 Alright, hello everyone. Welcome to Actinflab live stream number 33.1. It's November 23rd, 2021. Welcome to the Actinflab first-time participants and returning participants and listeners. We are 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 with feedback so that we can improve our work. All backgrounds and perspectives are welcome here and we'll be following good video etiquette for live streams. We are in the final two weeks of November 21, discussing thinking like a state, paper of choice, and we have some tentative paper selections for number 34 and likely will take a break and start with 35 in the beginning of January 22. But we're really looking forward to an exciting 2022. So if you have any ideas for people to invite for other streams or if you yourself would like to join, then just let us know. Today in Actinflab stream number 33.1, we're going to be learning and discussing this very cool paper, thinking like a state, embodied intelligence in the deep history of our collective minds by Aval aka Serval Gwennan Kalu. And it's just a discussion video, even though we have the author on the stream, which is awesome. We're pretty much in the dot one going to go over anything that we want to, we can return to some keywords and bigger questions that we raised in the dot zero. We can add notes for next steps for questions to think about over the coming week to touch on again in the dot two or possible ways somebody could continue this avenue of research. And we will be discussing this paper in next week, 33.2. Okay, we'll go around and introduce our panel today. Everyone can give just a short check in, they can say hello and something that they remembered from the paper or something that they were excited about to discuss and then pass it to someone who hasn't spoken and then we'll conclude with the author. So I'll start. I'm Daniel. I'm in California. And just as we were discussing right before we started, I want to explore this aspect of states as anticipatory cognizers, not just reactive cognizers. And I'll pass it to Dean. Hi, I'm Dean. I'm in British Columbia and in Skonstein, a Cedar box today out in my cabin. And I like the paper because I want to see whether or not there are parallels between what happens on an individual level, what happens on more organized levels of individuals. And I'll pass it over to Blue. Good morning or good, whatever time of day it is in your world, everyone. I'm Blue Knight. I'm an independent research consultant in New Mexico. And I like this paper for many reasons, but my favorite reason is the scale free aspect of self assembly. So it's self assembly at a much larger scale than I'm used to. I normally study cells and subcellular organizations. So I like how active inference unites the things that I like to study with statehood and larger scale organizations. And I will pass it over to first time participant, Andy. Hello, everybody. My name is Andy. I'm in Richmond, Virginia. I'm very interested in picking up the mechanics of how intelligence works through a collective. Right? Are there universal things that we can get by just the movement of information through relatively closed systems and how to modify that, the cognition that goes along with it? Now we'll pass it along to Stephen. Hello. Good morning. Stephen here in Toronto. Yeah, I'm interested in this paper in terms of the scale, the social impact of looking at states and how active inference scales up. And also maybe a little bit of provocation in terms of how much are we looking at the dynamic states of organisms or macro organisms and how much are we looking at the state of a dead set of artifacts that helps us work with our niche and the scale of working with macro artifacts in a niche as much as macro cognitive agents. So I'll be interested about that. And I'll pass it, I believe, to Daniel. Thank you, Stephen. I think we've all, we non-authors, have gone. So, Serval would be awesome just to kind of recap. Hello, and what brought you to writing this paper as we kick off the discussion. So, hello. My name is Avel and also, I'm training physics, epistemology, weird things. And what I'm interested with this paper is to show that if we accept a naturalistic understanding of cognition, which is saying that it's based on actual systems, actual dynamics, then we can't resist the conclusion that social systems are organized in a way that is cognitive. And I found this basic consideration pretty important because then there are no limits like systems that are cognitive work in the way they are structured to work. And we can change things and we can understand how change will affect how the thing behave, which is pretty urgent because I think it's pretty likely that the nation-state systems that is dominant today, nation-state capitalism system, is dying and is basically bringing us with it. So we have to explore the state of alternatives pretty quickly and efficiently. And this needs reflexive understanding on what the state-state is. Great. And I think it's a pun or homonym that will return many times. What are the states? Because we talk about partitionings of states, a lot inactive inference, internal blanket, external nested systems. And now we're talking about the nation-states of how are social systems cognitive? We use that state-based partitioning to talk about other cognitive systems, ones that are designed or evolved. And so we're applying active inference to study the cognitive properties of states. Stephen, and then anyone else? And I'll just keep on adding notes or questions and people can look at it or suggest things in live chat. Go ahead, Stephen. Thank you. Yeah, I think it's also cool with active inference, is that kind of ontological challenge whereby we're dealing with non-equilibrium dynamical states in a nonlinear system, which gives rise to pullback attractors, which have this non-equilibrium steady-state quality where, however, often states are, or say we think in quantum theory or something, they're like, it goes from this particular state to that particular state. So there's also that interpretation of what can be done with states when you start to have nonlinear dynamics at play. Thanks, Blue. So something that I mentioned in the .0 video that I'm interested in maybe exploring a little bit deeper is like, what do states aggregate into? So what is like the grander scale if we're all just self-assembling into something and the quantity of states that are produced? I can't remember who said it. Maybe it was Avel that states arise as a result of war and faction and division. And so as we are progressing forward in time, there's an increased amount of faction and division. And it doesn't matter how homogenous a population is, they will find something to divide about. And so this creates more city states within city states within city states. So is there some kind of equilibrium, like a tractor space for like the optimal number of states? Or is it all going to be like unified? Or is it all going to be dissolved at some point? Thanks, Stephen. Actually, that's a good point there, Blue. There's also something interesting about democracy. They say that for democracy to work, it almost has to be a split of around 45 to 50 between different opinioned parties. Otherwise, if basically one party is supported by 65, 70% of the people, generally speaking, it doesn't work very well because then there's no dynamic, I suppose, to move backwards and forwards. So I think that's quite interesting. So when we talk about the minimum number of states or entities to exist for democracy to function opinions, it's interesting. In the system that Mike Levin works with, back from our July live stream, the computational boundary of a self, he works in Polinaria, which are regenerating organisms, but you have to have a minimum number of cells and particularly like a minimum number of stem cells. So I think it's like a thousand cells at base and like 120 plenarian stem cells, or you don't get regeneration. And so like what kind of dynamic is required or how many people and even for like, you talk about hunter-gatherer organizations like 50 to 500 people, they think made like an optimal hunter-gatherer colony, like less than 50 and you're going to be eaten by coyotes or whatever. So where is that optimal number in statehood? What does that look like? Yeah, what aspects of the system are necessary, sufficient, incompatible at odds with certain kinds of regimes of state? And it is really funny how we talk about policy selection, policy sounds like it'd be more from international policy than it would be from reaching for your cup of water. And then we talk about regimes of attention. Usually people hear regime, they're thinking some historical regime of a state, whereas regime of attention doesn't necessarily carry that connotation. So linguistically, there's a lot of contact points, and that's why I think it was an awesome contribution by Surveill, was to just draw those out, ground it in the physics of critical systems and of cognitive systems. And that's what is contributing to hopefully some of the ways we'll explore it's going to be used going forward. So Dean and then Stephen. Yeah, this might sound like pushback or challenge or whatever, but it's just I have a personal issue with the entire idea of scale free, like that there's some sort of invariance across scales. And I think that like when we see, for example, a lowercase a morph into an uppercase a, even though they are of the same categorical type, and fit in the same place in the alphabet, there is a there is a change that goes on. And so there is a scale variance aspect to this that I think. So again, this is my sense of it as opposed to sort of the agreed sense that there's a scale free process in play. The other thing I would I would wonder a little bit about is when we take a constructivist view, a city as built, I'm not sure what happens to the bidirectionality that active inference of words. So again, I'm not saying that we can't find qualities in both, but I'm not sure that a city as a reflection of something constructivist is necessarily how living things reflect and predict. But again, I'm open to having my mind changed. Thanks, Dean, Steven, and then I'll have a comment on the constructivism. Yeah, I was just also think it's a good point we blew about the Mike Levin's work and the question of, yeah, how things are scaling up. And of course, the big thing that Mike Levin did mention that, you know, they, they've got these quite a lot of ideas around how things grow, multiply, and relate, but no idea how they stop. And that's kind of also interesting in terms of, you know, when we think about scale, it's like, has the organism reached its point because the top scale, the maximum scale has kind of been reached? Or is there some sort of, is it that there's, it's the dynamics of a mid scale, which is like an action scale? So it's not, it's not that it's ever arrived at this thing, it's just arrived at this mid mezzo action policy landscape, which is available. And, you know, is that availability then understood from what's inside? Or is it something that's checked to something external in the environment? So that, that I think is an interesting question. Thank you. Capitalizations. Yeah, you have to, the state capital, capital punishment. There's so many other funny word plays, but to the constructivist comment. So one sort of simple constructivism would be like, I mean, there weren't bricks there, and now there are. So the city was constructed. And now it's the scaffold for us to interact in. So you asked Dean, where's that bi-directionality? So I think there's a few ways to talk about the bi-directionality. So first is niche construction can be a symmetric relationship. We talk about the agent modifying the niche and the niche is modifying the agent. And so it's like, you have the blanket as the interface. And then there are action and inference on both sides of the blanket being used as an, as an interface. And so it's more like a co-constructionism. But if you know, Lego set A is building B while B is building A, that starts looking more like interactionism, because it's not just instructions put the brick here. It's a specification of a protocol and an interface for interaction. And that's what opens the door to improvisation and to other kinds of dynamics that aren't just simply like assembly of the 3D printer. And then the other point, and then Dean is, it reminds me of Dennett's 1995 book, Darwin's Dangerous Idea, where he talks about basically two ways that you can talk about the appearance of design in natural systems, which are cranes and skyhooks. And so if you look at a skyscraper, you think, how could that have appeared? One approach is they built a crane to put that last top piece on. And if that crane was too big to construct, then a smaller crane built that crane. So you can have a cranes all the way down, like small things were built and scaffolds were in place that were removed later to result in this skyscraper. Or you can just cut the knot and just say, well, maybe there was a skyhook and they just latched it to the sky and then built down. So that's like top down constructivism appeals to skyhook technology, whereas bottom up emergent process based co constructivism is more of a crane based approach. So I think that's a very good point, how even with constructivism, we can still retain a lot of the important aspects of constructivism, like physical construction, but we don't need to fall into a unidirectional or a top down trap. So Dean, what do you think about that? And then Stephen or anyone else? I think the important point, Daniel, is what you just mentioned, which is the trap of prematurely collapsing to the top down privilege. For example, if you go to just outside of Denver, you end up in a place called Red Rocks. And then the question becomes, did somebody enact Red Rocks because they put a stage at the base of an amphitheater or did Red Rocks enact everybody who was attracted to that place? And so all I'm basically suggesting is when we look at how things are constructed, we don't want to lose sight of the fact that sometimes things construct us as well, so the skyhooks aspect of it. Thank you, Dean. Stephen? Yeah, and also, I mean cranes and skyhooks, it gives that idea of how it links in, they're also dead items, right? So they're both tools or they're both objects. And we can bring in skyhooks, which might be informational instructions or cranes, which are, you know, places to carry the bricks that are going to build things. But I think that when we talk about that Markov blanket and what goes both ways is, yes, I think when we're dealing particularly with dead things, like bricks and that, we construct an environment. Now it's a bit more complicated when we impact an environment that also grows and changes and is alive. However, when something comes back the other way from the environment, is it constructing us or is it scarring us or is it giving us lots of little punches in the gut in which we heal? So I'm wondering whether the other direction, it can't construct us, but what don't kill us makes a stronger type of thing. Thanks Stephen. Andy, please. And then Dean, I think in the context of cognition in the state, the specific, the element that's missing right is what's the relationship between the nation and the state, right? So the nation would be the bottom up information or the bottom up cognition and the state could be theorized now as the top down. And then broadly speaking, right, the nation creates the state, the state, the existence of the state changes the circumstance, the changes in circumstance forces the nation, the people to rethink how they're operating in the world around them because their circumstances have been modified. And now you have this constant bi-directional relationship between bottom up and top down in the context of state versus nation and modifying circumstance over time. Thank you. Lou. Dean, go ahead. Oh yeah, Dean. Yeah, Andy, I like what you're saying except where it's already collapsed to an A to B relationship. And I think what active inference allows for is to keep things up on a generalized, abstracted level. It doesn't have to be, here's the nation A and here's the people B. That's the difference essentially is if you want it to be an A and B thing, go ahead. That's good for argument's sake. But in fact, what active inference does is it keeps one side of this open and not necessarily specific. It keeps it in order to be predictable. It asks the counterfactual, what if, as opposed to having that pre-presented as the relationship. And it's new. It's a new type of relationship that's arising. And I think that's explored in the paper, in the deep history section. There's really the specifics. There's the minute particulars of the specific states that we have and typified by West Falia and a bunch of other historical events that the specifics matter. We're in a complex system. The trajectory really does matter. Lou. So I totally forgot what I was going to say. But I would like to maybe challenge Andy, what is the difference between the nation and the state? Is the nation the people that make up the state? And then the state is the collective of the people. Is it a part whole concept or are they two separate larger entities and people make up both? So words, depending on which definitions you go with, they get a little bit squishy. In the context that I use them, state is a construct the political organization of a people is downstream of the shared identity of the people, of eight people. So the nation would be broadly the largest sense of shared self, distributed among a relatively closed system in this regard. And the state being the construct by which these people organize their affairs inside of their nation. That's a relatively new idea, right? The idea of a nation and the nation state is relatively new as far as like an overt construction. Beforehand, in the age of absolute monarchy, there was no separation between the state and the people, right? It was all embodied in the person of the sovereign. So different historical context, different periods of time, different cultural context. Now you might get different definitions, but in the way in which I was using it, the state would be a policy requirement of the culture in which it was born, right? That's why American democracy functions significantly better than South Sudanese democracy, even though they have literally the exact same institutions. So Daniel, maybe you want to flip to slide 33, I think, like where we kind of show in the paper all of the things that go into the production of a state. I'm curious as to where nation fits into this. Is it in the collective identity or is it a relationship between nation and collective identity or just what your thoughts are on that? Yeah, absolutely. Collective identity to me would be the definition of a nation, shared sense of self. Everything else are things that you may or may not need to state to accomplish that may or may not be more efficiently conducted in the auspices of a state, right? Control the monetary flow. You probably need something as sophisticated as a state, but to transfer resources back and forth, you don't, right? You can grow grain and you can trade it for meat. Go ahead. Just the collective identity I'm just thinking about in organism and the liver and, you know, the bone cell, it's like collective identity doesn't mean identical phenotype. And so makes me think of a cultural memory where each person, maybe there's a few touchstones that 80% of people know about or increasingly though, there's less and less shared information amidst the informational changes that are happening. So what does that mean when everyone's only going to have a little partial facet of the collective memory on board, but maybe they can enact broader processes that are truly collective? So how do we respect differences and shared aspects across multiple levels? So blue and Dean and then survival. I want to hear what you're thinking about this. So just like Daniel, how you were talking about like the bone and the liver cells, like making up an organism, you know, I think that that speaks here to what the specialized roles and like the density of settlements enables that like functional specialization and like you can have a blacksmith and a shoemaker and a farmer and everybody takes on their own little role and then goods can be redistributed. So it just like reminded me of that and like that functional specialization and the importance of it. Thank you, blue. Dean and then survival or pass. Yeah, when I was watching the 33-0 and I saw this slide, thank you blue for asking for it to be brought back again here. This is again a specific to a specific because it's historical and it's reflective and the fact that we can look at things and say this is what happened. We can look back on them. I'm not sure we want to necessarily conflate that with active inference, which is about the recursion. So this is what happened. We can see how things moved across a timeline from point A to point B, a city-state evolved. But is that recursion, which is what active inference is about, which is variation upon variation? Again, I know how we can diagram it and represent it, but is that representative of what actually occurs in recursion when certain thresholds appear and then just as quickly disappear? Interesting. So survival, if you'd like to add anything. And then Stephen. So there are a lot of threads that were opened. What are you asking my opinion about? I think part would be just on nation and state, similarities, differences, what is and what if, and then more on this sort of integration slide. Where does active inference come into this tying of the knot? How does active inference, about nation, I don't think it does. Nation is simply a collective identity that is stated to a political entity. It's a community of peers that happened to have some stakes and power over the behavior of the institution. What are your states? And the fact is that nation are not universal. You have a lot of states that are not nation states that are not city to community of peers. For example, the Roman Empire, it was a empire, which means that you have people who get the power that form an entity group that is the Roman. And you have all the rest of the people that exist in there that also ruled by the empire that is dominated by the Romans. So you do not have community of interests that is stated to a state. You have a netting group that dominates other ethnic groups to a state. Nation are not, they are not invariant aspects of what a state is. However, they are a collective identity, or at least ways to build boundaries between a group and an ad group. They seem to be an invariance of efficient human collective action according to Astrom's principles. So there may be more tendency to build nations, to build nation states when you have more important between group selection. And it is said that, I don't know of any computational activity work that's associated, but it is said that modern Europe is pretty extreme in terms of between group conflicts of war. Because the diamond agreement is that the geography basically enables trade at great scales, but prevents the constitution of an empire in Europe. So historically, there were a lot of states in Europe, as compared as say in China, where you have a big state that is, I don't know, in terms of surfaces, maybe two or three times like Western Europe is, and more integrated than one you can have in South America because of the mountains. So you have a pretty ideal regime to get a lot of intergroup conflicts and to get a selection regime that favours nations, organizations that somehow collate with states. And that's what happened to be instituted to, I don't know, the French abolition. I don't know the story of that. There are a lot of pathways in Europe for nation-states, but in Europe you've got nation-states that originated, you've got pretty aggressive kinds of imperialism that originated, and both can be traced, I think, to an intense regime of between group selection and of war. And Europe results that tend to say that war explains in the sense predicts more political complexity, which is a debatable construct that basically maps this kind of integration of a society. So the history of Europe, of modern Europe, Sicily, and Medieval Europe, and everything past the urban empire, would have selected for both nations and aggressive kinds of imperialism. And this is the kind of organization we get to live in now. Thank you, Sirval. Just like the ecological regularities through evolutionary time scaffold the evolution of an organism's cognition, this is a transposition of that argument to the biogeographical and contextual and ecological factors that give rise to and scaffold the development of states as things, as cognitive entities, there are some parallels there. So it's an awesome point. Thank you. Blue, and then Stephen. There is a pretty important thing to say that is a state does not actually correspond to a nation. Maybe states have original nations. Maybe they are aesthetic nations now, but a state need not entail a nation. You can have a state of a specialist in the control population, you can have a state. Well, that's kind of the usual deal actually. So, yeah, a nation is a collective identity of inner community or an interest. A state is a control system built from a social and it can be and often is captured by the elites that is ethnically, like socially in terms of the cultural norms, pretty distinct from the people in controls. Chicken and the egg. Egg as the secreted state. Chicken as the life form, but then there's a co-evolution. So that paradox stands there too. Blue, Stephen and Andy. I'm going to defer and come back around to my idea. So Stephen and then Andy. Yeah, I think the whole gaol question is very interesting here. So often in the European Union, you've got the European Union, it's kind of a very union of pluralistic partners is the idea and you've got this tension where one side is basically saying, no, you're not your estate. We want to be a nation, an independent nation as this tension, right? Because we know we're a union, we're a union. And when actions are working, that works the union, right? And so I think that there's a scale where people try and work at and the rate at which people work at. What's the dynamics of people in the same way that when you have politics seems to have a very slow trajectory, you know, things shift, you have the state, you have your identity of your party. However, near in an election, a week, this phrase is a week as a long time in politics and it is near an election. The dynamics start to become much faster. Things can change much quicker. So this scale question again comes in is what is what is a unit of identity? How does it manifest at different rates? And just one other thing in terms of identities in participatory community based action research, okay, they talk about community is actually a unit of identity. It's almost like there's so many different types of fluidity at the community level and overlaps between communities that it becomes a unit of identity to actually work with in measuring stuff. We tend not to do that at the higher level of the nation. It's just the flag is a unit of identity. It's a very which hasn't maybe changed in 100 years. So it's an interesting as well as the you know what how much can things change and how much are the ability things to change caught up in the artifacts that embody them. The artifacts that constitute them in some case. So Andy, any comments or to blow? Yes, thank you. So I was going to, well, let's go see what you just said, right, smallest action, element of action in this context. Identity is the smallest element of action. However, it is guided or bounded by its ability to affect change. So if you deeply identify with one other person in this world, the two of you are the small that you have you identity becomes basically shorthand for interest. We two people by calling ourselves a group have a certain worldview and a certain desire to see our interests achieved. We have the capacity of two people to accomplish that. So what is our ability to affect change at that scale, relatively small, once you're talking about anything larger than interpersonal relationships, right? But it doesn't change the fact that your identity is shorthand for your interests and that we all at whatever level or scale we're operating at are pursuing our interest to the best of our ability, founded by relative power, right? But this idea of nation versus statehood, survival, you're absolutely right. The state is a thing independent of a nation. Sometimes they correspond and typically we say that that is that that's a good fit because otherwise it's an empire, right? Otherwise it is somebody else's ideas of what constitutes right and wrong in terms of projected values or institutionalized values that are now imposed on you because the state is not representative of your cultural worldview or your shared identity. It is something provided to you by somebody else's in the form of empire. But the question that I had, right, going through the slides, there was this idea that the state as a top-down hierarchical control system, what is then nation or what is then identity? How does that function in the sharing of information and cognition relative to the state? Thanks for the text-driven question. It was a core argument of the article just to quote that states can be and should be understood as hierarchical control systems. So, Avel, if you want to give a direct response, otherwise Dean Blu. So, there is a direct response which is that to talk of hierarchical control systems, basically to talk of top-down control, this is, I think I'm referring the same point, the same thing in saying those both words. And of course, there is no cognitive systems that is pure top-down, but they are, they do not work. They live in labs and they are not useful. So, basically, the organic organization of whoever happens to have a say in what the state is, is at least the ascending dynamics in the collective brain. They are what affect, in Bayesian terms, what change the priors and the policies, et cetera, et cetera. So, in the case where you have a defined nation, defined entry group that happens to control the state, which I'd say is exceptional from what I've read, you can say that there is an meaningful nation-state fit and the nation is basically the brain and body of the state in the sense that, well, the ascending dynamics that influence what the control systems does. In most case, you will have, I'd say, a pretty much stronger decoupling because basically only elites would get to say in what the state does and in cities where, in most cities, the elite is not to define an ethnic group. In that case, you will not have a clear fit between a nation and a state, although, of course, the committee of power that defines a state, if you conceive it as a nation, will define what the state's prior are. Well, I'll participate to define because, of course, the tool, et cetera, et cetera. Thank you. So, awesome. A lot of hands. So, Dean, Ben Blue, Steven. So, I want to just circle back to what Andy just said. Andy, when you, if you go and look at this livestream three hours from now, even when you see a perfect reflection, we'll assume that this YouTube video comes out exactly as it intended. I think what you just said proves that we're not talking about scale-free, talking about scale-friendly, because what you will see is not what you just did. And so, I think that that has a huge bearing on this idea of what we're talking about here. We can shape shift into what Steven Silat has pointed out here, is that that fact that we can fill between these two levels of scale doesn't mean that it's one size fits all. It quite literally means is that we can fill, we can change, we can modify. And I think that's recursion in its essence as opposed to taking a reflection and knowing that the next time we look at that, it's going to be different because there's some elapsed time between the moment we said it and then the moment we heard ourselves say it. Dean, I think it's an awesome point with scale-free and scale-friendly. Scale-free, people usually mean to say that the mathematics or the formalism doesn't have an a priori system of interest or scale, like this is a centimeter theory or this is a kilometer theory. So it's scale-free because it doesn't have that a priori reference scale. But then scale-friendly, it's almost like, how about you tell me what scales you've identified and let's make them friends through interaction and iterated modeling. And so we're friendly to scale and scaling, but that doesn't necessarily mean it's free, although from the other side, it's also fair to say that it's a scale-free framework in the way that a Bayesian computational framework or even a linear model is scale-free. It doesn't a priori say that kilometer is a default unit for linear aggression either. Blue and then Stephen. So I think what I'm going to say is going to touch on the comments that were made by both Andy and Seval. First of all, in terms of this relationship, the mutual interest being the fundamental unit of identity and collective identity, this really goes back to expansion of your cognitive boundary beyond yourself. And when we were talking about the computational boundary of itself, again, I'm going to always refer to that live stream, but there's an informational sharing when there's an identity sharing. So when you expand your cognitive boundary to include more than just yourself, like if you have a really deep connection, a really deep relationship with one person, you have a great degree of mutual information between you and that one person. And then so say you have a family unit or a network of close friends, there's a large degree of informational sharing. And so in this sense of building a nation, like the fundamental unit of identity, then maybe is actually informational. So it goes back to like Bayesian principles where there's this informational sharing that then allows you to update your generative model. So just to kind of speak to that and then also maybe to just like maybe help Andy make the link like between downward causation imposed by like a government or like a hierarchical control institution and like the emergence that happens, like when people come together and share information in a collective, there's a super cool paper by David Krakauer called the Information Theory of Individuality. And it talks about like in order to become an individual, and there are many different kinds of individuals, but in order to become an individual, there has to be like a bidirectional information flow that's both emergent, starting with like the fundamental units, like the people sharing information, there's an emergent and then a top down. So bottom up and top down. And like when you have this bidirectional information flow, that's like what allows you to form a boundary, I guess, around your collective, something like that. Thank you. Blue, Steven, and then anyone else? Yeah. So picking up what Dean said there, and so yeah, there's an actually blue and everyone else is, the formula is scale free at the free energy principle, and then you get into applied active, so then you get into active inference, which has a sort of a scale free quality, and then you get into adaptive active inference, and then you get into, you know, inaction of adaptive active inference, and then you get organisms like us that want to apply active inference. And I think that I think it does like really start to go scale friendly, once you start to go up. And one thing that I find interesting, I've been, this actually is nicely ties into this infusion space work that I do, but I look at the face and the body. So there's a particular scale at which my mind is embodied, and you can see it. And the bit that you see is the bit that we show on this screen. You're not looking at my foot, or my shoulder, or my hands, you're looking at the face. The face is the embodiment of my mind as an identity. Okay, this is my projective identity, and it becomes very useful. So it's scale friendly at that scale. But, you know, until the other scales of my body aren't, you know, my liver doesn't do a great job of my identity. And then we got, okay, what scale do I have to get to? I'm talking around, if I'm just talking one on one, I don't sit here while waving a Canadian flag or whatever. Now, if I'm a crowd of a thousand, now maybe we need to have a big flag on a stick, and that becomes my face. That becomes my projection. But that's a scale friendly, because now I've, again, it's kind of gone in and recapitulated in the environment. So that's kind of interesting. Thanks. Thank you, Stephen. Serval? Don't really get the debate about scale free, scale friendly thing, because what a scale free theory is, is a theory that is, in principle, a formalism that's in principle, it's not a real different scale. So for example, the opposition to the weak and strong forces that only have a measurable effect at very small special scales, gravitation and electromagnetism that can influence things that are pretty much afar. Well, not electromagnetism, because you have the mutual concerning of person and strategies, but in principle, it could affect things at distance because of the error square, the dependency. Yes, getting. And the fact that you have a theory that is scale free does not entail that every system where you express the theory is scale free. For example, you have the gravitation theory that is scale free, but if you study the solar systems, it has a scale, which is about the distance from the sun and the photographs thing apart. If you study the electromagnetic motor, it has a scale which is the size of the motor. And if you study a biological organism with a scale free formalism, it has a scale which is whatever scale at which the organism happens to exist. So I don't get what the debate is about. Dean, how about give a thought on that? Yeah, I got a thought on that. So I think we can see what a reflection is. We can see what a triangle is. We can see a very small triangle and we can see a very large one. Those are reflections, but that is not recursion. And as soon as you get into variability upon variability, you're going to have to come up with something more than a reflection. And I think that's why you have to come up with something that more than scale free. I mean, we can see scale free and nobody's arguing against that. But what I guess what I'm pointing out is that if we limit ourselves to that, we're not going to see the entire picture. All kind of intermediate, I think they're both very valuable views. Scale free is certainly more aligned with the way that people do talk about formal models, whereas scale friendly is a little more of a Deanism, which is always fun. I see scale friendly as emphasizing the active role of the modeler in connecting the levels. So it's not enough to just say like, I have a stretchy ruler that could apply to different scales. It's not an a priori scaled ruler. And then making it more of this process of approaching the mechanisms and the processes that are interacting amongst scales, rather than just thinking, well, we'll put the linear regression on the little thing, we'll put the linear regression on the big thing, and then we'll have a scale free applied theory. This is more like, let's take that system as a indivisible unit of function and then like interact with it in a little bit more of an active way. And it can still be scale free for those who like that way of talking about it. But for those who want to get to know it, maybe scale friendly is approachable as well. Steven? Yeah, I think there's also active inference and adaptive active inference and inactive inference. So I think like Daniel made a good point there is keeping the active inference flow through and helping that is really useful, right? So that we don't lose the form of the side because that's the danger, right? We clip off the, but there is something, for instance, at what point is adaptive active inference possible? Like is that the point when you need to have the brain within a reason or there needs to be some generative process that's regulating active inference for it to then adapt at different temporal scales? Now that raises some other questions about exactly where, but there's even the idea of what adaptive active inference and active inference or adaptive inactive potentially is, starts to then tie in well, you know, and if I can only do certain adaptive active inference with the brain, then that's the scale of an organism that has a brain and below that scale, it can't. So there's something that becomes transitionary or contextual. Dean, on this point and then to serve all for a combined response. Yeah. So again, I think knowing that a triangle is a triangle regardless of the volume of its footprint is important. I think when we include updating, when we include time, then we should include scale friendly. So I'm not discounting scale free, but I'm saying that the updating process, whether you want to call that adaptation or inaction, the change aspect of it doesn't get resolved simply because a triangle is a triangle. That makes sense. Bucky Fuller said angle is metaphysical and general. It doesn't refer to any specific substrate to say that something is a 60 degree angle and frequency is always bound up in space and time and the specifics. So any realized system is going to be both. Surveill. So Stephen, I think talk about adaptive active inference. Active active inference is something that KShop introduced in 2018, I think something like that, tries to reproduce an active notions of cognition from active inference. What it says is that you have agency when you have a graphical modulation of active inference system, you have higher level of active inference that modulates lower levels of active inference. The question at rise is basically a question that is not solved and I don't even know what's my opinion on it in the ontology of sense and naturalism, which is do we have different level of existence or do we have only scales? And even at this level, when we admit that we have things within things, within things, within things, and that's what the world is, it's complicated. The thing is that to circle back to a question that has been raised initially about how active inference relates to other values of cognition, the way KShop tries to formalize an active concepts of cognition is entirely wrong. Because activists, they do not say that organisms are hierarchically modulated. Active inference theorists say that the brain is hierarchically modulated and they apparently projected it on the activists. Activists say that a system that is cognitive must be operationally closed, which is it has a set of, it has a structure or an organization, I don't know which because of the recognition that basically reproduces its own existence without any clear support, without autonomously prompt its environments. That's not how they say it, but whatever. And you have a specific formulation that is computational, it was formulated by people from the Basque Center of Life Sciences, I think the name is, which is constraint closure. So you have constraints that are slow processes that shape how faster processes behave. And you have a lot of them and collectively the slot of constraints shape faster processes so as to maintain themselves that is constraint closure and that is what's the latest version of on the elective, not agency, but autonomy notion is, and this entails not a hierarchical modulation, but meshed nets of processes that make something else exist at in-between scales. And the notion of what is the antics that use to give this is pretty complicated, I don't know. You have people that think that level is just an epistemic construct that is built by scientists and this is pretty consistent with a hierarchical approach. With a meshed approach, this everything becomes ambiguous, the meaning of every single terms become ambiguous. So I have no idea what will be a correct way to refer to this. Scale 3 says what it says that it is not interested in specific scales of behavior and it also says implicitly when we're in the context of active inference that it is not interested in a specific anticlevel, what applies for cells applies for tissue organism which applies for societies and which in the case of vanilla in France also applies for dust, vortex, flames and whatever. So I'm not sure, I'm not sure what the national scale is supposed to do and how it relates to the question of whether states are properly cognitive because we don't need a proper version and proper national scales and proper anti-distinction epistemic distinction between what is recursive and what is only a scale invariance. We only need to state what being cognitive is and to check whether states are and of course what this means for the broader epistemic constructs with builds around this. So I have to like go back to the active inference live stream number 30 when we talked about how to count biological minds and really this like raises so when you're talking about the distinction between adaptive active inference and like mere active inference and in this this theory about how to count biological minds is everything that's biological do all things then have a mind like even down to like the single cell organisms and in this paper like there are a lot of people who have this like mind, life, continuity hypothesis and a lot of people really ascribe to that so in this adaptive active inference versus mere active inference I couldn't find the line like I really pressed on Matt Thims about this and and he responded beautifully like he also couldn't find the line so at what scale or what level does adaptive active inference transition to mere active inference and what I'm fundamentally getting at is like okay so are you doing mere active inference if you're a single cell and you just move forward and back like towards a repulsant or an attractor like is this mere active inference or is this adaptive also and so like adaptation to the niche like to me I think that happens at all levels but there was this distinction made in that paper that like adaptive active inference happens when the collective forms some like emergent behavior that makes it more fitted to the niche environment and so like where is the state then a biological mind does the state make a set of organisms like more fitted to its niche through this like collective adaptive behavior or is it maybe maladaptive because overall like as a species the more divides we have like the more war we'll have I don't know. Thank you Stephen. Yeah thanks for this meshed nets idea I think that's definitely true and that's kind of fixed in with the kind of meshed architecture. I wonder if like Blue has mentioned about this different scales of active inference and identity is quite important here and the nation state is kind of is maybe identity driven and I kind of wonder if we'd scale that down I was thinking scale down I was talking about the face there earlier but for instance the baby in the womb okay is the release during childbirth of the hormones which triggers the woman to break her waters basically like a broadcast from the identity of the womb so okay body I'm here now I'm a child bearing womb I'm declaring myself in the same way that maybe the pituitary gland but it does it all the time declares itself certain times to crash out you know create insulin so could we see that then this meshed architecture comes at a scale where it's possible once the you know and then it's like and in the same way then the nation state maybe as we go up in scale the expression of signifiers to broadcast out becomes greater and greater you know becomes a bigger and bigger part of it and maybe the biological processes become less a part of it well definitely it's the question of realism and instrumentalism whether it quote is a broadcast or whether we'll just choose to model it as so one of those is definitely safe territory and the other one is a little bit harder to make claims about what the territory is versus the map that we choose to draw around the territory but very interesting points Dean okay I'm one I just want to make sure that I'll get feedback from people on this so to me adaptation okay adaptation around what is which can be scale free or if we're talking about active inference do we also include the counterfactual what if and I think if we include the what if then I'm not sure that adaptation looks the same it's not reflective so I'm not pushing back on survival as much as I'm asking are we are we overfitting are we saying that recursion doesn't matter and are we saying that scale friendly doesn't matter because we only want to concentrate on what is that's my question Dean I'll give a thought then Steven and then anyone else active inference agents that are capable of doing adaptive active inference as a title so just a recently proposed word as survival brought up their internal generative model includes a model of perception and action as well as of cognition so that's meta cognition the what is of meta cognition at one level can be a what if at another level what is is the agent can actually be imagining a counterfactual with respect to perception and action so cognition is that friendly scale it helps us bridge the what if and the what is because um there's there's so many other angles one other path to walk through that would be like um in Anil Seth's book being you there's a great discussion about how counterfactuals actually are like the basis of cognition like how do we know that we're looking at a real mouse like on my computer table versus like a picture of one at the same resolution because actually there is an implicit counterfactual I could move at a different angle and see something different to the physical object but I wouldn't expect the shadows to move on this jpeg and so actually the difference in our perception between two things that visually can look identical relates to deep situational awareness that actually blur or at least reimagine the relationship between what is and what if Steven and then blue thanks actually this paper is serving as quite a useful bridging uh provocation um because as you just mentioned there Daniel with um okay so what if you know a way of what if and counterfactuals so there's there's the cognition which is a particular scale of perspectival or even implicit what if the counterfactual but what if this and that happened in the future well the other question is what am I becoming and actually if we want to look at scale three if active influence is based on free energy principles and free energy principle was pretty much the only form of non-linear dynamical system out there but now we've got some of this stochastic stochastic chaos work also like if you go right back to non-linear dynamical systems and a lot of discussion around non-linear dynamical systems people are talking about thinking about that as a process of becoming so how does it gives you away what dimensionality can use well the the process of becoming starts to tie into that um and cognition as we see it doesn't maybe the process of being is more a process of becoming but the process of what I broadcast if using that term very crudely but what what my identity conveys is a much more steady much more of a an artifactual state if that makes sense especially if I take a photo of it and put it on my profile on facebook thank you um here's an article on being and becoming and that distinction as well as I'm sure a panoply of german words has been discussed for a long time about the relationship between being and becoming and just to give one thought before blue that generative model of the active agent includes what kind of agent am I what are my expectations preferences affordances etc so we get in the deep active inference framework built in what if and then what and then who would I be and how would others see me and what would they think if I were to do this and then this were to change that's all part of the state space of what's possible these are like the grammar in terms of the very austere outline of perception cognition action and external dynamics that allow composition and new design motifs for those kinds of questions to be asked blue so if you wouldn't mind flipping back to the twisty slide I think it's 33 um and when you're thinking about like are we talking about what is or are we also including the what ifs in the model like when we talk about what it's like to build a state so so this is like how states have been built before but but how can we like take this and like as Daniel was like referring to at the beginning of the talk like how can we take the what has been and incorporate like what if via active inference to predict like how to optimally optimally form states in the future so so is active inference incorporating the counterfactuals are we going into what ifs or like what about what's next like not just what if but but what's next to predict like the the cognition essentially of a state um and how it's it's going to think going forward great question and what's next we can have our expectations is we can have our preferences they can interact in a different way so dean then andy yeah blue I think that ties in with the point twos that we tend to do all I would all I would all I would add to what's been said is that if if we're looking at as as is we're looking at this reflection and and can we ask off of that of what are we becoming of becoming which is a function sometimes of scale many times of variation so I know like I keep beating this horse because I want to include the variation and the fact that time changes our picture that's that's all I want included in this and I don't think that that as is a perfect copy as we update so I maybe I'm maybe I'm explaining this wrong but I think as change occurs that perfect reflection that perfect copy doesn't move forward without variation concentration sometimes sometimes we just toss it out the window andy that's the what if sorry thank you dean andy and then I'll give a thought on the as change occurs so uh in my travels I've run across this idea that somebody tried to quantify functions of a state that are universal right what are the things like how to how do you maintain health care not necessarily which way is better but distribution of resources education health care things that we associate with the state maybe not every one of them every state exists or every nation exists or let's stick with state maybe not every state does every one of them certainly not all of them do it to the same degree of competency and they are all bound within the expectations of the people who um comprise that that polity or that that entity but to get after your question blue right if we if we want to get to what if how is this different how do states form differently what is the optimal thing that I think you've got to get down to like um a core set of variables for which states are supposed to provide solutions to or address um and then from there we can see well what if they did it this way what if a state prioritized this and not that what if a state prioritized this function over that function and why would they right what are they responding to I don't think that they're responding to um necessarily uniquely physical or material requirements I think there's some interplay between um intellectual or uh value space requirements as well as material requirements all mixed up in time and space given which parts of history we choose to preference so um in a particular moment thank you so one note on the as change occurs so here's a very influential book with a a broad follow-up lineage on the geometry of biological time by Winfrey and it brings us back to the discussion of kairos and how change occurring in chronos we become like pro crusties we have to stretch and pull and change the sequence of action to fit onto the scale as defined whereas thinking about the space of action the geometry of biological time is defined more by action like embryological events rather than how many seconds have ticked on some atomic clock in some bunker so that um as change occurs can be through time but it may not be um the kind of time that gets ticked out on a clock and that's just a very different way to talk about actions and inference because it's dynamic but it's not just like okay the clock's running and things are changing it's like there's actually a timeliness to action and so we're talking about the dynamics of the context and the timeliness and then maybe there's ways to map chronos and kairos or maybe one time we play more into one or the other so dean and then anyone else yeah just real quick thank you andy because I think what you did is you applied a partition to preference and priority and you made sure that we didn't say that they were one and the same so in real time we see we see an evolution uh throughout the discourse so I think that's that's a real example of what kind of happens as we update so one other random thought just from earlier and then of course anyone just bring up a thought or anyone in the live chat so we were talking earlier about like capital with the ol and capital with a l okay so this was pretty interesting capital with a ol comes from capitolium the name of the temple of jupiter at rome on the capital and hill so dating in english from the 14th century so that is referring to like this building that's like very much like city on a hill and um city on a hill is used as a multi-scale illusion to the united states at the very least um like capital hill but of course broader notions of what the state stands for and then there's capital with a l capital on the other hand has a far wider range of meanings and application the word comes from latin as well from kaput meaning head and then it has a lot of um polysumus uses like capital punishment which we talked of being the seat of government like sacramento is the capital of california but then like venture capital capitalization so that's just interesting um i hope it's not simple word play when there's a homonym but it's very interesting that they both draw back to rome which was alluded to earlier when thinking about the seeds of what today's states are on the world stage so andy then steven yeah um survival it would institutions of the state right like ministry of foreign affairs be can be considered an example of embodied intelligence in the context of cognition for the state uh yes uh something that blue said earlier is that um people proposed uh collective agency to be based on the capability of a certain kind of regime of behavior to influence individual fitness what i would say is that this is wrong the geological organization it is by definition organization but it is constraints that exist upon the collective behavior system if you look at state say it's a steel beam it is a sector it is something that affects how the steel beam uh interacts with the world and this is something that is entirely invisible from the perspective of a single particular steel which is the natural lower level lower scale perspective you can articulate so the view that our organization is entailed by uh things that affect the niche of their constituents is fraud what organization is is uh constrained to specific scale on specific uh behavior on a specific dynamic and um the relation it has uh with the body the body uh which certain scale uh certain scales of cognition is about is uh the set of those constraints uh that's correctively enact behavior that we say is cognitive so one uh mystery of foreign affairs or something is doing a thing it is embodied in the sense that it participates uh in a greater network of constraints uh which collectively build on realization uh which entail the behavior the specific each specific decision of the um ratio of line policy so for example if there is some kind of um collective say uh panic moral panic against communism is likely uh in the absence of any formal statement against communism that uh decision be taken by that decision be best against uh specific people and words and um teams but targeted communists and this is not entailed by anything at the formal level at the level of symbolic computation that could be construed as constituting the highest or most uh recognizable forms of state cognition yet this participates in um a collective behavior uh that's adopted from the border system uh this is basically the nation state distinction familiar uh the aggregate behavior of specific sets of position constraints uh past processes will build uh the slower processes and basically entail the behavior and this is um not something that needs to be knowledge from any higher level and this yet is uh cognition and these grounds of view of and recognition are something that uh foreign states um are something that is uh grounded in not the body of family single individual but in the body of the state as the uh set of its constituting constraints thanks sir val i also put up a quote that people can pause the video and read from the gulag archipelago that specifically refers to organizations as organs of the state which is a little bit of a almost sinister meat grinder way to describe the state and it refers to this interplay between the awareness and the actions of individuals and the systems outcomes at the state level steven yeah i think as well as this this is really useful i think this embodiment and our intuitive ability to embody um does then recapitulate the influence process so as well as these terms uh the seats of government or whatever coming from say we see it coming from roman literature but over time it's been encoded and encoded and maybe abstracted out and we use words and sometimes you have to look up the etymology to sort of find out where they came from but we we do intuitively see a difference if someone says someone's going to take their seat at the senate where you get a feel for that but you don't say they're going to take their seat at the palace they're going to take their they're going to they're going to sit in the throne right because the whole point of the palace is that those seats are not everyday seats right so there's there's a but you get that because of how you know space is treated around one of those seats like you don't you don't put a load of seats around and roll up a load of thrones you give space to it um and there's actually a lot of interesting work in physical theater where you literally can work with space to convey how much space you give someone will convey how much power they have that reminds me of um the term maronym so that's like when you say there are 10 mouths to feed it's referring to a system by referring to the sub part of the system so it's kind of like marrow topology is about the parts and holes and neurology and so the maronym is when revealed through our language we say there were 20 seats on that committee um and the fluidity at which that is semantically unpacked shows that there is a relationship like when you say there was 20 mouths you think there was 20 people and that includes a lot of our assumptions um you know 10 pairs of hands or you know 10 hands what does that say about our assumptions about you know ableism or about what the default is in our way of speaking like that Stephen and then anyone else yeah uh and well ableism definitely comes in then as these these codes as people are starting to look at ableism that that's a very good point and also there's that intuitive understanding that you know there's there's the idea that a team generally is really more between five and ten people do it optimally be a team and then you've got this idea of a group so you have an intuit when you hit 20 you've you've gone past a small group into a large group and there's a whole work around large group interventions and getting aspects of the system in the room which is about when you can no longer look at each person in the eye and hold it which is sort of after 15 16 people so again you hit this to some extent we don't need to be taught that these things evoke the reality for us as this structured organism in space in place now it's probably different for a dog on four legs maybe they can only sniff around four dogs right and that's a big crowd for them so you know and is that about their mind or is that about their their morphology so it's it's definitely an interesting question thank you serval and then anyone else i like the way you talk of the throne as something that whose function is communicated through medically embodied use because the community activity of the state is embodied in constraints that are not human but those constraints needs to somehow affect the behavior of humans of because those fact faster dynamics are basically the dynamics of human individual and they need ways to constrain it and influence it to exist as constraints and a major way they do so is by building the niche humans living so building the way humans uh embody their perceptions relate to each other to each other take decisions etc etc and to simply build a monumental architecture to convey that specific buildings of specific functions and are very important and that you have to understand the underlying philosophy to exist as a functional member of society it's a way to do so and that's a way which the throne basically affects people to believe in the kind of exceptional qualities of the sovereign which are necessary for a state state that has a sovereign to exist so the throne and the crown are or any other kind of uh token there so it's um it's embodied refers to a wearable or a bodily position it refers to of course in a cultured uh and a semiotic context um it's enacted that relates to declarative speech acts i pronounce you married the speech act is the the action and so when we're talking about writing and language driven systems which is a key part of again that that deep histories section declarative speech acts become very important when you're served the subpoena or when something is ruled and the hammer comes down on the table with like sort of like the finality of the act um like an onomatopoeia relative to the declaration that was uttered in a certain ritualized context um these are great examples about how we can use like the four e framework um and also is going to say it's extended because the crown and the throne they're literally off of the body and then what's the story that's like the total um uh like oh king got owned emperors new clothes where the extended cognitive apparatus that signals the cultural position of the king is revealed to be not what it is and so it's like to strip the person of their role it's just like stripping off their clothes and these are just all kinds of very accessible phenomena that we can ask how does this support a four e or an active inference approach relative to some other approach like signal processing yeah i'm sure you can make the neural network that recognizes a crown versus other images but that is like a sub component of visual perception within this broader picture and then also it helps us ask um what if and how we could design embodied and cultured enacted extended etc systems that reflect the statistical regularities of who we are now and how we think things are and how we expect and prefer things to change that's a good yeah i think that's a very good point and maybe this is where you know the survivor's talking about constraints there and also i suppose it comes back to the idea of when to stop until i talked about this a little bit with dean uh yesterday is and when do we know when to stop is one of the big challenges about you know ecological context and i keep thinking so you know you have the hammer comes down you know if you have a state you know the hammer comes down it you you you viscerally have an embodied understanding of what's being communicated by them you know that comes down something's been hammered out and in the houses of parliament in the uk they have this big mace they call it the mace it's an ornament an ornamental mace but of course a mace well i don't know if this is exactly true i think of a mace normally it was something you club someone at the head with right in a battle right so it's there is maybe that that big stick con idea that's being brought in but well i one thing that really came to mind when i was when this was coming up was what happened during nazi germany when the new technology of sound systems came out so the big thing that happened was these big rallies that completely overwhelm people is no one had had just like we didn't have with social media before no one had had an embodied experience to calibrate and in some ways you maybe never can because it becomes slightly beyond the concepts of embodiment what it meant to be in this big arena where you heard sound coming from everywhere and then the idea is that that was part of the reason where it become disruptive maybe we weren't able to constrain and it just became over a cancerous almost overwhelm but there's something about new technologies and the ability to constrain and find a state that is to some extent sustainable and when it becomes out of control and is there something about the embodiment that daniel was talking about there thank you steven survival and then anyone else so i want to emphasize that uh as in their formalism i'm criminal too the notion of the constraint is not the certain tale the notion of an interdiction of a tap down order or of of a stop on a behavior a constraint is a slow process that modulates that modulates a fast process for example if we agree on the sound of some word say lamp uh so we have a relatively slow process that is our couple behavior that will constrain our relative faster processes in the sense that when one of us use the word lamp the other will understand what it refers to so it will basically allow her the dimensionality of the possible behavior i have because i will only pick behavior that are consistent with the fact that you said the word lamp as i understand it um so the way that we culturally build our niches um it's not uh something we understand as constraining it's something we understand as creative and yet it corresponds potentially to the um the concept of constraint as would be used by a physicist or stem biologist which is uh the sense i use in that uh paper thanks it reminds me of a paper i could find it with like uh it was like developmental constraints or opportunities and it was just saying you can think about like okay i guess there's some morphological creature that just extends in a blob but any other one that has facilitated enabled affordances to grow in one direction can be framed as the constraint from the other dimension and so there are two sides of the coin to talk about what is the constraint that enables it's like chess it's the enabling constraints on chess moves that allow brilliant chess games and a culture to actually arise so survive with the direct response then steven and blue yeah thank you uh i think blue uh had to hand first dimension uh that's uh what i wanted to say is that um to have something that exists as an input construct who have interesting behavior you need a symmetry breakings for example you need um partial of gas to not be independent the default assumption of a physicist is that um nothing everything depends on nothing each particle evolve independently essentially so this means that uh i need basically um the mean and the variance of their individual trajectories to to have all of the insert information on my system this correspond to like pressure and temperature and depending on what the trajectory is um well this there is no clear mapping actually uh whatever uh but if i somehow forbids um if i somehow build uh correlation between the behavior of those particles then uh there are new possible behavior pretty much you have new possible elements of this set space uh that exist you are the creation uh as far as a physicist that is concerned with behavior within certain set spaces um is as far as uh covered by press but the physicist sees question because there are things that were not part of the set space it drew at the beginning of the system that become existence uh something that was not in that now is and what is interesting is this process of becoming uh as uh i don't know who uh as we discussed earlier uh because uh you have kinds of symmetries building and symmetries breaking that are constructive and yet they need to be constrained because if everything just arbitrarily adopts any kind of behavior in the set space that is accessible that is possible in principle you have just a philanthropy you have what noise you have literally nothing there you have just particles that blah blah blah blah blah blah and uh the facts uh of existence that we have things to talk about it entails the concerning of the possible set space to everything is accurately to some of the things are more likely and uh it entails a meaningful map um between the possible states at the macro level at the uh lowest possible level description and the states as we are interested in in a specific uh at a specific moment so thank you was um writing on the way that constraints are not restrictive but creative thank you the absence of restraints statistically constraints would just refer to you know an unconnected Bayesian graph just independently vibrating variables and so to have anything that's tracking or anticipating something else we're discussing the imposition of constraints using that term Steven and then Andy you mentioned i see what Daniel you put there enabling systems slash constraints so i think um there is that i i totally get and i think it's a very good point about you know you've got the statistic stochastic chaos and you've got the general dynamical processes and there's there's going to be some way of constraining that through different levels um there's also then uh you know if we take consciousness as being at the brain the brain stem and not in the frontal cortex you know and it's then sending stuff out you know i can see that biological dynamic but you also have this enabling system process of scale i suppose that in a way that i was just referring to and these things can start to be like the ball rolling down the hill so to speak and you know it starts to accumulate um its own because it's it doesn't need to extract information to make inferences to make things happen it it's just happening through thermodynamics and dynamics um i wonder how we can think about the challenge of these macro systems then the enabling system overriding the emergent dynamical system thank you andy is there a fundamental difference between thinking like a state and thinking like a tribe or is it just a level of sophistication um with what specialties are are enabled once you get a state level thanks for the questions sir all states states has administrative systems a tribe does not uh so like you have a lot of ways societies can be organized that are somewhat in a continuum and what i call a state is a specific control system that well is in a predatory state either it exists or it does not and in the case of tribes i don't think um most of what we call tribe have an integrated decision system more like a shared um ethnic uh cultural norms and if they have intuitive decision systems it's certainly nothing like uh administration it's so fascinating because the tribe can be implicitly existing without a state as sir all just noted and it makes me think about web three and dow's where it's almost like the states are being defined first and then the people are rushing in to fill so instead of the historical emergence of specialization and elaboration and formalization on top of an implicit background of the people and then in a complex interplay as we've been discussing this is almost like a computational state that people are migrating to with a preset governance system sometimes one that's incredibly constraining and people may not even know how constraining until they're there so that's an inversion of how we've seen the um members of the state and the state's evolution engage so more things change the more they stay the same but it's a different river for sure so dean blue steven i think i found something that i can i can say i can agree with with serve all and and maybe some of the people who really like genetics because i i think constraint really matters and i and i think the idea that ratio is not phenotypical drift through neutral mutations even though mutations are a ratio of the non mutations i also think that ratios and randomness accommodations are not the same thing so i think constraint allows us to be able to see the difference between those two things one is when a triangle is a triangle and the other is when a variation plays upon a variation and i think we can accept both that's what i think the constraint piece actually helps us pull the two things together thank you dean blue then steven so just to kind of go back to what um andy was saying about the difference between a tribe and a state but also like when do we know when to stop um and processes that constrain and when i think about this like bi-directional information flow like if the emergent energy is too strong like it then it's out of control right there's no constraint but but like when the constraint is too strong like i think about like you know even in mental action the idea that like you have your grip is too tight and so like whatever you're trying to control like pops out of your grip right like so like you can squeeze a balloon and then boom um and so there has to be like some balance and i just wonder if that balance is like quantifiable like does downward information have to equal upward information flow like is there like a balance between those two things that allows like a self emergent or not a self a self organizing entity to maintain in the state of an attractor because of like the information flow is too much in one way or the other the balance goes off and it it'll steer away from that attractor space steven and then i'll add a comment on that blue yeah i think that there's the the the ability to constrain is it is going to be at different levels in in the states um you've got certain things which are just like constitutional we hear this in the states there's constitutional rights right they're fixed they can't change it's been there for 200 years or whatever and then you've got more negotiation and in the uk you've got the government has to go and talk to the queen every couple of weeks or every week so in theory the queen could constrain now again they don't just have to send the letter they have to go and meet in person and talk there has to be so then there's this intersubjective dynamical constrain and then of course then for scaling you've got the kind of we can't really do it as organisms so we put in an electoral system but the electoral system and the democracy in north america comes a lot from the tribe the the sort of scale tribes of north america as much as it does from france it's quite interesting and one thing they have in toronto we have the treaty of the dish of one spoon and the treat of the dish of one spoon is basically they sat round in a circle and they have a spoon they have a dish with one spoon and you only take enough for yourself and you leave enough for nature and it's past rounds and that's the treaty of how to look after the great lakes and actually it was a lot of that kind of slightly scaled version of tribal work which then informed and said okay europe you're killing each other all the time what about something a little bit like this so actually democracy in this new u.s. was partly from the not really tribes anymore but that kind of scaled piece so i think there's there's there's that need to sort of draw back into something which is more contextual more dynamical as well as the kind of i don't know the more kind of law-based approach thank you so people have been mentioning the getting a grip which was of course from the paper about how states get a grip on their subjects and that was drawn from the work of of yella and ridfield talking about optimal grip and skilled performance and all these other cool inactive and ecological psychological ideas we've talked to so it's like of all the trajectories that the golf club can take we're constraining that to some lower dimensional manifold that's getting a grip and i totally agree i think a good example of an empirical system and about what kind of reciprocal interactions exist and what attractor states does that beget uh is the federalist anti federalist discussion at the end of the 1700s in the united states with people saying if you have the affordance for a standing army at the federal scale then we'll see this happen at the lower state level or if there's the ability to engage in foreign diplomacy or print debt at the federal level then we'll see this happen or not at the state level and so i think that's where we go from sort of the what has been and what is to the future that we don't know about the what if where we are in that scale friendly level where we're talking about designing or modifying or preserving or intervening in those causal threads that do link and co constrain the smaller subunits with the larger processes and how do we talk about the space of design there and integrate our preferences and expectations together with just sometimes the dazzling and tragic and complex history of just everything we see on this map so we'll have just um a few minutes left let's from anyone who'd like to let's just hear like some last thoughts what's something cool from today or what's something that you're looking forward to discussing more next week you know otherwise it's been an awesome dot one so steven and then we'll go around to anyone else yeah following on what you were saying there daniel and i'll use this as a sum up it slightly opens up a new thing which maybe we'll get to the next session but you know how do we see what happens with the nature of tech companies and the huge affordances that that puts into our landscape um also the constraints if we take the McLuhan tech trade of how it then changes the way we behave but overall they're introducing rather than um a constraining or a sort of a wise constraining way it's putting out there what's out there and it's it's forwarding huge amounts of new novel interactions be it microsoft be it google be it facebook and then they're getting hugely powerful just through permeation into society so you know something that's interesting is then where does you know i think that may reflect one of the challenges we have at the moment with corporations getting so big um so thank you and this is a very very good talk and lots of things are sparking off it so i appreciate it thank you yeah like nation states used to be like states were the big players in town there was big ones and there were smaller ones but that was sort of like the ultimate level of analysis web two we saw websites stock tickers apple amazon etc become bigger than many states and intertwined and amongst them in a way that sometimes is even more advantageous though they don't have citizenship themselves they have user identifiers they have revenue flows they have governance models and then now into web three we go and platforms and protocols maybe even decentralized ways of accumulating governance and economic systems that will also be at least initially larger than smaller nation states and web two tech companies so there's going to be like legacy like web zero one two three four all of these systems are going to be on the same territory but they're going to have different maps so they're going to be coordinating very differently um would anyone else like to give a closing thought i think there's a lot of stuff in the dot too that we'll be able to like jump off into like survival next week i want to hear about where we go next like what are your next personal research goals and what are other opportunities for individuals who want to like learn more or research or apply in this area or under this framework blue and then anyone else so just a final thought i'd like to like discuss more or maybe get more into like what is the state anticipating um and how do they anticipate uh what's coming next just predictive behavior yeah and it will be fun also to like just connect some of these to the core terms in the active inf ontology just like what are the sense states of the states what are the action states of the states let's you know what is attention is that the most wanted list what are the ways that we can just take the terms that we already have and start to build up this table of mappings so that we can and maybe it's going to fit on multiple levels so it's not going to be just the one and only map of maps but it'll be a draft versionable participatory set of mappings and that'll be fun dean yeah what i really appreciated about the paper was the ability to sort of look at things through the scale in variance lands and now i'm really curious to see what would happen if we got off the map and started moving around and then actually incorporated scale variants because i think that's really the and then what happens piece that this paper of paper affords it gives us a good grounding and now we can step off of that and think in the hypothetical space and i'm really curious to see where that goes excellent and if you have any last comments well thanks for letting me play i really enjoyed the conversation and a lot of this was new to me right i'm i'm new to the even the idea back of inference but being able to discuss it in concepts of the state uh incredibly interesting so thanks for letting me participate excellent well it's always a uh two or more way street so we all had an awesome time thank you everybody for participating we hope to see you here and or participants through listening in another live stream in the lab working with us on projects so see you all later talk to you next time