 Hello, everyone. Welcome to Act In Flab, live stream number 33.0. It's November 18, 2021. Welcome to the Act In Flab, everyone. We are a participatory online lab that is communicating and mute if you're not going to be speaking. Thank you. We're a participatory online lab that is communicating, learning, and practicing applied active inference. You can find us at the links here on this slide. This is a recorded and an archived live stream. So please provide us all with feedback so that we can improve our work. All backgrounds and perspectives are welcome here. We'll be following good video etiquette for live streams and respecting our awesome facilitator blue. At this short link, you can find the streams past and upcoming, and we are about to be heading into the second half of November 2021, where on the 23rd and the 30th, we'll have two participatory group discussions on this very paper we're going to discuss today, thinking like a state. And then we have yet to set the December paper. So if you know someone who might be interested in joining or you yourself would like to discuss a paper of yours, get in touch with us. Today in active stream 33.0, the goal is to learn and discuss this paper, thinking like a state embodied intelligence in the deep history of our collective minds by Avel Gwynin Colou and the video like all dot zero videos. Even though we do have the author here, it's just an introduction to some of the ideas and how to pronounce them, not a review or a final word. So we'll introduce ourselves and then go into aims and claims, abstract and roadmap, keywords, figures and formalisms, et cetera. So let us know, especially if you'd like to participate in the upcoming 33.1 and 33.2. We'll start with just an introduction of saying hello and what we're excited to talk about or resolve either in the dot zero or in the upcoming group discussions. And then we'll culminate with the author and transfer facilitation over. So I'm Daniel. I'm a researcher in California. And I'm just excited to check in on the progress of this important topic. People have been theorizing about societies and cities and states and everything in between since as long as they were there, probably a little bit before. So it's just a fun topic. And I'm really excited to see how active inference can give a new lens at resolving some of the longstanding questions in that area. I'll pass it to Steven. Yeah, thank you, Daniel. I'm Steven. I'm based in Toronto. And I'm really interested in the multi scale implications of state and niche construction, particularly for me, because I'm interested in working around, like all social topographies, which is probably at the scale of events and moments in people's like lived experience and in their architectures and places. So I'm really excited that this is bringing in this idea of state and larger scale, but also bridging that with active inference from the kind of faster, smaller scale. So yeah, lots of good stuff here, lots of ontological questions as well, which I think is really cool. And I'll pass it over. Should I pass it over to Blue or should I? Yeah, over to you. So I'm Blue. I'm an independent research consultant in New Mexico. And I really like this paper. And I really like working and talking with Avel about ideas, because these ideas are kind of at a bigger scale. I am interested in self assembly of things like at the cellular and maybe even subcellular level. And like this takes it to the very big level. And I think that active inference really kind of unifies this. It's a skill free framework. And so that's kind of what I'm excited to talk about today and maybe not resolve, but discuss and get some other perspectives on. And I'll pass it to the first author of the paper or the only author of the paper, Avel. You'd like to just introduce yourself, say a few words. So I'm Avel, also known as Servo. I have a training in cognitive science and physics, mainly interested in philosophy and a lot of things. What we want to hear is that I founded and help organize Kairos research, participative citizens laboratory, which aims to understand how to act out to navigate conspiracies and show historical dynamics. And part of this is to understand why we have states and how those states that we now have are like. So the goal of the paper is to basically understand how states have constituted and what it tells about their continued structural identity through ages. And this, of course, the topic is big, there are several kind of states. But the very first thing we need to do is to understand what your state is and where it comes from. And that's what I'm trying to do there. Awesome. So moving on to the aims and claims of the paper. So this paper thinking like a state embodied intelligence in the deep history of our collective mind. In the paper, it states that the core argument of this article is that states can be and should be understood as hierarchical control systems, essentially similar in their core physical architecture to brains. And we will leverage active inference to ground the study of intelligence in the way an organism maintains their structural autonomy to formally assess the collective brain hypothesis. And the big question of the article is what kind of intelligence is displayed by city state systems? And how is it embodied in their structure? So we'll go through the abstract Daniel, do you want to take the first section of the abstract? Yes, Steven, first. Oh, just one comment just because I know you mentioned kairos there. And I've done an activity with the kairos blanket. It's like a mapping thing where you stand on parts of Canada and to look where it came from. I don't have that connects. So if I'm writing you kairos is working at that kind of regional, national scale to look at questions around that. And I'm why I hear that's informing how your research is being approached. Is that correct? I did not understand the question. kairos blanket, correct. Yeah, it's an activity. It's an activity. Actually, not to do the Friston blanket literally where they put a big blanket around parts of Canada on the floor in a room. And then people have to stand on it. And they keep taking away bits, which were basically, which were not no longer in the territories of the First Nations. And then you see how much is left. There's not a lot left, so to speak. So it's talking to that idea of statehood and settlers and how much is left. And I wonder if that speaks to the wider work of kairos and questions of states and how do we understand how we fit into that? Or maybe I'm miss misunderstanding. So that fits in the work of kairos because it tries to bridge scales. I never heard of this kairos blanket. This is not related to what we're doing. I mean, it helps embody impression of how well settlers have appropriated indigenous American territories. So this is something that's cool that fits the mandate of kairos. But the core idea of kairos is it's the Greek notion for timing of three in a Greek. And kairos is the instance and it's quoted actively like it's an instant where you can seize opportunity. And this is basically what embodied agents do they build skills and they build capabilities and they enact them when it's appropriate. This is the goal of kairos to understand and reproduce this core ability and a lot of people have used kairos you have anarchists that claim that are called kairos you have a Christian concept that is called kairos you have a lot of people that make the use of the concept because it's very broad and it's very it's not something that has been imported in the Christian concept of time really in like what we have today. So if you want to talk about this you use your gestures the great word which is kairos so Yep, just one thought then I'll go to the abstract which is that kairos as opposed to chronos like a chronometer with decimal time kairos is like timeliness and it brings action and sort of a little bit of a moment that transcends the tick tick tick tick tick. So it's a cool term. I'll read the first third of the abstract. This article aims to show how the deep history of early state societies entails the development of a collective form of cognitive agency. It relates classical works in the anthropology of states, in particular, Scots seeing like a state with the inactive account of biological and cognitive organization. Thanks to the unified ontology for self organization dynamics across scales offered by the active inference framework. Stephen, do you want to take the next piece of the abstract? Yes. Okay. So active inference conceives of cognition, a synchronization across individuated sensory motor states, entails that biological or socio cultural constraints display a minimal form of cognition by shaping the behavior of faster dynamics in a certain way. When such constraints collectively define a basic life form, an integrated operationally closed system, they can therefore be said to embody adaptive knowledge, properly speaking. And then I'll just take the last part. It says the inactive inference account I articulate here strongly motivates and methodologically grounds a hoist, a wholeness approach in the social sciences. Indeed, it grounds a study of human societies in the role of structural constraints, whose meaning depends both on the broader systems activity and in the historical context of their emergence. The present account of the dynamics of early urbanization and state genesis aims to illustrate this approach. And a quick just walk through the roadmap of the paper, we're going to go through the introduction section. And then there are three main sections of the article that embodied intelligence in nested minds is the first. The second is the deep history of cities and states. The third from the urban metabolism to the collective mind. And then we'll go through the conclusion. And first up in this dot zero contextualization video, we're going to discuss the keywords, which I think there were no official keywords, but these are the words that I created for the paper when discussing it at the conference. So we pulled out intelligence, autonomy, embodied intelligence, active inference, the free energy principle, the concept of thinking through other minds. And we've done that in a stream before Daniel, is that true? I think we have right. And then niche construction. So I don't know I don't have the facts on the on the exact number of the thinking through other minds stream, but we can find that out somewhere during this during this talk, we'll look it up. So about you want to maybe speak to we'll give you the first stab at talking about what you mean by intelligence and autonomy. We've pulled a couple quotes out of the paper, but maybe you'd like to say a few words. So if I remember well, intelligence, etymology, specifically, it's the capacity to make links to connect dots of data and to be able to understand it. I don't want to commit to any specific notion of intelligence, because I'm not aware of any meaningful first principle definition of it. There is a lot of work in measuring intelligence like IQ, which maps to specific communities and has literally zero reason to generalize to other life forms than humans. And we could have you to other life forms than humans in occidental societies. The notion of autonomy is much more specific. It's important to okay, so etymologically, one's ability to build ones own rules. So for example, if a city state has laws and the organs that make the law are within the city state, it is autonomous in a stricter sense. It is also recruited in the study of life and cognition and the way cognition is grounded in life by inactivist scholars, where it is related to a personal closure and the precarious circumstances. So it is something that is able to operate and able to recreate the conditions in which in which it operates, while there is no exogenous force that help him do so. For example, if you have a candle, it helps. There is burning fire, which generates a flow of oxygen so as to maintain itself. But if someone does not keep fuel going, this dies. So this has no really way to recreate the conditions in which it operates. This is in contrast to the simplest bacteria there is, which is able to look for food practically. So the active concept seeks to capture that. Nice. Daniel, do you maybe want to read the quotes from the paper directly referring to intelligence and autonomy as described? Her intelligence is the general ability for a system to understand the world either in a reflexive propositional way or as enacted in their ecological activity. And autonomy refers to the property of operational closure under precarious circumstance, i.e. the capability of biological systems to continuously maintain and recreate their own structural identity as a result of the collective activity of their constitutive processes. Right. And these are processes that are actively generating and sustaining identity. And I think Avel spoke really well to operational closure in the what you just described. But we just want to refer anybody who's listening to this paper by Montevil and Masio in 2015, this biological organization as a closure of constraints, as closure of constraints. So I want to emphasize that there is, I think it's pretty old, the approach of publishing. It has at least 10 years and the Montevil paper is an attempt to formalize it in logics. So it is not necessarily the reference on the side of what is the reference paper on this. Awesome. Thank you. Maybe let's move on to embodied intelligence. And Steven, if you this is perfect for you, you want to read the quote from the paper describing what embodied intelligence refers to? Yes. Okay. Embodied intelligence refers to the ability for understanding adaptation, sort of a mixture of both, that is imprinted in an agent's physical structure rather than in some general computation of ability. Yeah, that's significant. And then we have the woodlice alias. Yeah, don't use the gist of FYI. You can use Keybase if you want a back channel. But that's the woodlice example that Carl Firsten provided in the 2018 example. That's a little shout out there. Awesome. Yeah, so the does anybody want to say a few more words about embodied intelligence, maybe ML or Steven, or if there's something that comes up for you guys, feel free to raise your hand. If there's something else you want to discuss. And the slides are the slides that we prepared for the talk. They're in the Keybase chat as well. So I'll mention just to say, I mean, because we're kind of introducing it at this stage, I won't start raising those and those are questions and we'll save that for the next one. But I just note that this is useful just to highlight how this is being structured because you've got this word intelligence and embodiment and I know there's talk about the brain or the idea of the brain and the idea of the mind. So I think it's interesting here that this paper sort of bridges a bit across the brain, intelligence, mind, culture, kind of area. So when it's talking here, it's kind of seems quite an open description and is really making sure it's clear to speak to the difference to this general computational ability, and that it's somehow a different type of knowing that's being done, maybe with the word, you know, morphological computing sometimes is used in that in that world. Nice, thank you. Maybe Daniel, do you want to explain the free energy principle and active inference? Sure. So as on the slide here, the free energy principle states that dynamical systems, given sufficiently regular boundary conditions, which are required for their continued existence, spontaneously self organize, so as to minimize their expected variational free energy. Expected variational free energy through time, effectively maximizing evidence for the world model they enact. So the expected variational free energy, which we talk about in more detail elsewhere, it's kind of like all aspects of the system through time are what have to be handled in order for persistence. And using some technical formalisms, that turns out to have equivalence with evidence for a generative model, where one is persisting. Active inference explicitly brings action into that inference process. And it's the process by which dynamical systems maximize their evidence for the model of the world, which they enact. This can occur by a system altering its environment to fit expectations, ie niche modification. So it's the difference between the free energy principle, merely saying that persistence systems are going to be isomorphic with those that have high evidence for the world model, which they enact, and then active inference as a corollary process theory says, and the way that they're going to do that is through inference, updating their internal model, or through action, which can be just sort of like a flock of birds action in the sky, or it can be niche modification like Stigmergy. Nice. Thank you. And then we can go on to thinking, Oh, Steven, you have a comment to say? Yeah, just one piece here also sort of flags up for me is the use of the word model, because we talk about their own model of the world. And yet, often we think of models as something that we've created in the world, you know, on a piece of paper and things like that. So again, this is sort of ontological challenge in the sense that it's it's kind of like a self model, but it's it's a self model that's not a representation. So our language is kind of couched in our history. So anyway, I think that might be something that sort of resurfaces and have to kind of keep restating because people might have this representational kind of inbuilt way of, you know, those metaphors are quite embedded. So it's hard for people to necessarily switch to what it means to move away from representation, representation approaches. I have one question as well. The reason why we're talking about active inference and free energy principle in this paper. So, Aval, what does active do for this paper? Are there other contender theories in the game? How else are people trying to take a unified approach to studying the specifics and the generalities of states? So the approach I have, I define as an active inference with the N between parenthesis, it's a synthesis between what is said by an activist and what is said by an experienced theorist. The core framework is an active, I'm interested in characterizing states, state societies as a network of constraints. So as slow processes that are exist at a holistic scale, not within individual, not between individual, but in some sense, over individual, and that shapes the activity of individuals, so as to maintain self. That's the core argument, so state is our organism. But now if I want to get there, I need some strong and says, I, if I need to get there, even if I need to formalize the role of states which are at core information processing machines in its, I need to formalize in a stronger sense the relation between basic synchronization and epistemic states. And that is precisely what active inference does. So I need the two, I need elective formalism, the elective approach, and I need active inference formalism to make the link between the ecological activity of states as an inferential process, and the body, the constraints that actually consist, that actually state is consistent. Nice. Thank you. We can, I guess, move on to thinking through other minds. Steven, did you have a comment? Just one other comment as I was interested by the, you know, using the term constraints. And in active inference, it sometimes thought of as the higher level processes are in some ways constraining the behavior of lower level processes rather than necessarily dictating them. So they're, you know, it's a little like, I'm doing the faster jig. And someone's just slowing me down a little bit. And so on. So, yeah, just curious that that word constraints has brought in, so often people think of, you know, intelligence is all about doing things and, you know, not constraining things. So anyway, just curious about that. Yep, that's related to synergetics, Hakan, not Bucky, and about like a wave in a ripple, like the wave constraints where the ripple can go, the faster fluctuations are constrained by the slower ordering parameters. So here we have thinking through other minds. And this is another body of work, which we won't spend a ton of time on but for organisms that are making generative models of their niche. Sometimes that niche contains other organisms such as theirself. And it matters how those other organisms also include generative models of oneself. So thinking through other minds is describing how in a shared epistemic niche shared informational niche with shared salience cues, for example, that two different agents are able to coordinate because they are thinking through each other's minds. Like if I put up my hand, then blue will think that I want to speak. So it's sort of like a little bit of a theory of mind, but with more of a act in flavor. So both like the idea of constraint and this concept of thinking through other minds, bring me back to Mike Levin's paper, The Computational Boundary of a Self, because of the nested like cognitive cones, right. And when he kind of talked about it, he talked about, you know, the higher order or like the larger level, the higher level organism, or collective constrains the behavior of like the individual, like sub level collectives. So that was kind of this idea of constraint and then also, like thinking through other minds, like specifically, we talked about how connections between cells, individuals, but but connections that that essentially breed information sharing. So when you have like a shared informational niche, that enables this this process kind of of thinking through other minds, where you think more as a collective. And so that expands the cognitive boundary of the self to then include other, you know, individuals within that cognitive boundary. Steven? Yeah, and that ties into the work of Andy Clark and the extended mind. So there's a so the way that that gets there's a there's there's a bit of talk about how do you think about that extension happening and how much is it an extension of the individual and how much I suppose is what's been talked about here is how much does what has been extended start to become its own sort of cognitive agent? Nice. And maybe very appropriately, of El has returned and we are moving on to talk about niche construction. And I think of El might have a lot to say about the niche and its role in extending the cognitive boundary. So maybe do you want to give us a few a few insights into can we help us think through your mind? Modify the audio. Modify the audio visual niche of our joint attention. Anyway, I can't change your model of the niche. This is not what I'm going to be. I think I'm not trying to that activity, at least, whatever. So the there is a body of work in biology, evolutionary biology, which is centered on how organisms modify their direct environments in a way that somehow fits their, their ecological needs. So a typical example is how beavers will be dumb. So as to get I honestly don't know what beavers get out of them, but they get something. And the basic arguments in the thinking through other minds framework is that the way you modify a niche has a cognitive value. It's externalize expectations. And by accelerating expectations, it allows expectations to synchronize between individuals. And it's, there is a very clear analogy or identity between once you accept there is a cognitive value in niche construction as the externalization of expectations. There is a very clear analogy and even identity between niche construction and excellent clinicians. Beaver are just extending their permission, their expectations, the domain of the world that somehow participates to the fact that they live, they keep on living and understanding the world. So there is a pretty basic way in which an organism can expand the extent of the world, the domain, which is operation closed and which participates actively in their maintenance. And this is basically what there is to be found in niche construction by shaping their niche organism shapes, their world, they shape their condition, they shape their body. And I'm not sure I'm trying to give a general insight in niche construction. I don't really like it, but I'm not sure how we're going to do this because I'm not sure what is the point that is to be made the following way. Just that niche construction is an important mechanism of extended cognition, like the recent Stigmergy models show that individuals don't need to carry on board or internally the entire cognitive apparatus. They only need to know how to interact and potentially modify a niche. And then also, I think this is important when we're talking about cities and so on and states, because our niche can, as per thinking through other minds, include others like us or distant from us, but also it includes the abiotic niche, so climate change and resource utilization. Those are niche modifications. There are some radioactive mountains that have been niche modified. And so it provides a continuity between the kinds of ecological changes that humans make with the cognitive approach to ecosystems. And so that's why it's a very rich framework to build off of. So I just want to read this quote from the paper really quick. And then, and then Stephen will come to your comment. So it says, as stated previously, the network of constraints defining the structural identity of basic life needs not be printed in flesh. It only needs to influence faster processes so as to continually recreate itself. And the references for this are there on the slides if you care to probe further. It says, such constraints can be embedded in the informationally rich socio-cultural niche, which enable the human ecology by affording adaptive coordination and the development of complex skills. And I think one of the main things that this paper does, and I'll feel free to correct me if I'm wrong, but it kind of takes the idea of cumulative cultural evolution and brings the niche into this as a cognitive extension. So I think that's one of the key moves made here. Basically, I know cultural evolution. I have known cultural evolution for like five years or so, acting in France for one year only. So cultural relationships, I see things. And I'm currently writing a paper on that could be interpreted on the reductionism in cultural evolution. And the fact is cumulative cultural evolution in cultural evolution is the replacement of flow fitness variants by high fitness variants within variants, I mean of cultural tokens that are implemented within minds or behavior. Because there is not a strong ontology on what the replicator is in cultural evolution. And a basic insight I want to convey is that culture does not stop in the brain. Culture certainly does not stop in individuated cultural tokens. And one of the goals of the paper is to show that social culture evolution, it has a knowledge extension or intention, I don't know. It exists not only within the mind of individual and within the burial effects of cultural tokens. It exists instructors that are not apprehensible by any individual and that are not entailed by any specific cultural token. Thanks, Stephen. Also bringing that beaver example is to some extent the beaver constructs a dam and to some extent maybe it grows a dam. Maybe in the same, you know, so by adding pieces it's growing something a bit. And at what point does it sit, stand back and say, oh that move that one a bit to the left, move that a little bit to the right and sort of improve the construction and how much is it an evolving process? And I suppose there's also that idea that once something has grown and it's changed the niche, something else, you know, just like the brain effectively has to grow before it can do its thinking. So maybe there's something about how these structures have grown and been constructed. So I don't know, I just mentioned that. Awesome, thank you. I guess maybe we kind of got ahead of ourselves talking about niche modification and cultural evolution. I love this picture because it really shows like kind of how much we have offloaded our cognitive selves into the niche environment. Like we don't need to know how to use directions or what restaurants are good or we don't need to know actually any of that anymore because it's all like riding around with us in a smartphone in our pocket all the time. So we've just put our brain, we've offloaded all of the cognitive capacity of our brain or at least a good portion of it. So what do we think about now? It makes me think about that always. It says in the paper, we claim that niche modification is critical for cumulative cultural evolution and advancements such as writing allow active inference to occur between agents and stigma accuse embedded in the niche. Avel? Did I say what you did I write? What you just said? Okay, what I wanted to note is that what you described the way we offload the information on our social and environmental and technological environment is the classical collective brain picture which is entailed by mostly Boyd, which I saw in Henry of relative cultural evolution. What I want to show is that to go a step beyond we do not only have a self that is augmented by environmental cues. There is a self outside ourselves. There is things that shape such a life and that are not their agency that shape such a life and that is not human that is institutional. So that's the step beyond the collective brain in the vanilla collective brain view which is maybe an anti collective brain. There is actually such a thing as a collective brain starts an image and we can study it. Well we can study it now through Wikipedia and the internet like our network of the collective brain. Stephen? Yeah it's interesting as well in terms of we have all this extended information but there's this there's a very good book called The Paradox of Choice. So the the thing is we have all this information but you still have to do an action policy on what to do and the challenge is you can have so many you know it's like going to the supermarket and you know you can buy 50 different types of olive oil. It's just olive oil at the end of the day but now you've got to choose between 50 and you've got you know so there's a sense of abundance but there's also a need for an action policy which is of course what active inference talks about and ultimately it's not the volume of information that's going to be the game it's the ability to act and in appropriate ways and make sense of how to be in your niche so that's kind of interesting. Awesome thank you so we are going to move on we're going to go into the next section embodied intelligence in nested minds but just to quickly summarize here like what was the main the brought up in the introduction so we've gone through the keywords but just to clearly just stated out autonomous systems are defined by a set of processes that create and sustain their identity they embody their understanding and adaptive abilities within their physical structure they enact a model of their world and seek to maximize evidence for this model through active inference and they can maximize model evidence through modification of their niche so just to kind of sum up what an autonomous system is and what where we are at this point and then we'll move on to discussing embodied intelligence in nested minds so a quote from the paper it says here we argue that innovations are instead an emergent property of our species cultural learning abilities applied within our societies and social networks our societies and social networks act as collective brains we outline how many human brains how many human brains which evolved primarily for the acquisition of culture together beget a collective brain oh this isn't actually from from the thinking like a state paper this is from the innovation in the collective brain paper and this framework this collective brain framework grounds a study of intelligence in the way that an organism maintains structural autonomy and allows the formal assessment of the collective brain hypothesis so the collective brain hypothesis has been put forward but not formally assessed up until this point Avail do you want to talk a little bit more about that because i probably just botched it about how the collective brain hypothesis is not formally assessed yeah so in cultural evolution you have what i'd say is the dominant theory is gen co-evolution gen culture co-evolution theory which is basically centered on how we have cultural violence that have an adaptive role and there are a set of cognitive bias or cognitive mechanism but actually cognitive bias behavioral bias actually actually that help us identify what are the most adaptive cultural violence and this has a feedback effect on gene selection which has the effect on cultural selection and basically all of the interpretation is entailed by a process of a co-evolution process between genetic and cultural evolutionary information processes and the fact is there are claims there are numerous claims in the literature that there is some sort of collective computation that's entailed or collective cognition that's entailed by this so there is if i remember well Henry Shenmue to Krishna who wrote the innovation in the collective brain paper and there is i think Boyd and Smaldino Boyd which is basically one of the two founders of the initial formalism and Smaldino which is one of the most computationally savvy a member of the current literature that wrote a paper on a collective cognition no collective computation distributed computation because revolution has distributed computation i don't remember the exact data but it's i want to explain to you about its words like they have a model of individual bias in selection and they have a narrative interpretation of this somehow entailing some kind of computation and there is nothing in between there is a few paper of less interactors that's kind of interested in the co-evolution of sexual networks and cultural bias accents small art name them there is literally not one paper that explains how is this supposed to be cognition what are the mechanisms what is the cognitive architecture there is nothing right or it's marginal and i did not heard of it and the goal of the paper is to basically explain how we can grant a view of cognitive mechanisms that are entailed by the social cultural networks and social cultural constraints that are embedded within society and it's not a model because to build a full causal model of any brain or organism is not tractable but it's to prove that it's meaningful it's there is a positive way in which we can model societies or at least process within studies as a collective form of cognition and a key example of this is the state because we have access to the mechanisms we have access to different papers with which they understand and predict and control reality awesome that's a perfect i think transition into the next slide do you want to read this first Daniel this paper uses the framework described to illustrate that the co-evolution of top-down control and cultural niche construction enabled the emergence of the city-state complex and how these collective organizations maintain structural autonomy awesome thank you um and so that's kind of the the summary i say of the of the section the embodied intelligence in nested minds section does anybody want to give a thought or we can go on to the next section just what you just said about how we know the mechanism avil of the state um i'm thinking we see it's neurotransmitters and its action selection policies so it's like every system has advantages and disadvantages for its study this is one where many of the key features are designed and so it's a lot more like investigating the connections of a program that was written rather than the potentially like map territory debates over like well what is the bacteria really doing are we really measuring the bacteria but it's like we know how certain processes in the state happen but not all them and citizens usually don't know Stephen yeah so following on to what daniel said a quick question in a way is how much you maybe trying to have a sort of a realist approach on the idea of the brain or it being a a brain or how how much is it an instrumental tool or a metaphor so be curious as you know you take that correlation because you've got brain you've got mind you've got intelligence you've got extended cognition is it that you are really wanting to make it say there is like a brain that's kind of in enacting its own action policies or um so that'd be that might be helpful so i think actually like maybe we should hold that to the end section from the from the because i think we're going to get more into how is a city like a brain um as we get closer to the end of this contextualization we'll see for dot one and dot two because we'll cover just the background here but it's a very good question yeah very good um it's going to provoke a lot of discussion so that's why i'm like well maybe we should get through the rest of the rest of the the um talk going on first so um it is so we're moving on now to the deep history of cities and states uh here is another quote from the paper uh steven do you want to read this for us yes okay cultural niche construction allowed humans to alter their socio cognitive ecology in a way that allowed extensive functional specialization and the evolution of supra individual agency who mediate in institutions that makes me think about the evolution of functional specialization in ant colonies and it's like you can't have the super specialized task you know you can't have somebody who specializes on just sourdough bread until there's somebody else who can help them with some other tasks and that's the economic inspiration for why they call that process in ants division of labor downstream of adam smith 1776 the economic division of labor and the cat maker and the shoemaker and all of that so once you have a lot of people living together they can specialize in a new way there are new institutions need it needed to mediate and that results in more being different i love how daniel like says ants and gets like giddy and excited uh steven yeah i know it's this term super individual so that's kind of this i suppose in the ancony you think of the queen but in our world we we have these and actually i've been thinking about a lot of this myself in terms of how we relate to these bigger scales so this idea of a super individual or super collective ties into the idea of the state you know because you know we think of you know the word state is actually used literally in the united states to be you know a region but there is an idea that it's trying to maintain something so um that term i think is a is a is a useful one to sort of pull out and uh and it's a super individual agency so that ties it into active inference in a in an interesting way it would be interesting to know the term socio cognitive ecology it's not one i've heard before in quite that way so i don't know that could be something just to just to maybe ground that term in this context well we can return to it and i added the book cover for leviathan with the person made up of other people awesome thank you so um in this section of the paper avel explains how state niche construction niche construction depended on um serial monoculture the technology of writing material infrastructure dense settlement dense settlements and engineering societies from the top down as documented in seeing like a skate seeing like a state by scott um avel do you have a few maybe words to say about this i mean more than the brief kind of overview i just gave so i'm not sure i don't find the key uh subject like on the importance on the morphogenesis of states from a social scientific perspective just we're just talking about state niche construction like what what did that depend on all of those things um musically your state is um i think we could have a lot of definitions the orcans says it i think it's okay it is said in certain cities that it is uh essentially a monopoly of on legitimate violence so if there is a group of men that has weapons and that has the exclusive rights to use it and that this right is probably accepted it is a state uh i'm most interested in the mode of operation of a state so for me it's a chain of command basically it's an emissive systems that makes people do things uh so these those two definitions they entail um the loss of human autonomy they entail that you have uh at least a subset of humans that are constrained to follow the laws laws that are explicitly imposed so that's in clear uh deviation from what is done say in Iroquois society where um at least by the partial account I have mostly no one has the power to coerce anyone else and any decision is wanted on rational debates so this is not what a state is like a state is when you have a guy with the biggest bird and he can say okay you go kill this person because i want to i want this weapon and um you have um definition of the states that sorry uh you have definition of the state that is uh grounded in the sovereign i'm not sure when it initiates i think it's with leviathan which i did not read but it's basically that the state is a central actor that has power on anyone which is um sorry enough uh chats things um so state is something that has power on anyone and it's not tied to a specific individual um so that entails uh the capability for a set of for institution or a set of individual to um basically coerce other people to do things and this entail um for example enough at least an authentic session of resources for other people to not flee or uh enough uh imbalance in wealth for other people to get to you even though they know it will strip them of their autonomy so it's a body grounding on which the literature that project literature on states and state emergence is written and this entails agriculture trade a lot of ways for basically wealth to be centralized and powerful people to be able to constrain everyone else awesome and that like speaks very well to um how states get a grip on their subjects and themselves um and what is the the common grounding maybe that um is used to get like to get a heterogeneous um understanding of states oh sorry emergency alert your attention please your attention please update in action policy requested from the state exactly exactly so so now states get a grip on us through like our our our cognitive offloading into the niche right um steven did you want to say something yeah i mean this also speaks to that challenge of being able to get a grip on things which are beyond the adjacent scale like they you know they're trying to get you know they have to use they have these technologies that can reach out to everyone but at the same time it's kind of they can't easily get a grip so there's um there's some interesting points there around um you know because there's a few scales between us and the state right um and that's that's interesting and technological development like little tiny chips that track where you move that makes it a little easier to get a grip on where somebody is moving so it's a co-evolution with the states and technology going back to agriculture and writing and ledgers and now we're in a new phase of ledgers from zero one two three nice steven yeah and of course the challenge is that they get a grip on what you're doing but they their their inferences about why you're doing it yeah they're decontextualized so they think someone's doing x or y and uh that may be wildly different you know they may be underestimating the risk or they may be radically overestimating the risk of behavior awesome so i'm going to just flip through these next few slides so uh dense settlements the what the function of the dense settlement did was allowed for economic codependency and we talked earlier about the functional specialization that was like you know necessary for this economic codependency and it also facilitated the development of nested social communities and daniel wrote is there a scaling relationship here and i think we have to wait and address that maybe in the in the dot one and uh moving on it says enforcing state control over urban systems mandates building acceptance into the population and this is like how do you let the state be in charge of you how do you grant grip to the state and this is the people must perceive competency in the authoritarian regime and there's a redistribution of wealth like for example paying taxes and also symbolic legitimacy which is achieved through the church largely or maybe maybe there's other ways but but was in the in the early development of cities and states largely religious based so what um there's a lot of things that that we've talked about so that tie into um formation of cities and states and its large-scale coordination collective identity crop redistribution which i thought was interesting like some people grow it some people eat it some people take it some people sell it um and then the functional specialization specialized roles and the control of the money and the flow of the money so all these things together um unify to give a state grip and allow this kind of top-down control um is is there any comments on that or if not we can move on to the next section just that if those are the necessary and sufficient features when there's a crisis of competency or distribution of wealth or symbolic legitimacy what does that forevote of algae have an answer i'm sure waiting and waiting in the wings yes the answer is basically a structural structure of demographic theory which is a central result in clarionics which basically the states afford the possibility for the elites to ignore the social maintenance and just capture wealth and when they do too much it causes political instabilities so um and this entails uh fluctuating um fluctuating cycles of integration and distribution in states uh developments and by the way we in the sense of France the u.s and likely global capitalism are in a phase of this integration which uh needs which entails that we have uh you don't need to understand how to organize outside or uh in spite of dominant situation and to understand where they come from so we're going to move into the last section of the paper which is from the urban metabolism to the collective mind and we're going to go back to the main question of the article which was what kind of intelligence is displayed by city-state systems and how is it embodied within their structure and uh maybe steven do you want to read this quote from the paper yes okay so answering this question in terms of inactive influence entails revisiting our accounts of the urbanization dynamics as the results of city-state systems actively managing to get a grip on their world i minimize their expected variational free energy cool and there are a couple more um quotes from this section that we can move on to uh daniel do you want to read this first one here basic autonomy is indeed formally defined as operational closure or more precisely the capability of a network of constraints to canalize faster processes like metabolic flows into maintaining and recreating itself i like that's kind of a twist on the kanthian organism concept the teleological definition the organism is like the level it makes sense at and this takes a more operational and mechanistic approach towards closure of mechanism and uh constraints and autopoiesis rather than merely n-directedness which doesn't immediately suggest a mechanism and then since the material and socio-cultural landscapes of urban systems canalize human activities so as to maintain itself the process of urbanization clearly contains the constitution of cities as basic life food comes in waste goes out cities alive awesome and i love this picture of like the elephant is the city um survived did you want to say something also to point out that um the first point you made is that um the montevil and most sure definition or my account of them is a twist on the kanthian idea that organism is the level uh at which it makes sense and it's basically the movement uh that montevil and mocio and moreno make uh because uh and activism at first it's grounded in conceptual philosophy in phenomenology um this this is the grounding in which uh varela and maybe matriana i'm not sure articulated uh the view that cognition is in fact grounded in life and what uh mostly people and the basque center for life sciences i think it's called try to do is to mechanize it to operationalize it to ground it into specific uh chemical uh bare chemical processes so it's uh you precisely captured uh what these accounts is meant to do awesome and good good job inferring daniel based on based on the um you know stigmurgic cues left in the niche uh and and we can move on to a couple more quotes from the paper um this is one that that really resonated with me and it's paraphrasing jablanca and lamb in 2006 this paper that's here uh the evolution of information in the major transitions says a full-fledged evolutionary transition typically entails the emergence of new forms of information systems mediating regulation at the collective level and or the heritability of collective traits um and then another quote from the paper uh is also taken from this gaudy and crawl paper in 2014 it says the emergence of city states with their professional classes and their mediating institutions based on writing technologies constitute a clear case of such a transition especially in the case of the late agricultural civilizations and i i totally get this and also it really speaks to how stigmurgic cues left in the niche environment can can contribute to the formation of our collective brain i mean writing like we can leave stigmurgic cues for generations yet to follow um and some writing in stone right like so this is like still we can read it many many many many ages and generations later um which i thought that was really uh poignant steven you know one thing i'll be interested in in terms of the state is um how much different areas of society or the state um are trying to achieve similar types of outcome so when there's um and when there's i mean we've just had that case of um that video of a cortes being murdered and uh you know so like within the state you have internal conflicts you know within within the state itself um but also how much does the people at different scales have to be aligned or how much will the state itself kind of do what states do so to speak because its own mind awesome thanks daniel yeah i agree about not every single action in the state necessarily being the same like if you looked at the body you said well no we're supposed to be making glucose why are there cells that are breaking it down or like there can be sections that are doing processes that seem directly opposite to each other but that can be part of a broader regulatory process so it brings into question you know what are the official processes versus the things that are in the gray zone of the state and how do we go from this model into more useful forms nice thanks um and so we can move now to another quote from the paper that is from uh paraphrasing uh this paper that we actually just did in a recent live stream um how to count biological minds by uh mathu sims and in um in the thinking like a state paper it says if the nested hierarchy of regulation systems enables adaptive behavior at the system scale it effectively enacts a collective mind regardless of whether system components retain their cognitive ability um and this speaks back to what we were saying earlier about the the top level kind of constraining the lower levels like maybe there's like a um once you contribute to the collective then you know you've contributed to something greater than yourself and your your own cognitive ability perhaps doesn't matter so much any thoughts on that or we can move on to some more topics from the paper uh daniel do you want to read this um quote the closest analogue of our intuitive picture of a mind in urban systems is their most disembodied mechanism for adaptive regulation administration indeed just like neuronal cognition it entails the top-down regulation by a specialized population with the help of a specific purpose information system makes me think about those tubes that they used to put the physical packages in send them and now there's special information channels but you better be listening when that channel calls so that's like Futurama like they have like i mean the bureaucracy like the administration and i mean administration i'd love this quote from the paper i do think like administration is totally like disembodied from like reality and bureaucracy is like so imposing but like in the Futurama they all have like you know it's like space age modern transmission like you could just click docusign on the internet and like yet they're using like tubes literally like to to send the things off um and it takes seven years to get approval or something ridiculous like that but it like it does work like that right bureaucracy is is kind of crazy um this next quote says the institution of state-sponsored legal codes affords a way to directly transcribe administrative expectations into individual minds rule of law is therefore a core tool by which states create their world through building the basis of their own intelligent behavior into their socio-cultural niche and this is brilliant right like just yesterday my landlord called the codes officer um because my neighbors have weeds right like i know that i'm allowed to have weeds it's beyond the expectation of of like you know our neighborhood that nobody has weeds and everyone knows that this is the expectation some people don't maintain their weeds and we all have the right and duty and obligation to you know impose that on our on our niche environment Daniel so the uh expectations the normative expectations a lot of laws are like you know it shall be within two weeks after this giving notice that this happens so they're expectations and then they're called codes and so maybe think about predictive coding and it's like we can predict that this technology is going to be used this way so i know it's not exactly what people mean with the neural rate codes etc with predictive processing predictive coding but laws are normative they're saying what should happen and that helps behave as expected you want to be on the right side of the road driving in the U.S. so that's like an expectation and then it becomes an inter-subjective preference that is sponsored by the state so it just kind of cool again showing points of contact between people's day-to-day life and their intuitive explanations for how these systems work and then it's like yes we have an ontology that picks up right with you and then goes a little further than you might have expected to awesome so well yes that's exactly the idea knows until the actualization of expectations the fact that they are normative is i think incidental it's epiphenomenon to their core role which is to shape behavior you have a project in um labial studies i don't know you call that which is called i think homo-geuridicus or homo-egalis or some closely associated formulation which has a behavioral approach to know and we need to do that more yeah sounds good so we just have some summary quotes maybe from the paper and to kind of summarize the ideas that were presented here daniel do you want to read one or some of these okay i'm just picking which one i would want to read i think the second one is a lot like what we were just talking about with the state defining expectations and then modifying its own niche that operational closure to implement those expectations lest it dissolve then i i like this top quote our notion of intelligence is based upon active inference a scale-free framework applying to any system with individuated boundary states and then it's like even the word state is a pun as aval pointed out before etat being state in french and we talk about internal external blanket states boundary states that sounds like it's going to be at the boundary of the empire and then this is what it looks like to apply active inference to a real system to go from a sort of basian graph boundary notion to being like well what if that boundary state country was intermediating and insulating between these other two so i like how even the word state comes at the point i like that i really like that too and i like that like that that's kind of a play on um like the frist in particular physics like specific particular physics but also like physics as related to a particle um and and the idea of a state actually something when i was proofreading this paper was like you know i know that you're talking about like city states but it's not immediately obvious to the reader that like you might want to say like city states to kind of flesh that out like from the get go because it is it can be such a play on word right it's a boundary state it's a blanket state it's an active state it's a you know city state daniel maybe one other thing we can talk about in the dot one and two is states as we know them are new and so it's only been a few generations like you're a great great great grandparents that's a long time ago it's a lot of tacos ago but it's a blink of the eye in terms of evolution and in terms of deep history and so even pre westphalia that was not so long ago and there's so many world traditions and so just realizing that we're in a very non-stationary moment with the development and the implementation of states it's really important because it can be easy to be like oh it's a grind all the states in the un and you know 50 states in the us it's like even 70 years ago there weren't that many states in the us and there was different countries in the world so things are changing it just sometimes it's at a different timescale which makes sense because they're bigger and slower entities than we are survival i want to point out two things first the plan is not intended but no i will claim it is and no one will check the second is that the received you is that states emerge with agriculture it is what godi and crawler said in the paper i cited which is not accurate or we don't know whether it's accurate you have a pretty good seeing review of the literature of the state imagines and the fact is we do not know what humans say it is what like before written history we do not know so i and most importantly it's something grabber and when grow to a leading anthropologist in history anthropology like to rave about there are kinds of causation that are authoritarian there are kinds of causation that are not so even if we are signs of i don't know great ritual states or anything we don't know whether those kind of organization were like what we know today because western nation states are extreme in kind of extractive and cohesive behavior that are shaped by war within europe and between europe and colonialized countries so i prefer i much prefer to ground the history in which i talk in european history in modern history after west phalia after the birth or in the roots of the birth of the nation states because it's pretty hard to talk about the roots of somerian states we need to do so because those are states we need to do so if you want to talk about what states are but if you want to talk about what our states are we need to go in west phalia we need not to go in somer or egypt or whatever so just to kind of play off of that extractive and corrosive behavior and that states are often born out of war something that you mentioned in the paper but i don't think maybe gets enough attention is the concept of shared identity because in this shared identity like that's what makes us faction often to our own state we're going to divide and become something new and dania was saying how we have so many states now so many more states than we've ever had before and this like factionalism and we all have like this distinct identity is this going to breed more war or is every state going to become so small because they'll just faction and keep dividing i mean you can get something the most homogenous human culture you can imagine and they will find something to divide them so so is this good is it bad if we keep dividing is it going to be like a homogenous like bunch of sub factions or is it just going to lead to like more division and more war if i discuss that it will take a lot of time i understood the session was yeah we are much of track definitely definitely after we'll save it for the dot one okay so moving on the collective mind does not directly emerge from inter individual interactions but from slower evolving dynamical constraints which supervene on but do not reduce to human activity therefore the proper scale of analysis of collective cognition is an integrated activity of entire urban systems and should more specifically be grounded in the broader processes by which they maintain and reproduce their structural identity understood as an operationally closed set of constraints underlying their metabolic cognitive and ecological activity you know replace urban with brain body just you could say the same exact thing the brain the body have to be an integrated unit finally blue what does it mean to think like a state i'm going to put your overlay over the video so it'll look like your estate go for it the cognitive activity of states is defined mechanically by the standardized administrative systems they use to understand and manipulate their world and functionally by their drive to build and enact grip over the material and sociocultural niche so there it is the question the answer we made it through from hz and i like it's not what is it like to be a state it's not the nagel 54 what is it like to be a bat this is more of a cognitivist interpretation that helps us actually potentially find useful ways you know like i don't want to do a you know in in x country you grip the state people can kind of write their own jokes on that theme nice um and we can just think about what maybe these questions lead to maybe what um a good understanding of this work might enable avil mentioned earlier the um necessity for this understanding because of the the kind of divide divided space that we're in right now um what are the next steps for the fp and active inference uh what are the goals what are you still curious about if you want to participate in either of the discussion videos get in touch with us we would love to have you on to talk about these ideas in more detail do you guys have any closing thoughts yeah just in dot one let's uh dive in and open some threads take some notes in dot two let's see where we go maybe somebody would like to pick up on one of those threads or apply one of these ideas and say oh i was working on a history of my region and so i want to see how i can apply that so we applied active inference to get thinking like a state and then how do we apply thinking like a state for something else whatever it is that somebody is interested in seeing avil thank you for coming and kind of guiding us through our contextualization video it's always a pleasure do you want to have the final word i don't have anything anything smart to say still yes but these are the words the paper does speak for itself quite brilliantly i will say all right authors who come on dot zero are the best no it's all good thanks for joining blue and facilitating see you later survival and thanks also to steven so hope to see you all in the dot one with a dot two or beyond so bye bye