using it here in the context of it's just accent bias. Right, so that in that case, the, the Fremady is the other or one of them killed the other one because they pronounce shibboleth incorrectly they said it's simple instead of shibboleth I think. And the point is that, you know, from the second where we're in utero, we listen before we see anything. We listen to our conspecifics before we see them. And there's a lot of evidence that we show preferences immediately for people who sound like the people that we heard initially sound. So we established these a kind of orbit around our social circles, using that familiarity and sound. And a lot of racism and a lot of other social bias can actually be deconstructed into relating either more to shibboleth or the accents are actually a bigger driver than we realize and we consciously realize. And so Emily Cohen and Mark Moffitt, they use that as a way to suggest, well, what if we had accents kind of before we had words? And how would that, how would that look? And the idea here is that, well, other animals have vocalizations, dogs bark, chimpanzees make vocalized sounds, vervet monkeys, they do these up and below calls. So they have, in some cases, sophisticated vocalizations. Obviously, parrots have very sophisticated vocalizations only next to us. And so would we, would there be a circumstance where our answers or could have used that to identify with one another in a more cultural evolutionary perspective? So why do we do it as babies? Why do we do it as children? Well, the argument will that probably is indicative of something that's more evolutionary ancient that has more has more legs to it. And the reason why that's important is because we'll be, we started off as an imprimant and we somehow defeated the chimps and the monogamous in terms of resources or whatever. And how did we start to kind of cooperate at scale? And there's been some discussion over that, but introducing shibboleth closes a lot of gaps, for instance. And Joseph Henryx, the story of our success, in his first book on cultural evolution, he goes through how we became a species that learn from one another. And he said he admits that majority of this happened when we live close together. But if you think about it, we must have learned from each other while we were still in these sparsely communities. We transitioned from these sparsely communities to these higher dense environments. But it'd be very difficult to imagine a densely populated area of chimpanzees and mammals because there'd be a lot of fights. So how do we kind of overcome that? And shibboleth is one example by kind of extending our networks. But yeah, it's a long way to answer your question. Interesting. So kind of the historical and developmental primacy of accents and then the kind of linguistic and symbolic signification that we have as a scalable technology. That's why seemingly an open-ended number of people can have a word that they might all pronounce differently. Or a simple abstract, simple like the triangle that you showed that that might mean something different to everybody. And that's where you kind of invoke the platonic concept, which is like the word separate from any accents. Like does the logos have an accent? Zach, Zach, Zach. Yeah, yeah. We tried to, I spent a lot of time on the image because I wouldn't be careful because it's not that there's one light that we're all accessing. It's not that there's one perspective, a true perspective, as far as we know, that we're all that day. The angles accessing when you talk to me and that I'm accessing when I talk to you. We just have kind of, we both have a technically transcendent perspective that sufficiently overlaps to the extent that our communication now is what you might call productive. I do the same thing that we can properly communicate. We're co-constructing that little triangle or blimp. Just kind of, yeah, modification of Michael Evans, I kind of like. Yeah, accent and coding in the production of speech, accent decoding in the recognition of speech. And then that's what allows, for example, token based language models to be in our cognitive ecosystem and get some meanings or allow themself to be projected meaning upon, but cool. Okay, so where does active inference come into this for you? Like in the paper and in the presentation, you mentioned a few places, but you brought in so many historical frameworks and different directions. So just in your own work leading up to here or from where you take it, like what do you see active inference in the free energy principle doing in that line? I see it as operationalizing the Jungian notion of symbol as self. And I think that does a lot and it does a lot because it gets at that foundation, one of the foundational principles which is seeking self. So it's not that we're born and some children aside, I'm not going to speak. I'm not super interested in this or I find that I belong in this family, they give me food and I'm just going to stay here forever. Those kind of things don't happen. And so what it does for me is it frames the motivation for symbolic communication because if it's conceived of as an extension of your identity, if this cultural unit is in some ways more foundational to human identity than even a chimp social unit is for its identity, which I think it is, then it no longer becomes an option. So it kind of accelerates the epistemic ontology that I think the fringe principle underwrites meaning kind of search for self seeking information, information gain. And I think you could, and I didn't, but I think we can rephrase this as morality as information processing. So in that context, in the context of I'm a person, I'm essentially a concept within a broader symbolic social reality. And it starts to, I start to minimize uncertainty by computing cells by finding the blanket that I'm behind. Then morality in that case becomes a signal to what those identities are and where I belong. So in that case, someone was deviant, okay, this person is evil, then that's a big deal. Let me pay attention to this. Someone yelled at me when I crossed the street because I was crossing the street at a wrong time. That's going to stick in my mind because they've pegged me as something. So that was the use of morality as creating something. And so yeah, and so this is one of the reasons morality is I think so loom so large, I think in our minds. And I think it is important for that. But yeah, you also have insight into these concepts. So yeah, if there's a word I should be using, please stop me here. Yes, I think it's very interesting to apply active inference to the Jungian or to the symbolic journey to the search for meaning and all of this. So if we were just talking about like visual perception pixels on the screen or photons on the retina, a sophisticated cognitive entity would do structure learning and propose latent causes. Like, oh, maybe it's not just a million red pixels moving from left to right. Maybe it's a red car. And so then there's essential or the symbolic car. So that would be like structure learning and world model again also for the kind of entities that we are. It's like a self in world model because one of the most important things is agency and understanding. Well, what are the movements that I have agency over, which is why we're able to kind of ignore isocades, because we know or suppress proprioception associated with our body being in the position that we expect it to be, because those bodily features are under our control. And so we're able to basically calibrate them out, not need to be distracted or attend to them. And then that allows us to be sensitized to changes that are not caused by us. And so it's like, what is the eye? What is this symbolic and embodied strange loop that's causing all my thoughts, my thoughts. And so it's like that can be framed very clearly in a uncertainty reduction, structure learning, external state or hidden state framing. And only through kind of a lot of modification and coercion, can that be cast in like a micro economic reward maximization framework, which just leaves more to be done than before the application. And ACTIMF is like, well, no, this is at least a uncertainty reducing way to approach that question of self and potentially multi scale self on discovery. Yeah, what are the points I was going to make with neoclassical dynamics like similar to what Derek did in structuralism was, well, because this is such a, this is a general equilibrium kind of, I think of it as like a thin framework, meaning it reduces the economy down to essentially get no movement because all the resources are allocated and there's really nothing left to do. But the, what's interesting is when people identify where that fails so neoclassical market failures, it's the most interesting side of neoclassical economics, it's how you get pollution, it's how you get technical spillovers that have, you know, positive reaction. I think the interesting one is in the concept of cultural niches, social trust. And so there that, and also the, and how firms coasting how firms don't exist. So you have social trust at the level of level of a country a lot of social trust at the level of the firm. And within that that minimizes uncertainty for the employees or it minimizes uncertainty for the consumers because they know the product they're getting. And I think we can also think of it as minimizing uncertainty for you think of someone who's 12 and they're not in the marketplace yet, they're just learning things. How do they participate? How can we think of them participating in the market economy other than just a kind of a long time consumer until they start participating? Well, they have a series of symbols and they have a series of ways of reducing their uncertainty. And so this is, you know, this is universities, this is, you know, companies that maybe not thinking in companies per se, but they're thinking of, you know, a business person or a scientist or, you know, they have these similar things that can be used by their hotel. Another question was you described this kind of perspective swap that we engage in and you described it as there was a perpendicular component to what, what was that social setting that was perpendicular and what was perpendicular to what? Yeah, and so here, you know, so here, I guess I'm kind of on the edge of my non knowledge, but what I was, the perpendicular would be so typical primary relations but also, but also a lot of our social relations are intimate. And so that's perpendicular meaning in the sense of, you know, for these are the people that you, that you, you know, these are interactions that you have that are kind of based on your subjective perceptions and meaning those, meaning those interactions that we can kind of carve out of a symbolic reaction. So for, I guess for people, it's difficult to do because as soon as you engage with someone, you know, you're thinking about what you should be doing and, you know, there's a bunch of unconscious things of, okay, I should, you know, I should shake their hand or you know, I should, you know, if I'm inside, I should kiss their cheek or, you know, I should take my shoes off now so there's obviously kind of a top down implicit symbolic occurrence is happening. However, you know, for non human animals for, for primates, you know, for dodging, etc. That world is, you know, it does have, it does have history. You know, it does have, it does have memory, but the perspective is still, is still subjective. It is still a, you know, a subjective memory. There's no kind of above the fray objective top down sense for which they can coordinate their, their actions and behaviors and their kind of goals. And so I, on one hand, I don't think it's saying too much because I think it's basically just describing. You know, it's kind of visualizing how symbols are different than symbolic interactions are different than non symbolic interactions. On the other hand, I think it's interesting because I think it can help, it can help us imagine the context and maybe model the context for how we first started it. We have, we first started to do it, which is that we were in an environment where both environment, both in a, in an actual environment like in a, in the place, but also our kind of cognition was at a place where we could start to imagine our behavior as if a third person were looking at it. And so that, yeah, that's essentially what I was getting at. Cool. All right, I'm going to go to some questions from live chat. So if you have any thoughts on them, please feel free. Otherwise, of course, however you want to see them. Okay. Upcycle Club wrote, How can tokenization improve the performance of active inference agents that use tensor networks as generative models for sequential data? And then just follow it up by defining tokenization is the process of breaking down a text into smaller units called tokens, which can be words, sub words, characters or symbols. Do you have any thoughts on this? Repeat the question. Yeah. How can tokenization improve the performance of active inference agents that use tensor networks as generative models for sequential data? I was just interested in the question itself, but yeah, I have no, I have no thoughts, but other than I can't wait to, you know, to examine the wording of that question and maybe have an opinion on that. Yeah. I appreciate whoever asked that. Whoever asked that, I think that I would have insights into that question. Indeed. It's on stream. We'll come back to it. Okay. Dave asks a question. There's a few background pieces, but it will end with a question mark from what I see. Okay. Whitehead in process and reality presents a framework that allows a kind of chaining across local spatial relations that allows a group of observers to attribute objective locations to objects in a shared space. A group of minds can apparently share concepts somewhat contrary to what we'd expect from a strict avoidance of the fallacy of misplaced concreteness. It'd be nice to speak meaningfully of shared concepts while avoiding the fallacy of misplaced concreteness and also avoid Whitehead's rather unsatisfactory reliance on eternal objects concepts residing in the mind of God. Could the trick be made to work by positing chaining across locally I and Val first and second person mappings where you and I take into account the partial discrepancies between what you and I mean by a given term or a given illusion while we retain and reuse the structural commonalities between your and my usage of the term? Yeah, I mean, this is certainly much more in my wheelhouse than the first one. And yeah, I mean, I start off my paper talking about Whitehead, so I certainly deserve a question that involves it. I think precisely, and I think that that's kind of getting at what, you know, the way I present our discovery in language where approaching the idea of, you know, the fallacy of misplaced concreteness, like we believe in these platonic forms, it's this western bias, it's this thing that Plato did that was wrong, that we continue to do with, you know, class of economics, but even though it's wrong, it's, it motivates us to create language because by believing something, by believing in these forms in a sense, we end up reifying we end up creating things that do exist. So I guess the way I see it first of all is that the myth of objectivity is that we believe in these platonic forms, it's this western bias is this thing that Plato did that was wrong that we continue to do with, you know, classical economics, but, but even though it's wrong, it's, it motivates us to create language because by believing something, we end up reifying we end up creating things that do exist. So I guess the way I see it first of all is that the myth of objectivity is that will by wanting to believe that there is a right way to speak by wanting to believe that there's a right thing. There's a right thing I mean by the right thing. We end up kind of bringing that about. So where I'm curious kind of parse his question out is at the tail end where he talks about, you know, the difference, the distance between I guess the two subjects I knew, and that kind of how we kind of inch closer together. Well, if if both me and you believe that, you know, we're speaking more or less the same language, you know, we're both talking about active inference for like even a bit at lower levels, then we can navigate a closer and closer space and establish something like a reference where words have common meanings, etc. We can establish some common ground. And I think yeah so I think on one hand you can think of it as this, if I understood correctly, towards the end, we can think of it as this kind of closing the gap between our two perspectives, but I think you also have to. It's also necessary that we're both conceiving of this. You know, I don't say God like because we're trying to avoid what has use of God like but God like is useful because if we just think of that there's a, a that there's this fallacy and there's this perspective that can see the true form. So we in the sense we need this kind of mythology I think, or we at least needed in the past to kind of get at something like language. Awesome. I'll add a few, a few notes on that and ask another question. So in thinking about these appeals or abductive hypotheses about hidden causes, it's like, wow, what is causing all of these horses? I mean there's small ones and big ones and they have different patterns and some have four legs, some have three legs, one, you know, had this other number of legs one time. So it's like what could be causing all of these sounds, sights, smells, and there's basically two extreme structural hypotheses. One of them is like the single, like there's the, that's the platonic horse. It's like, I don't even know what it is. It's very different than any of the horses, but it's just somehow is the structural cause. It's the formal cause of all these horses. The other extreme structural hypothesis is like horses don't exist. Everything causes, like there's just sort of like an infinite component model. Now complexity is always found in between the one and the infinite. And so that's where we actually do develop more well articulated accounts and what really does cause a horse to be a certain way in its accent, in its real embodiment. But we can see that kind of modest or absolutist appeal as being like, as you put a good null hypothesis. Like potentially one of the best starting null hypotheses, not going too deep into value judgments. But I mean, why not just start with saying maybe there's one thing and then we'll keep adding latent causes as we go. But then to confuse the appeal to a single cause with the actuality of a single cause is kind of this challenging dance. Because if you take it as real, then it's real as real. And then to hold it at a distance though, but it's hard to hold something at a distance when life and everything is on the line. And then one other piece, what you said about like, well, the God like appeals of some of these like white heads, formal approaches. And you seem to express somewhat of an ambivalence about this kind of an appeal. And that reminded me of the quote from Hamlet. What a piece of work is a man, how noble in reason, how infinite in faculty, in form and moving, how express and admirable in action, how like an angel in apprehension, how like a God. Action and inference in Hamlet, early active days. Okay, I'll ask a question from the chat. Muse equal wrote, isn't this indicative of learning and intuition being entropic post observation. We disassemble the observed construct into segment frames of historical time through interaction, prior knowledge and memory. Where, where is learning? Where is symbolic learning? In what ways is symbolic learning anticipatory real time or retrospective? I guess this kind of gets over there. You know what you were saying with forms is kind of causes of hidden states and kind of how do you, you know, update, you know, update beliefs, like once you have the beliefs, like how complicated is it to, you know, to actually update them. Trying to think of an example that we went through. But I, yeah, I mean, it's a, and now I'm just thinking of my daughter because she's learning how to speak. Right. So she's clearly symbolic learning. And, you know, kids are much more comfortable with variation, you know, with dentistry. I'm not comfortable, but they exist much more entry than than we are. So they can they can navigate the space I think probably better than we can conceive that they that we can, or that even one can because they're in it so often. And how do you kind of build symbols? Well, I guess. Okay, so where I came from from symbolic learning was Hofstetter and Sandeir's book on analogies for getting services in essence. And so their claim is that, which I think aligns pretty well with active inference, but I'm curious to hear more feedback on it. And then their claim is that we learned primarily through metaphor that enemy. And so it's through, I think through analogy, they said so. They look at you look at surface differences rather than essential differences and essential would be more formal tonic form. Look at more of circle differences. And so, you know, my, you know, my dad is to me there's a dad towards someone else. You know, I'm crossing the street so I suppose learning, I'm trying to get some good examples but, you know, we build, we build structures of symbols. We're using kind of like against like so, you know, this is like, you know, this is an action of sitting in one chair. I sat, I saw someone sitting on a bench outside. You know, I saw someone sitting on the ground. So it's the producing I guess, if you a generative to model of, you know, in this context in this situation, this is happening. These are the, you know, these are the relevant features of the environment. And this is then the object within that environment. Here's something going on here, and I'm bringing in, and I'm using that as a prior to establish something. And then at some point, and at some point we start getting concepts and then once you have your initial concept because that takes some time. Once you have your initial concept, then you compare concepts to concepts in this kind of metaphorical fashion. I didn't give a good example of metaphor. We start to move outward that way. But yeah, I guess that's the best that's the best I have for you at the moment. Yeah, this role of metaphor is very interesting and something lightly speculated upon but not quite brought into active inference generative modeling. Yet, but metaphors like, well, it's like the highest point on the building, or it's like the center of the circle, or it's like what's inside or outside of the circle. So those would, those kinds of metaphors which can be bootstrapped or grounded in in the immediacy of our developmental and evolutionary history can then be brought into. Purely cognitive or symbolic spaces. So we could talk about a clash of symbols or about friction or inertia and grasping all of these. Tangible. Examples that are kind of out there, but then we bring them in here. And I think that may have something to do with what Dave was discussing with the eye and vow. So it's not, that's kind of the Martin Buber. It's not just I it. If we were talking to an inanimate object or an object that we didn't think had a self in world model, then it'd be like ranting to, you know, this puppet, or just like kind of just but but under no expectation that it has a mind of its own. But then the eye, vow, or inactive inference, what's sometimes called like thinking through other minds or thinking with other minds, where it's kind of like, well, I'm modeling my own speech in relation implicitly or explicitly to what they'll respond to. Right. And knowing that they have that kind of recursive activity as well. That is purely cognitive. It's lifted from the Accenture. It's brought into the structural and the symbolic. But it's always realized by systems with accents. And I think to just the way that you introduced and highlighted that more direct experience of like how somebody is saying that, not what they say. It's easy for us now to focus on the semantics of the what, which are so overwhelming. But that almost obscures the kind of bootloader that actually got us to the developed social systems, which as you pointed out, clearly some things have changed such that we don't require that direct intimacy. Yeah. And so that that's a complex tangle of cognitive and cultural phenomena. It's not like only being ascribed to the ability to do metaphor or think through other minds or use symbolic logic and learning. But all of these are coming together in this kind of unified mystery. Yeah. Yeah. Maybe I understand that earlier point. So the first an article on singing sing on from the same him sheet. Or you think of it as, you know, with language you're engaging in these. Fast-paced dances with people, you know, exchanging, you know, exchanging beliefs in anticipation of, you know, you're into, you know, interlocutors beliefs, and you're constructing something. Yeah. The only thing I want to say is when you asked earlier where I felt active inference was useful, I guess in nutshell, it can, it can better understand the dynamics between cooperation and competition, you know, outside and inside. And yeah, that gets us some attention with how we think of communication, whether it's cooperative or whether it's kind of just ecological. But it encourages a, it encourages that dynamic where we're in, like if, you know, if we're saying something and maybe I'll, maybe I'll disagree slightly with me on a point. But we're so we're still here is trying to be cooperative. So we're trying to say things that we both understand. And sharing that kind of same cultural musketeer attempting to, and at the same time, proposing either iteratives or modifications of those beliefs, so that I'm able to persuade you of my point. When we get out whether communication is kind of cooperative, to take some of this cooperative thing to its dance versus it's kind of contentious. Then you can start to tease out those, you know, those those factors. Okay, short and interesting question in live chat upcycle club writes. What do you think about petroglyphs? Petroglyphs are rock carvings made by pecking directly on rock surfaces using a stone chisel and a hammerstone. What do you think about rock inscription? Rock inscription. I'm a big fan of why I imagine what you would be getting at is, is that so is that symbolic, you know, probably if human does it if maybe he's getting it or she's getting it if, you know, if a bird has done it, is that symbolic. Yeah, I don't know. I'm curious, but yeah, I'm a big fan of sculpture. That kind of reminds me of the Osamandias poem, like, I was here, and all of that entails like there was an empire here. Now, of course, in the poem, it's all degraded and decayed. And so that's kind of like the media and the message. But the inscription on the stone is like, I was here, I chose to do this, or even if you don't, sorry, what? It's almost like graffiti or something. Like, like, why, why are we? Why are we doing this? Why are we? Okay, I get it. Yeah. Somebody who uses that graffiti tag did that. We could have different stories. They could have different stories about whether they have free will or all these different like secondary things. But it's like the fact of the matter is on the rock, it was inscribed. So I say, yes, I mean, I get I get that now. So what I see there is, you know, what I see there is that is, you know, that's soul, right? That's in the sense of, you know, something like Whitman would describe it is that that is we do that because we have soul. I mean, not that in a not necessarily a religious context. But as in you can exist, you know, Daniel Freeman can exist as a conceptual identity outside of space and time. And the way you do that is that you leave a mark, you know, you leave your legacy and, you know, you know, et cetera, et cetera. But yeah, so we exist as these symbolic selves, both as part of these cultural units, but in some sense outside of it. And so, yeah, do you write a name on a wall? Do you, you know, do you start a company? You know, do you want to be famous? Do you want to be on a movie for whatever reason? Yeah, this is that that is an extension of identity. Yeah, so if that's what that's getting at, then yeah. Well, in sort of closing, where are you going to take this line after the frontiers paper? And this what's the next frontier or the next journey? Also, yeah, so that that I don't know. So yeah, I'm open to ideas. But yeah, I was I was obsessed with this idea for a while. And I think I've nearly gotten it. I think I'm kind of like 95% the way there or maybe 90%. And I think this discussion has helped me. But yeah, I'm just kind of curious to see where it might go. But yeah, I mean, I'd love to see it formalized. Love to see taken seriously. But yeah, I have no specific, no specific thoughts. But cool. It was great chatting with you that I appreciate the setting up and appreciate the dialogue. Thank you. I really appreciate hearing the comprehensivity and just the relatability of the curiosity and how many things are brought together in your paper and in this discussion. So see you around till next time. Take care. Thanks everybody.