The Shape of Systems to Come (Session 3)

We are live. So should I start? Yes. Okay, so let me just share my screen. Okay, can everyone see this?

We did for one second, but now it's not there. Oh. One second. Something went wrong. Let me reconnect. One second. Thank you.

Tony, it doesn't allow me to share my PDF screen. Like you don't have the window available? Yes. I mean, let me do it one more time. One thing you could do is just share the entire screen, the first selection, and then just know that your whole screen's going to be shared. So minimize everything and just have to be. Now basically close all your pawns and stuff so we don't see the other windows. Oh, sure. Now can you see it? Yes.

Yes. OK. So what I'm going to do today, I realize that probably if I just am in, well, that kind of also causes some trouble. I realize that the best thing to do at this point is kind of a step back after last session that we talked about the sharpening of the distinction between manifest image and scientific image through the controversy between Szilardzian scientific realism and empiricism. I thought it's not quite obvious why basically we emphasize on this particular controversy

and what are exactly the implications of the scientific image for self-conception of man, and respectively for its self-transformation. Which is to say, basically, we are still not, we haven't still discussed what it means to basically have access to the intelligibility of the non-manifest. So I thought the best way to proceed is to somehow step back and talk about basically

what is exactly the scientific image and what it does basically to the self-conception of human. So with this new plan, I expect that we wouldn't be able to talk about the manifest image this session, especially that any remaining time after this discussion about the scientific image, I'm going to devote it to what I promised last session about models, talking about models, about basically what is exactly the complexity of nature, how can we detect it, and what

are the most optimal models to track dynamic tendencies of physical world. So next session then would be the manifest image and the synoptic image. If we can, I will talk to Mohamed and Tony. We can have like an extra 20 minutes so we can wrap up the discussion of the first module. Okay, so just to confirm if something goes wrong, if you can't hear me, someone says something because all I have in front of me is just a slide. I can't see anyone.

Everything is going well, Reza. Everything is fine. I'll let you know. Sure. So, the development of natural sciences after Copernicus signals a new phase in the articulation of intelligibility insofar as the human confronts the intelligibility of that which exists independently of human apprehension. This was the main point of scientific realism, Szilard's scientific realism, that we elaborated last session. The intelligibility of the world challenges precisely those activities and assumptions through which the human encounters himself in the world

and takes this framework of self-conception as sufficiently intelligible. Indeed, the challenge that the scientific image poses against the human's standard apprehension of nature, in which he conceives himself, should be understood as an index of intelligibility that overrides the apparent intelligibility of the human's position with regard to himself and with regard to nature. So we confront ourselves with an index of intelligibility that overrides and surpasses the intelligibility that surpasses the index of intelligibility

that stems from our spontaneous image, namely how we appear to ourselves and how world is appearing to us. To this extent, the scientific unbinding of the intelligibility of a world outside of the manifest image should be taken as a vector of transformation for cognitive inquiry, one that unilaterally forces thought to change its structure and rules in order to be intelligible, both at the level of internal configuration and at the level of purpose, namely at the level of intelligibility of practices. Now, the excavation of the scientific image

and the articulation of intelligibility of a world that lies outside of the manifest horizon have a number of immediate consequences. These interconnected ramifications can be roughly enumerated as follows. First, the scientific image points to the structure of reality he has a complex system of non-manifest entities. This was basically how Salars extract the distinction of the scientific image from the manifest image

through the debates surrounding scientific realism. The scientific understanding suggests that this structure is not only comprised of different kinds of systems, but also is endowed with radically emergent properties, which confer new functionalities on the set systems and further diversify their structure. And this is basically the definition of intrinsic emergence. Intrinsic emergence meaning that a pattern emerges in the system, and the emergence of

the pattern in the system comes under the closure of systems principle. the extent that what is appearing to be new, the emergent pattern, generates a new functionality for the system. Namely, the system uses this new pattern to basically produce a new functional class or a new behavior. Moving up and down these structural functional levels requires inductive leaps.

This is from James Crutchfield. And change in explanatory models and respectively alteration in conceptual schemas that map the objectivity of a given language onto the real object as a system of identities which undergo transformations. Now this is because when we are talking about physical systems and complexity in nature, as we talked about, we are talking about different classes of structures. And these different classes of structures or regularities constitute different hierarchies of structure and function.

Now each structural level and respectively each functional level needs to be understood and needs to be detected, measured, and quantified via different explanatory strategies. We cannot overextend our explanatory strategies nor models all the way down or basically allied different explanatory levels, quo distinct hierarchies of a structure and function. These interlevel moves can be explained in terms of transitions between different

physical computational classes. This is something that I will hopefully get back to at the end of the session. that basically regularities of patterns and structures can be explained in terms of distinct computational classes. When we are talking about computational class in this sense, we are specifically referring to physical computation as different from logical computation. Physical computation or intrinsic computation can roughly be described as a computation

in the sense that it accounts for how structures of a given system constrain and support information processing. But the difference with logical computation is that it divorces semantic of utility from information processing embedded in processes. Logical computation needs to be understood in terms of semantic of utility in the sense that, for example, 2 plus 2 equals 4.

We have output 4 and we want to basically create a computational transition, an algorithm, that maps certain inputs to the output that we have in mind. But this output, the utility, the prediction of this output, and hence what I talked about semantic utility, is what is absent in physical computation. Now each new computational class transcends the antecedent domains on which it operates.

Radically novel computational classes, and again we are talking about intrinsic computational classes, radically novel computational classes become incomputable in terms of their lower level or antecedent conditions. Which is basically, this is a kind of structural functional equivalent of Thuring's Incomputability Theorem, Godel's incompleteness, and respectively Cantor's diagonalization argument. Hopefully, I can get back to this and review them briefly at the end.

In the sense, the detection and quantification of these complex hierarchies, sorry, in this sense, the detection and quantification of these complex hierarchies or types of regularities which constitute the structure of reality, or in fact articulating the intelligibility of anything new in the world, poses a challenge to thought, especially since it appears that all we can describe is expressed in the language of our current understanding.

Number two is that the scientific image points to an image of the human that is fundamentally counterintuitive and alien to the human's spontaneous self-narratives. We talked about these spontaneous self-narratives or basically self-conception under the rubric of epistemic given. in I think in the first session that we were talking about for example Darwinian naturalism is basically needs for example needs to be understood in terms of how it erodes the spontaneous image of human as being you know the the

product of the divine or and also it challenges you know the given it challenges in self narratives of man there are basically stemmed from and a a non-naturalized image of human being. The same thing about Copernicus, the spontaneous image of our place in the universe via basically a terrestrial image

and a spontaneous terrestrial image. And that's what Copernicus challenges, Kepler challenges, Newton challenges, so on and so forth. Now, if science sees or more accurately explains the image of the human in the universe from the perspective of the intelligibility of natural history, and if this intelligibility contravenes what we deem as intelligible from the perspective of our phenomenal self, even at its most basic level, i.e. the aperceptive unity of the I which thinks, then thinking can no longer corroborate its intelligibility by resorting to what human appears to himself or his manifest self-conception. Nor can it sustain its will to think by holding back the traumatic imports of the will to know, sublimated, and scientific rationality.

So in a sense, the intelligibility of the non-manifest that coincides with the intelligibility of natural history infringes on the intelligibility of our basically phenomenal self through which we make self-narratives of our world according to spontaneous narratives of the world.

But at the same time, so if the intelligibility of the non-manifest infringes on the intelligibility, the supposed intelligibility of our phenomenal self and any kind of self-narrative yielded from the perspective of this phenomenal self, thinking cannot basically justify or corroborate its intelligibility without integrating the

traumatic effects of scientific rationality within its own will to think. Moreover, the intelligibility of thought and its volitional compass that governs intentional action can no longer maintain the intelligibility of purpose solely by responding to interests and needs drawn from what human takes himself to be manifestly and is spurring what science says what he really is. So, in a sense, the intelligibility of the non-manifest,

namely the intelligibility of a world existing independently of how we think about it, is in a sense affecting the intelligibility of our practices. Because our practices are indexes of intentional action. And intentional action, as we talked about a little bit via that diagram, the adjoins of intelligence, Szilard's entry, exit, moves.

Intentional action needs to be understood in terms of the relation, a designated intentional volitional relation between thinking, the domain of thinking, and world. The good for man is a recipe for the convergence of, on the one hand, practical intelligibilities or practices according to different kinds of good, qua intelligible purposes, and theoretical intelligibilities concerning the knowledge of what man is in the universe and the other. But in so far as science offers an alternative image of man that outstrips what man thinks he is,

practical intelligibilities aimed at crafting instrumentalities as well as satisfying lives which is basically Plato's definition of good according to different kinds of good cannot remain intelligibly purposeful unless they can be joined to the ever-deepening intelligibility of this alternative image which has been and been extracted by science. The scientific image, so the third point is that the scientific image calls for a transformation in the various structure of thinking.

Since complexities and dynamic tendencies of the world cannot be approached by manifest resources and their generalizing rules or inductive generalizations, This was the point that we talked about in the last session. Then thought must undergo a de-stabilizing revolution in its own internal configuration. So just as, you know, in order for thought to be able to detect, track, and act on dynamic tendencies of a world existing independently of thought,

But it needs to design or devise its own dynamic tendencies. And this is something that I will talk about when I'm talking about manifest image, that these dynamic tendencies are not essentially, are not isomorphic to dynamic tendencies of the world. But there is, and that's, you know, Selaar's theory of picturing, that while we have the coherence, coherentism of rules of a given language, assertability is only adjudicated by the rules within the language, namely inferences,

differences, but also at some basic levels, the units, the picturing units of language are causally mapped to the non-linguistic objects, namely the objects of world. And at this end, the good language, basically the criteria of a good language, is a language that is capable of correct picturing, namely, form its, in French, your rules according to these causal mapping between a linguistic object and a non-linguistic object. And this is something that I will get back to it later. That is, thought must conceive an internal

dynamism capable of tracking the dynamic complexities of the world. In order to render intelligible the catastrophe of the non-manifest catastrophe here I strictly you know talk about it in terms of the kind of catastrophe that you know catastrophe theorists talk about for example runathon it's an index of you you know, singular change, quote discontinuity, that introduces basically a radical morphological change.

And this is what really the non-manifest is. The non-manifest is a catastrophe for thought in so far as it forces thought to change its very morphology it's very architecture in order to render intelligible the catastrophe of the non-manifest thought must devise catastrophes of its own not only in terms of inferences and conceptual activities capable of de-stabilizing its doxastic conservatism breaking away from its belief entrenchment and dynamically manipulating the parameters of its rules but also in terms of how thinking can relate to the world through action. The effect of catastrophe of non-manifest

for the mind is that of a cascading fallout leading to a noetic catastrophe. If the mind is conceived as the edifice of cognition with entry and exit doors, this was Wilfeld's Selaar's picture of thinking and mind or two gates which is a picture of mind presented by Chinese philosopher Mo Jung San connecting the neurobiological domain of cognition to the social domain of cognition so in Selaar's basically we talked

about that, you know, the structure of thinking is basically the output of world regularities is being mapped or being transferred to the input of language. then we have an intralinguistic domain which is the domain of inferences or rules of language of a given language with the language simply being understood as a symbolic repertoire then the output of the language

becomes again the input of the world And so when the world to language transition, what SLRs identify with perceptual takings and perception, then intralinguistic transitions are defined as the realm of movements between assertions and inferences. And then the exit level, exiting from the domain of the language to the world, for Salars, according to basically these rules, for Salars constitutes the domain of intentional action.

Mo Jong-sen is the same thing. It's what he calls a two-tiered ontology of the mind. This is something that I will get back to in our second module. A mind with a gate opening inside and a mind with a gate opening outside, one opening inside. One to the basically traditionally understood cognitive mind, which is basically constituted by neurobiological situation. And the other one is basically mind as a social edifice, which is what really the definition of mind is. so if the mind is conceived

as the edifice of cognition with entry and exit doors or two gates connecting the neurobiological domain of cognition to the social domain of cognition then the intelligibility of the non-manifest simultaneously uncovers the natural history of the former and incites revolutions in the rules and purposive actions of the latter now a mind that has access to the natural history of its constitution and is also proficient in modifying its rules and crafting its purpose purpose of actions as an adaptation to the order of intelligibility is both theoretical and practical is a mind capable of self realizing

itself, or a mind capable of self-realization. But insofar as self-realization is tantamount to re-engineering the existing reality of the self, and the reconstitution of its natural history by different sets of realizers, a self-realizing mind gestures toward the multiple realizability of the mind. This is basically the central claim of functionalist thesis. And functionalism, when we are talking about functionalism, and something that I get back hopefully again in next module, is we not only cover functionalism as being understood

as functionalism of 20th century with its various kinds, normative functionalism, mechanistic functionalism, computational functionalism, so on and so forth. But functionalism in the sense that basically coheres the philosophical project. A functionalism that basically is the continuation of preneal questions of philosophy. and has been around since Plato, since the Stoics, since the Confucianists,

to, for example, Kant, Hegel, you know, Sellars, Brandom, Turing, so on and so forth. It's the understanding of basically kind of a brief definition of a function in this sense is that it's or functional explanation or functional rule is that a functional explanation of an item is about is in terms of not what a thing is but what a thing does. Namely, it tells us that what a thing is is not really about its, you know, its thingness, but about its activities.

Namely, the definition of what an item is cannot be independently conceived from what that item does. Now, functionalism can be, you know, divided to, you know, a whole broad spectrum of varieties of functional explanation and functional roles. But basically like a rough summary of these functional basically rules are the ones that functions are understood in terms of the metaphysics of causation.

and then they can be discussed in terms of mechanistic functionalism, strongly mechanistic functionalism, or normative but mechanistically constrained or materially constrained functions. And also the third variety would be a strongly normative. For example, Salars is a functionalist philosopher on two levels, both at the level of strong normative functions in which concepts have a status. And this status is really their functional roles.

And dysfunctional rules can be explained no longer in terms of causation, but basically only through rules, inferences. So it kind of circumvents the metaphysics of causation and basically rather still the shadow of teleological implications of functionalism. But he's also functionalist at the level that for him normative functioning is also materially constrained or multiply constrained.

or it's constrained by how basically mechanism and processes organize this, for example, a normative function at different levels of basically its material substrate. And now another thing about functions is that when we are talking about functions, we are talking about the realization of function and this realization of the function, quo an activity or behavior of the system or what a system does, can be understood in terms of how a set of realizers organize this behavior,

namely lead to this realization. Now, we have for every function from a mechanistic, from a functional perspective, we can have different sets of realizers. But of course, these different sets of realizers can be, you know, in a very complex way need to be ordered. This is something that I don't want to get into. to. But for every realization, for every realization of a specific function, there are different sets of realizers. And as long as a set of realizer has equal powers and capacities to

yield or to lead into a specific realization we can understand that realization in terms of basically that's that set up and that's didn't in terms of and that set of realizers so this allows us this understanding of realization in terms of set of realizers allows us to imagine realization of function in terms of its multiple realizability namely its instantiation its generation by way of different realizers which might not belong and

this is usually the case which might not belong to or have anything in common qualitatively or quantitatively with those realizers that currently generate that function. So, to continue. But insofar as self-realization is tantamount to re-engineering the existing reality of the self and the reconstitution of its natural history by different sets of realizers, A self-realizing mind gestures toward the multiple realizability of the mind. A multiple realizable mind that is also capable of re-adapting the intelligibility of its actions to the intelligibility of alternative realizers.

An alternative image of the human as a complex system whose functions can be realized by alternative entities or systems is in reality a mind that not only redefines its meaning in the sense of how it can be realized and implemented in different contexts, but also one that repurposes itself and its actions according to these meanings. Artificiality is the truth of the mind set in motion by the order of intelligibilities. For what is artificial other than repurposing what is naturally given without violating natural laws

and the ability to be realized in an alternative organization or complex to the one that currently constitutes it, that is to say, realization in the artifactual. So, two things. You see, functionalism also, So a thesis, once you really ramify its consequences, you see that it is also a thesis about meaning and purpose. of a simple way to put it, it is very much similar, you know, to the Witkinner Science

thesis that meaning is determined through use, you know, the kind of the core assumption of pragmatic. And in fact, pragmatic functionalism is really at the core of functionalist program. In the sense that once mind is realized in different context, namely by different sets of realizers, its meaning also changes. And this is kind of a pragmatic implication that there is no such thing as a given meaning. Meaning is produced through use. And use here is simply the implementation and embedding of the mind within different context of use and practices. Now also the change in meaning of the mind also coincides

with change in the purpose of the mind. So a mind that is capable of self-realizing itself is not only a mind that is capable of changing its meanings according to realization in different contexts and by different set of realizers but also a mind that is capable of repurposing itself to the extent that it exceeds and outstrips what it was supposed to be, what it was supposed to mean and what it was supposed to do.

The cascading effects of the catastrophe of the non-manifest culminates in the functional evolution of the mind as a self-realizing geist. For this reason, the intelligibility of such catastrophe should be seen as an explosion of enablement, both in the order of self-conception and in the order of self-transformation. In the wake of the Copernican revolution, the catastrophe of the non-manifest is what sparks and fuels a noetic deracination and drift, a noetic uprooting and a noetic drift. The deracination of thought and its noetic drift is commensurate

with what Plato calls the form of good as the form of forms, since it sets the scaffolding for conceiving the realm of intelligibilities as a complex system of recipes for crafting a world which includes not only instrumentalities, but satisfying lives. But what kind of life would really satisfy us other than one that involves a self-knowledge which has passed through all the stages of disciplined reflection on the nature of things, that is to say, their intelligibility. And that really needs to, the import of the scientific image needs to be understood in terms of unbinding different stages of reflection on the nature of things.

A philosophy that snobs the multi-staged reflection on the intelligibility of things and the implications of what intelligibility can do to the mind is the very definition of a mindless contemplation. It is a thought whose impotency insinuates the expiration of the intelligibility of its purpose. So these were the immediate, albeit roughly sketched, ramifications of the order of intelligibility that theoretical sciences unbind.

Now, inspecting the ramification of the scientific image as the excavation of new order of intelligibilities concerning the non-manifest world, we see ourselves faced with a number of puzzling questions. so these are the kind of questions that we try to answer throughout basically this whole new rationalism series our current module and the next one can the intelligibility of a nature outside of the manifest domain an intelligibility that coincides with the epistemic alienation of how human appears to himself

be coherently integrated with the intelligibility of human practice understood as a self-instituted purpose. Or let's phrase the question more classically, if not more ambitiously. Can what is not what human thinks of himself, that is, a partial conception of human as a complex impersonal system, be reintegrated with the pursuit of the good, not only good as intelligible instrumentalities but also as satisfying lives. You see, this is a rather paradoxical question charged with tensions in that to most, if not all of us, a satisfying life must also respond to our psychological needs.

But the scientific explanation of the human, particularly in light of neuroscience, damages the very world we have created out of our manifest image to fulfill our psychological needs. If science, now the second question is that if science sharpens the asymmetry between what the world really is and what it appears to us, then how can we respond to this world in any intelligible way through a regime of actions that is not confined to a conception of ourselves based on how the world appears to us. The third question would be, what human really is

becomes not only commensurate with, but also the impetus of humans ought. if basically the intelligibility, okay, let's reformulate this a little bit. You see, if our normative arts are by default, not by default, are initially in line with the intelligibility of how we appear to ourselves and how the world is appearing to us, quote a spontaneous image of man and the world,

then can this normative out be reconciled with a different order of intelligibility that is basically ever-growing and also at the same time shatters the purported, the apparent intelligibility of our spontaneous image. my contention is that the answer to these questions is doubly yes in that aside from the possibility of these couplings and combinations these are the definitions of what good for human is and consists in but to elaborate both the possibility and the necessity of this positive answer

first we must accurately carve the problem at its joints and remove the seemingly paradoxical nature of these questions that stemmed from the conflation of different concepts, elision of different explanatory levels and problems. While the first module, our current module, points to the possibility of these goals, The second one, with its emphasis on self-transformation or what it means to do something with time, elaborates the necessity of such goals. Now, as a kind of a flash forward to the second module, Reason and Time, or, you know, addressing the question of self-transformation in accordance to a robust account of self-conception.

and also in order to highlight the orientation of what we have discussed so far and what we will discuss, you know, all the stuff that we have covered and basically where we are headed, the crux of these presentations is one of simplicity or perhaps triviality. Yet its triviality, its encapsulation into a simple form, should not occlude its deep ramifications. So this is basically the entire goal, rather trivial, but as I said, this triviality shouldn't obviate its importance.

This is the entire goal, and it would be useful to set it out so we know basically how we are diversifying the arguments and then cohering them ultimately to reach to this thesis. Thesis. The form of good, understood as the form of forms, is the augmentation of the order of intelligibilities of things and practices. It is an answer to the pre-neal questions of what to think and what to do, that manifests as the self-cultivation of that which is intelligible, and the self-realization of intelligence as that which knows what to do with the intelligible.

the apotheosis of the good resides in the advent of autonomous self-cultivation characterized by Chinese philosopher Mo Jung San as the practical elaboration of the death of God as an intelligibility that points to the possibility of humanity attaining the highest good and fulfilling the principle of samon bonum So, in a sense, as we talked about in terms of the tasks, the task of, you know, the projects of rationality in the sense of coming up with a robust account of self-conception, you know,

elaborating the relation between self-conception and self-transformation, augmenting the relation between the two from one that is necessary to one that is sufficient and also be capable of revising self-conception according to how and to what we are transformed culminates in basically a new understanding or functional understanding of the relation between truth and goodness from a classical philosophical perspective.

And this is what basically at the end of our second module we talk about this in terms of philosophies of self-transformation and specifically Chinese New Confucian philosophy for which The amplification of the relation between truth and goodness from one that is necessary to one that is sufficient and augmentative coincides with a new conception of humanity that is capable of attaining the highest good,

or basically fulfilling the principle of someone bonum. So this was basically I just wanted to tie the implications of our previous session regarding the scientific excavation of the non-manifest and tying it to basically the overall argument and thesis of the seminar. Reza, we have a question from Aaron who is an enrolled student, but because of family obligations could not be in a class and he's watching it on YouTube.

I posted a question on the site. It starts with, could Reza comment on the multiple realizability and virtual as considered by the lander slash the loose? I don't know if you want to take it up now and answer it or later, but I just thought to bring it to your time. Okay, okay. Can you just... It's on the sidebar. I posted it on the sidebar. Okay. Function, function... Just at the very, very end. It says, could he comment on the multiple realisability? Well, I mean, I don't know about the virtual, but yes.

You see, okay, I mean, first of all, my knowledge of Deleuze is quite rusty. And I have read Delanda, but I can't say that really I'm any knowledgeable in the work of Delanda. But from a kind of a very classical, not Deleuzian sense, a very classical understanding of virtuality, reality, actuality, order, which is a kind of scholastic definition, medieval definition, is that the actualization, namely the realization of a function,

needs to be understood in terms of powers or capacities that effectuate a certain function or certain activity. Now these powers or capacities, corealizers, are distributed across really broad spectrum of complexities and organization. So basically these powers are qualitatively different.

And in order for a function to be realized, what basically we need to do is to not only address or identify these powers, but also be capable of mapping out the organization of a functional organization or a hierarchical organization in which these qualitatively different powers

can orchestrate basically a single surface function, basically an actual manifestation of the function. Now, this is what they call dimensionally varied and multiply constrained account of function. Dimensionally varied in so far as powers that lead to the realization of a function are distributed across different dimensions, varied dimensions, qualitatively, structurally different, you know, hierarchies. And also multiply constrained because each power, as coupled with a distinct structure,

enforces a specific constraint upon the realization of a function. So in this sense, I mean, from, again, as I said, classical sense of virtuality qua powers, we need to understand that the actuality of the manifestation, actuality of realization, is really, can only be executed via mapping out the order of virtualities, the order of powers, with the understanding that the order of powers is not a totalized order, but one in which powers are qualitatively different, namely the order of virtualities,

Virtuality is being qualitatively different powers. So this is a kind of, again, as I said, this is a very classical reformulation of this multiple realizability thesis according to, again, a classical definition, an old philosophical definition of virtuality, actuality, or virtuality and realization. So, yes, that was it. Thank you. More questions before we move to our next discussion. Go ahead, Joshua.

here I'll be seeing myself shortly just gonna read I'm me I guess my my kind of in in kind of the how do I want to put this at start again arm so where it's like s so large like differs you know in between the sort of for so large there is a sort of normative functionality in the causal functionality and the normative functionality is... No, norm... One second before we move forward. For CELAR's, the normative function can also be detected within the causal function.

In fact, causal function, you can't really detect purely causal function. Causal functions are basically... Function is really about how a system ought to behave. Yeah. So this is the detection of, so even at the causal level there is a level of normativity. But the normativity at the level of causal mechanistic function is normativity that's ultimately metaphysical because it's reference to the causal system. Whereas normativity at the level of conceptual role function is one that is basically inferential, is not metaphysical anymore. Well, what I was gonna, I guess, building on that sort of idea maybe is our sort of

like capacities, our sort of like normative capacities I suppose that we have, you know, as maybe flow from sort of this Kantian notion of the sort of unity of aperception and that sort of thing. That, like, I mean, going back maybe to the O'Shea paper and how he discusses how the sort of material layers that undergird that sort of, like, unity of aperception. and sort of he points how sellers then build a sort of naturalistic explanation underneath that for the sort of like causal and materiality constraints as to how that is formed, right?

If I'm kind of repeating this somewhat correctly. No, yeah, sure. I mean, well, the whole thing is, I mean, as we talked about in the first session, you know, So really what we call, you see, it's also a good example for those people who are talking about like, you know, the AI cynics, basically. AI cynics, they say that a machine cannot think. Even what it actually does is thinking, namely realizing the function of thinking. It's not really the definition of what thinking is. But it is really the consequence of two bad arguments.

Basically, the argument itself is flawed. One is that at a functional level, it's stupid to differentiate between flight, realization of the function, flight between an airplane and a bird. Saying that, oh, well, an airplane doesn't fly in the sense that bird flies. The comparison is just from a functional standpoint is null and point. But also at the level of that what we see as what thinking is, trying to basically hold the ineffability of thinking within that whatness of thought,

the ontology of thought, that's also problematic in the sense that that very whatness of thought that we think that the machine is not able to think like that, is really the product of unity of a perception. It's basically the intricate consequence of transcendental psychology. In the sense, and we talked about that this is basically self, phenomenal self, a perceptive unity, is really a computational model. is a computational model that can be also explained in terms of computational terms. Hence, it's self-realizable in terms of computational functions.

I guess maybe I guess kind of what I'm, yeah, I mean, I don't think I disagree with you on that in any sense, but I guess I'm curious if through gaining some normative description, I guess, of how these things like functionally operate, and then through that gaining some sort of capacity to act upon the material level of that substrate, if doing so would in any way then ramify the kind of like normative results of changing the sort of material constitution of that particular thing.

Yes, and this is what Salars makes that cryptic passage that ultimately man is what really science says he is. And the scientific image replaces ultimately the manifest image. In the sense that particular traits of manifest image, and we're talking about when we are talking about manifest image next session, the particular traits of the manifest image, namely those traits that are attributed to person, and let's elaborate what person means for CELARS. Person for CELARS is not specific to a species being,

human in the sense of a species being, but merely in terms of capable of yielding a specific function, thinking, or deploying concepts, cognition, by way of those linguistic transitions, entry and exit. Now the thing is that what's really the core of a person is the logical irreducibility of its function. The logical irreducibility of its function is commensurate with the causal reducibility to its material substrate. So the scientific image is ultimately about causal reducibility.

Causal reducibility can be joined but not contravenes, not replace the logical irreducibility. Thank you. I think, yeah. And the thing is that the intelligibility of practice is a two-tier intelligibility in the sense that both, as Plato says, it's a kind of a multi-stage reflection on the nature of things, namely the intelligibility of things in terms of their causal reducibility, so on and so forth. that's what the scientific image is

but also these things are also when he says things these things also include our instrumentalities and arts so it's both intelligibility at the level of the logical intelligibility and intelligibility at the level of intelligibility of things scientific things Another question. So the intelligibility of practice ultimately becomes the integration, the adequate integration of intelligibility of things as such and intelligibility of ought, the logical irreducibility of ought

that needs to be basically synchronized with these intelligibilities, with the intelligibility of things. This maybe leads me to another kind of question. I mean, I guess insofar as our sort of normative oughts allow us to take action on the kind of things that structure those normative capabilities. capabilities. And this is pretty speculative, I guess, but I mean if then we can kind of alter that material

substrate such to allow for further normative capacities, but because that altered substrate is itself still materially constrained in distinct ways, don't you simply then have like an expansion of the manifest image because you might get to a certain point in which you get a sort of, because it's still finite, make it to a point where you still have, you can then develop a broader conception of the scientific image but that sort of normative conception still is in some ways relative to a more developed scientific conception?

I think, well, I think to an extent the answer to this question is yes and no. Yes in the sense that the growing of scientific... You're buzzing. You've gone back to the robot voice. Okay, okay. Let me reconnect. Joshua, you're the face right now, just so you know. Oh, lucky me. Don't put your finger in your nose. I really hope that other people also join in and ask questions.

I have questions too, but because I'm not a registered student, I'm just running the technical side of it. I don't want to jump in and dominate the conversation, but please join in and ask questions because it was a very intriguing lecture so far. It has been. Am I clear? Yes. Yes. Okay. So I was saying that the answer to your question is both yes and no. Yes in the sense that the expansion of intelligibilities of things, the domain of non-manifest carried out by modern sciences is basically both the rate of it surpasses far the efficiency of

of our normative realizations, but also in terms of horizon, also basically exceeds the normative capacities. So this is, yes, from this perspective, I can say a relative yes. But no, to the extent that these intelligibilities regarding new expanses, new expanses of non-manifest,

doesn't translate to the manifest. Manifest image, as we talked about, we need to distinguish what we mean manifest. Manifest is also a broad image. And then you need to basically carve it as joints in order to be able to solve its problems. One apart dimension of the manifest image is the manifest as being literally manifest, as being basically the domain of the manifest perceptible and its generalizing rules that we talked about last session. But also there is another dimension of the manifest, which is the logical irreducibility of conceptual activities,

or, you know, the category of personhood according to CELARS. Now, the thing is that this simply uncovering new domains of the non-manifest doesn't mean that it's being translated to the first dimension of the manifest image, namely the manifest perceptible. This is important. It only translates and it only needs to be taken in relation with the second dimension of the manifest image, namely the logical irreducibility of conceptual activities. At this level, there is nothing that makes this conjoining process of the intelligibility

of the non-manifest and the logical irreducibility, qua intelligibility of normative realm problematic. But yes, this becomes insofar as the rate at which and the both at the terms of rate of growth, rate of expansion, and also the range of expansion of these intelligibilities, we do not have any kind of even indication of any kind of robust project of norms, of instantiation of norms, that allow us to tap into the power

of these intelligibilities. And this is exactly what I was trying to say, that for people like Hegel and Johnson, the ultimate aim is really to be able to make a sufficient and augmentative link between intelligibilities of things or intelligibility of the world and intelligibility of purposes or practical intelligibilities. This becomes basically the ambition. Obviously, at this point in time, not only we do not have this, but everything that we have

shows actually we are basically our normative intelligibilities, our practical intelligibilities regressing in fact in comparison with our scientific intelligibilities or intelligibilities of the world. So it's basically the autism of the manifest at this point. I have a question. Could you talk a little bit more? This actually is something you just mentioned, but I didn't quite understand. Could you talk a little bit more about the relationship of the perceptible and the normative, those two dimensions within the manifest image, or like maybe directly to a resource, like

a, or a particularly useful text in outlining those different, those, the relationship of dimensions? Sure. Were you present last session, previous session? I wasn't present, but I watched the, I watched the rerun. OK. So basically, so you know what the manifest perceptible is, the manifest perceptible dimension of the manifest image, which surrounds the constancy of the accurate sensible properties and how we basically make rules out of these accurate sensible properties. Oh, OK. Yes. Are there the level of generalized observations or observable generalities?

Yes, observable generalities that concern with manifestly perceptible observations or observables. So we have that one. And the other one, a dimension of the manifest image, is the dimension of conceptual activities, namely the the inter-linguistic transitions the moves between assertions and inferences and at this level basically there is the question of logical irreducibility the question that conceptual function shouldn't be understood in terms of metaphysical function.

And this is something that I will get back to next session when I'm talking about manifest image and theory of picturing, which ultimately, you see, we know that, for example, concepts of mathematical languages are capable of mapping the objectivity in mathematics to the real objects. And they make this quite effectively. This is what makes, constitutes the reasonable effectivity of mathematics, or according to some unreasonable effectivity of mathematics.

But the thing is that when we are talking about languages from an epistemic perspective is that none of these theses about effectivities of mathematics, effectivity of a certain epistemological framework, do not say how is that we can gain traction upon reality in the first place by way of our linguistic resources. You know, there are different theses about, for example, how objectivities of a language are mapped onto the real objects or gain traction upon the real.

But these are all based on, at some level, when you really go and unravel their definitions, At some point, they stop and really take the issue as some sort of intuitive dimension. For example, a structural realism really delegates this question to the level of structures, as if we need to believe, have an intuitive belief that basically transfers of structures make, enable these kinds of projected mapping between linguistic objects and non-linguistic objects. Or for example, for a philosopher like Badiou, it is really the ontology of mathematics that allows for these. But really, none of these, if you really look into them,

can break that intuitive belief about, you know, what is really that allows us to gain traction on reality from inside a language, especially if we are going to define language in a coherentistic framework, in the sense that assertibility inside a language, the criteria of true and false, can only be adjudicated by the rules of a given language, namely inferences. If this is the case, then how really, as I talked to Jordan about this last session,

that how can different competing languages can express their adequacy, which one is more adequate to, for example, express a certain physical phenomenon, if that's the case? are we going to abide by, for example, how they respond to evidences? But responding to evidence, again, doesn't really address the fundamental conceptual question that basically how can we really relate language as a frame of determination onto reality? And corresponding

with, for example, evidences, observations, so on and so forth. And then I have one other question, and it's really too broad, but it would be helpful if you give me some pointers about how the normative might relate to the non-manifest, because it seems to me like you've talked about how there's a normative that I mentioned to the manifest image, but you're also saying that the scientific image when fused with the manifest image would produce more robust norms. And so I'm just wondering exactly, I mean, maybe that's what the coming lectures are going to be about. Yes, it will be. But the thing is that, you see, we talked about that, you know, it's simple as this.

It's almost like too simple. This idea that what we ought to do and respectively what can adjudicate what matters is decided by the intelligibility of things in the broadest possible sense. Now, the thing is that our oughts, if they are simply tethered to the apparent intelligibility of the manifest image, these oughts are inadequate.

Basically, as scientific intelligibility overstrips the resources of the manifest image, these odds that are in correspondence with the manifest image increasingly prove to be inadequate. Now, the thing is that, so in this sense, the ought needs to be, if ought belongs to the order of intelligibilities, namely intelligibility both theoretical and practical, this ought needs to be commensurated with intelligibilities of the non-manifest as well. with the understanding that the intelligibility of the non-manifest violates the spontaneous intelligibility of the manifest image.

Reza, can I ask a question? Sure. OK, so at some point, you talked about how scientific image essentially replaces the manifest image, right? This is what Selaar says. Yes, Selaar says that ultimately, the scientific image replaces the manifest image. But does it really replace it or absorbs it and colonizes it? No, it can't absorb it. I mean, if scientific image absorbs manifest image, it simply does it according to the odds of the manifest image.

That would be more like the utilization of the scientific image in the service of the manifest image, as we see, for example, currently in no liberalist agenda. No, for sure. No, that's not what I'm talking about. I'm talking about the opposite of that because, you know, from my understanding, and I mean, I will wait for the next session. It doesn't absorb those superfluous components of the manifest image because, as I talked about, and this was also the last previous session, is that the scientific image really needs to be understood as an index of intelligibility that overrides the apparent sufficient

intelligibility of the manifest so it doesn't really absorb it in shatter the pieces I'm talking in the literal sense and when I'm when I'm when I'm talking about manifest image you're actually maybe I'm not even talking about the last name category of manifest image but but basically for basically the category of perceptible or sensual which is physical and and it basically it's the only way to to connect and to communicate the content of the scientific image back you know that's not true really that's that's I mean the whole previous session was that precisely that's just a flawed epistemic reconstruction of scientific theories that science doesn't do like

this sign doesn't really work by way of inductive generalizations at the level of perceptible this was basically this is exactly what science isn't because the money because the because the domain of the manifest perceptible is the domain of accurate sensible properties and these sensible properties themselves need to explain how they are conceived out of basically the organization of imperceptible non-manifest entities their structure in fact is accidental and

contingent upon how they are orchestrated by way of swarm of basically imperceptible non-manifest entities that theoretical science tries to excavate. Thank you. We talked about this in terms of, for example, a cube. You see, those properties of the cube, namely, for example, color and its shape, its facing shape to an observer, they remain the same for different observers. roughly the same. But the explanation of what the cube is, basically what its geometry is,

that is anything that is outside of basically this realm of the manifest perceptible, the definition of the cube, the observation of the cube changes for different observers according to the conceptual framework by which they are approaching and they are observing this cube. So observation according to accurate and sensible properties of the manifest perceptible basically yields intelligibilities of the same class. What in fact, what science does is that it really, you know,

it proliferates and it expands the realm of intelligibilities rather than simply, you know, limiting these intelligibilities to the intelligibilities of manifest perceptible. You say limit, but maybe I'm not communicating properly, but on a very like, I'm not sure if superficial is the right word, but on a very physical and literal way, things have to return to the realm of perception for them to be understood.

I mean, you're still like, you're making sound out of your mouth for me to understand what I'm saying. That's the level I'm talking about. Yes, yes, I know what you're talking about. You are simply talking that as long as science doesn't have observations, it's not really science. And science are about... I don't even care about the observation side. I'm talking about the second window, not the window to the mind, but window to the outside, the sociable window, right? So that window somehow involves the manifest, somehow involves the problem and limitations of manifest, manifest image, no? No, yes, it has, it involves the problems of the manifest, but not the manifest perceptible. It's the manifest of conceptual activities. the second dimension of the manifest image well the thing is I come from the

art world so for us it's like so it's like that the the image kinda because the art has anything to do with science I definitely but what but you see what I mean the problem becomes the problem all about you know art art claim that it has a language and has a grammar right which which we can totally question and problematize, right? And art tries to compete with language or claim its own territory as a visual or not even visual, but a different type of language with its own vocabulary and its own grammar, right? SPEAKER 1,000 repertoire. But a symbolic repertoire, and that's what Josh's assignment was, he posted

on the classroom page. But a symbolic repertoire of art that is simply exclusive to the manifest perceptible is pretty misguided. And that really explains a lot about contemporary art and its fear of things like determinant concept, conceptualization, coherent conceptualization, and also normative determinations. I'm going to have to go and read. What is being exercised by the art doesn't essentially mean that it's true. It simply can be also conceived as a misguided procedure. others have more questions

or or it simply marks or it simply marks you know the the essential difference between art and science you know and yes if that yes I can agree that this is what art does but this cannot be translate to basically I what really ultimate frame of communication is and should be well so I I do have a question and it's basically building on Mohammed's question which is I I it is I hadn't thought of it before but it is a little bit confusing to me how what maybe how to think about communication

in the context of rationalism. And you know I can imagine for example one idea of communication in which everything is reduced to empirical perceptibles. And that would be for example, and that would be when people talk about ideas as infectious or something, right? Sometimes they think of something as ideas having like literal physical force or something. Yes. And so I'm curious what the scientific idea of that might be like what how how are you communicating to us in some way that's not and that's so okay your communications to us are causally reducible in what way are they logically irreducible what is at that mean is that is that a lot you call your reduce simply means that

a come language personal language doesn't communicate language is the medium of assertion and inferences You see, this is something that I will talk about in detail when I'm talking about the manifest image, in terms of that the words don't have meanings. The words are assertions. We assert. We don't express meanings or communicate these meanings a priori. If we are buying into that, we are simply again buying into some sort of epistemic given. So when we are talking about logical irreducibility, it's the logical irreducibility of the moves between assertions and inferences that constitute the meaning, meaning in the pragmatic sense.

Now, communication, how can communication, well, this becomes, yes, this becomes what Falar's, the whole, you know, the whole rationalist project and also, you know, the whole project of Solar's gestures toward this, that how can we conceive a language whose abstract resources are more complex pictures of reality, rather than being simply the pictures of, for example, of the manifest perceptibles. And this is something that, yes, this is, as you see, in fact, in the advent of the

the entire domain of cultural civilization, how short it might be, this communication or this framework of expression of language has changed quite significantly. Yes, that's the whole point. point is that we are using these you know metaphors about manifest perceptible when we are trying to communicate but simply if we are using them meaning that are they the most are they are they the best way to communicate or also are they

really as they think they are are they the most optimal ways of communication The answer to both questions is no, in my opinion. Okay, I have one last question. What is the conception? The very last one. Okay, maybe someone else should do it. Let's see if somebody else wants to actually join the conversation because I don't want it to be dominated by you and me and maybe like poor Joshua. Anybody else has a question? What I wanted to say, I'm not yelling, right?

No, no, not at all. I don't know if it's the volume of my voice. Okay. So in regard to what Mo said before, I was thinking of simulation or representation as a way of entry inside the intelligible. I mean, I was trying the other day this Oculus Rift thing that's like virtual reality mask. So I was simulated into some feelings and some conceptions about things that I hadn't experienced before. Let's say a monster chasing me or flying off a plane and dropping my body off and such. So this kind of representation, I mean, if we could somehow represent what we also find,

then this could be a way of rendering intelligible. And then the other thing I wanted to say, this whole catastrophe that thought must undergo was quite interesting. And I was thinking of somehow a paradox that's in neurology, somehow like an esoteric transaction of the manifest image, that's how I see it. It's the ghost limb, you know, when you lose an arm or a leg, and you see that you have no arm or leg, and you feel that you're moving it. You're moving your ghost limb. deviant models of self. Yes. Yeah, this was somehow interesting.

As you know, the body has gone through a trauma and a catastrophe, and it cannot change the way it works because of this trauma. It cannot accept that it has lost a part of it. Yes. but it's because precisely it's because the body is still working under the you know computational constraints of the phenomenal self yes that's exactly what happens if basically your I mean a very metaphor for this it happens if what we talked about with Josh and Mo in terms of you know practical intelligibilities, odds, that are not commiserated or synchronized with the

intelligibilities of things, multi-stage intelligibility of things, which part of it is really the scientific image, the intelligibility of the non-manifest. Yes, that's exactly what happens if you don't, are incapable of synchronizing your practical intelligibilities with theoretical intelligibilities. You simply create a simulation or a deviant basically alternative of the order of intelligibility that simply serves your supposed intelligibility

at the level of the manifest. Exactly like your deviant phenomenal self model that you are talking about, the phantom limbs and so many of these things. Yes. But in fact, the deviant ones, the deviant models, neuroscience, and this is what some of the stuff that Metzinger talks about, that these in fact point to the cracks and ruptures of the phenomenal self. namely the self of the manifest. Now, okay, now back to your question about simulation and representation. Yes, simulation, I mean, well, simulation is a broad category.

Simulation, you know, we have philosophical simulations, we have scientific simulations, We have simulations within, again, semantic of utility of the kind that you are talking about. These need to be differentiated. But, for example, you know, from the specific point of view, computer games that you're talking about, yes, I think these need to be understood as basically they alter both, you know, phenomenological experience and also in so far as these represented environment extend the domain of cognition they basically function

exactly like in the sense that you know for example theories of extended cognition talk about they they enrich the resources of cognition the representational resources of cognition, how basically new patterns can emerge through these interactions with environments on and so forth. Yes, I think from this perspective, yes, these are, these representations can enrich, you know, certain aspects of cognition, the extended aspects, extended dimensions of cognition within, again, enriching of the representational resources of cognition and how you can manipulate these representations but there are also simulations for example in computation there are intrinsic

simulations simply the model is embedded in and this is what I'm going to talk about model is being embedded in the processes of a system basically it the architecture of the simulation reflects the architecture of the process of the system itself and these are these need to be taken in you know in a completely different sense of simulation and are also of course you know the whole genealogy of philosophical simulations from Plato's you know cave allegory to you know like Nietzschean monads so on so forth and And these are again, again, facilitate between simulations that work between their, that

they can be, this simulation can be categorized by virtue of their emphasis on representational pool of simulation or conceptual pool of simulation. how basically it can set in motion a new conceptual framework according to how it manipulates representations. Which again, comes back to the geometry. That's what geometry does. So Reza, do you want to continue on? Or should we just go with the discussion format? How much time do we have? We have about 45 minutes.

Can we make it 50? Yes, of course. OK, so I just want to kind of go over some of this stuff very quickly. You see, when we are talking about, and this is what we talked about last session, that when we talk about the structure of reality, We are talking about issues like emergence, complexity, how something new emerges, how can we detect complexity, how can we quantify them, how can we detect the structures in the first place in order to render them intelligible, so on and so forth. And we talked about, for example, that, you know, there are how we define a structure at

the level of concept, namely the objectivity, at the level of objectivity of a given language, for example, mathematics, can give us new insights, new novel insights into the actual structure, the actual intelligibility of, for example, a given object in reality. But the thing is that two issues remain that need to be answered. One is that, exactly as I said, these two issues, one is the issue of emergence,

intrinsic emergence or radically novel emergence. If something new is happening in the system, if something new emerges in the system, how can we detect it? How can we quantify it if we haven't developed so far a map for it at the level of the concept or framework of objectivity? This, and also how can we basically explain and measure complexity? In fact, what is exactly complexity in nature or in physical reality?

Both these questions, science approached them via the problem of modeling. A fundamental point is that any act of modeling makes a distinction between data that is accounted for, the ordered part, and the data that is not described, you know, the apparently random part. the distinction might be new for example for either completely predictable or ideally random

unstructured sources the data is explained by one descriptive extreme or the other but the thing is that nature is seldom so simple you can't simply approach it and detect its structure by appeal to ideal states, either pure randomness or pure order. It appears that natural processes are mixtures of randomness and order. It is the organization of the interplay between order and randomness that makes nature complex. A complex process then differs from a complicated process.

A large system, for example, consisting of many components, subsystems, degrees of freedom, and so on, is a complicated system. Like, for example, an ideal gas. But a complicated system doesn't need to be complex in the sense that I talked about it here, namely the fluctuation, the movement between randomness and order. The ideal gas has no structure. Its microscopic dynamics are accounted for by randomness, basically appealed to an ideal state of observation.

Now, it turns out that the balance between order and randomness can be reached and used to define the best models for a given data set. The balance is given by minimizing the model size while minimizing the amount of apparent randomness. Now, the first part is a version of Ocom's eraser that we talked about in the first session. Causes should not be multiplied beyond necessity. The second part is a basic assumption of science. Obtain the best prediction of nature, the sufficiency of predictive behaviors.

But neither components of this balance in modeling can be minimized alone. Otherwise, absorbed best models would be selected and created. Minimizing the model size alone leads to huge error. Since a huge model, since a smallest model is a null model, that basically captures no regularities. It does not detect a structure. Hence, it cannot yield intelligibility. A huge model also is simply the data itself, and manifestly not a useful encapsulation

of what happened in the, you know, for example, in the laboratory and under, you know, complex observations so both model size and in used mark and induced error must be minimized together in selecting the best model so in modeling in scientific modeling usually the sum of the model of size and the error is minimized the sum of both but what is missing in this story of modeling of scientific modeling, which is the definition of classical models, is that what is missing in this story of what to do with data

is how to measure a structural regularity. Just how a structure is measured determines where the order randomness dichotomy is drawn. Now, this particular problem can be solved in principle. we take the size of the candidate model as a measure of a structure. Then the size of the best model is a measure of the data's intrinsic structure. If we believe the data is faithful representation of the raw behavior of the underlying process, this then translates into a measure of a structure in the natural phenomenon originally studied. but again we are facing ourselves with another deep puzzle one that precedes

measuring a structure how is a structure discovered in the first place if the scientist knows beforehand the appropriate representation for an experiment possible behaviors then the amount of that kind of a structure can be extracted from the data as outlined above. In this case, the prior knowledge about the structure is verified by the data if a compact predictive model results. But what if it is not verified? What if the hypothesized structure is simply not appropriate? The best model could be huge or worse appear upon closer analysis to diverge in size,

Not to mention that this idea of modeling according to our prior knowledge, our prior knowledge of how to measure a structure according to the structure that already we assumably have knowledge of is also an epistemic bias. At the very least, an infinite model is impractical to manipulate. Now, the situations indicate that the behavior is so new as to not fit into current understanding. Then what do we do? Now, this is a problem of innovation in modeling. How can an observer ever break out of inadequate model of classes and discover appropriate ones?

How can incorrect assumptions be changed? how is anything new ever discovered if it must always be as we talked about expressed in the current language now the thing is that contemporary physics does not have a tool for measuring a structure it appeals in order for it to detect and measure structure it appeals to ideal estates as we talked about pure randomness or pure order and then proceeds by way of for example inductive rules try to you know basically come up with a model of a structure that allows

for its quantification no one of the recent approach in in modeling to overcome these problems is the computational modeling. Computational modeling adapts and extends ideas from the theory of discrete computation, which has developed measures of information processing structure to infer complexity in dynamical systems. Now, computational theory defines the notion of a machine, a device for encoding the structure in discrete processes.

Now, this machine, this computational machine, the structure of the machine itself in this computational modeling approach, can be said to be the best approximation to the original processes information processing structure using the model size and apparent error minimization method discussed earlier. Now, once we have constructed the machine, we can say that we understand the structure of the process. But what kind of structure is it? Has machine reconstruction discovered patterns in the data? Computational theory answers such questions in terms of different classes of machines it distinguishes.

For example, these machines are classes with finite memory, those with, for example, infinite one-way stack memory, those with first-in, first-out queue memory, so on and so forth. but nevertheless in this approach the architecture of machines themselves represent the organization of the information processing that is the intrinsic computation the reconstructed machine is a model of the mechanisms by which the natural process manipulates information now we talked about this intrinsic computation as different from logical computation so I don't get to this

anymore one of these machines that is capable of basically allow us to to detect structures regularities of the structures across different levels or or different hierarchies, different hierarchies of structure, but also detect novel emergence of novel properties is what's called by and invented by James Crutchfield and called the epsilon machines.

So briefly, an epsilon machine is a computational machine, is an intrinsic model, basically a model that is embedded in the, encoded in the structure of a physical system and how this structure manipulates information. Basically, the model becomes the subsystem of the system under observation rather than simply becoming an analog, an outside analog of the system. This epsilon machine is basically constructs, detects, and measures structures according

to a different complexity matrix. because we said that the whole idea that is at the base of how we detect regularities in nature how we render structures intelligible is really the basically the the interication between randomness and order now so for any Any robust model needs to be a model that takes its metrics of complexity, its metrics of manipulation of information, information processing, within a framework in which order

and randomness are mixed. They do not exist in ideal states. The metric of complexity that allows for expression of regularities of complex systems at this level of intermixity between chaos and randomness is the metric of complexity according to statistical measures. speaking via a kind of a rather metaphor is the understanding that for forever for a computational machine of this of that works and operates according to the measure of a statistical

complexity the construction of the the way that it generates or detects the structures is according to the statistical basically calculation of the prior estates that have led to its present condition so basically the machine this epsilon machine or this computational machine as a as a model to detect and quantify a structure works in this way it's groups or it groups on different classes equivalent classes of a statistical probabilities that have

led to its present condition into equivalent classes of causal estates these are basically causal estate they have caused its present condition so it represents causal estates in terms of equivalent classes of equal probabilities so we have equal probabilities that led to the present condition. Now, it's infer how it is going to move to the future condition, to its future structure, according to basically making an inference

between how statistical equivalent classes can be reconstructed from present condition to the future on the basis of how the past led to the present. So you see, it's basically a form of memory complexity, a memory-driven complexity. It infers how it's going to move toward the future according to causal states or equal classes of probabilities that have led it from its past condition to its present condition. This is a very brief picture, a schema of basically of epsilon machines or computational modelings.

Epsilon machines reconstruction allows to discover new structures. And discovering of new structure is done by grouping the lower level states into equivalent classes of the same future morph. The equivalent classes then become the notion of causal state at a higher level. So, this basically at the same time shows this computational machine, which is a form of intrinsic model to detect and quantify a structure, shows that not only we can basically map the complexity of physical systems as a hierarchical complexity.

but also detect and quantify and explain the emergence of new properties, new properties which are incomputable in terms of lower level or antecedent conditions. now this is simply the so that in a sense the the intrinsic model defines a calculus of emergence a calculator of emergence where each instance of computationally produced emergence is identified according to inductive leaps

into novel classes of automata that is emergent phenomena are recognized by reconstructing the least complex type of finite automata or epsilon machines which adequately captured the complexity of the system dynamics in question now Now, analogous taxonomy of attractor types in dynamic systems, epsilon machines are arranged in a quasi-hierarchy of increasing complexity. Quasi, because they are only partial orderings. Moving up and down hierarchy of a given complexity in a system takes place when regularities, i.e., equivalent classes,

are detected in a series of increasingly accurate models of the data stream issuing from the emergent phenomenon. Now, this is basically a very basic picture of a kind of a modeling that is designed in response to some of the epistemic biases of classical modelings, which are also derived from the epistemic biases of certain empirical approaches, and also in relation to new definitions of complexities. complexity as being a measure of a statistical ordering,

namely how groupings of certain causes create a hierarchy of a structure and how these different hierarchies of structures can orchestrate a new, basically, level of structure and function. But also, these hierarchies can lead to levels of structure and functions that cannot be adequately explained in terms of basically their antecedent and lower level constructions. This one, the latter one, is basically the definition of emergence.

emergent of radically new complex properties. Radically new in so far as they express a variation of Turing's theorem regarding the non-computability of emergent phenomena in terms of lower level or antecedent conditions. I mean, Turing's non-computability theorem was itself inspired by... Godel's limitation theorem and proof of both Godel and Turing's theorem, you know, were employing, you know, a variant of the same, you know, Canturian diagonal method that,

you know, Cantor used in order to basically show the generation of, you know, new cardinalities. I know that John Bova asked me to talk about a little bit this diagonalization and in relation with to some of these, you know, approaches in computation and ultimately in modeling in science and how we basically, diagonal theorem deeply influences how we see the nature of complexity, how we detect emergent phenomena, so on and so forth.

A very brief, before we go to discussion, is that basically the way that these models work and detect emergent phenomena is that they are using, They are mechanizing a variant of the Cantorian diagonal argument. I mean, hopefully John will talk about all of this in his seminar. But, I mean, the conceptual framework of the diagonal argument can be explained in terms of basically that self-referentiality.

Concomitant with negation creates basically novel sets or novel structures in our case. It's basically, and this is really the definition of emergence and how it can be mechanized in the architecture of a computational model. The model creates self-referentiality. For example, a fractal is self-referentiality without negation. Basically, self-referentiality of a structure, recursivity of a function, of a computational function,

creates self-similarity of a structure, a fractal. Now imagine a self-referentiality that produces intrinsic self-dissimilarity. And this is what really the nature of complexity is. It's about self-dissimilarity. Namely, the emergence of the radically new cannot be explained, non-computable in terms of its lower level recursive processes. Those statistical equivalences, how they basically are undergoing recursion,

as they are going under recursion, they also randomize and recombine the process of recursion. This recombination and randomization is the equal of negation in the diagonal argument. So basically, statistical equivalences are, you know, propagated, but also at the same time the computational machine, this intrinsic model, randomizes the process of recursion, of the recursion of these statistical classes. And by randomizing this, it creates structures that are self-dissimilar. Basically, it allows for the proliferation of new classes of complexities.

Or basically, leading to emergence of new structures and how basically it detects these structures. It detects novelty. And it's then quantified according to these, basically, these moves between recursion and negation, recursion and randomizations. Recursion of statistical equivalent classes and randomization of them. So this was a very basic picture of, you know, some of the ongoing, you know, developments in the field of modeling and, you know, through which we can detect and quantify structures and render physical systems intelligible.

So, questions? Does a more complex system also become a more unstable system? No, actually, the complexity, as I said, complexity, first of all, complexity, when we are trying to define various conceptions of complexity, we have a lot of them to the point that it's almost sometimes even mentioning the concept of complexity becomes problematic.

You see, complexities, for example, in old definitions of complexities, its complexity by appeal to, for example, variations, dynamic systems, so on and so forth. But the thing is that the realist concept of complexity through which the nature of complex systems can be articulated is one in which complexity is essentially a feature of hierarchies of structures and functions, with these hierarchies being understood as distinct, qualitatively distinct level of regularities

of a structure. Now the thing is that it's the intuitive understanding that a complex system is about the organization of things, how things can be organized, and how hard is it for them to be organized. For example, we see that the complexity, for example, in complexity at the level of biology, We see that these different level of structures highly basically enmeshed to one another to the point that these hierarchies are nested at the level of DNA, at the level of, for

example, going at the level of cellular, level of organs, hormonal pathways, so on and so forth. The thing is that as complexity increases in this definition of complexity, namely the question of statistical stability and organization, as complexity increases, the emergence of new functions in a given organization or a given system needs to account for what is already in place. So those complex systems undergo instability in which as the complexity increases,

new variations of structures also can increase, and new functions can emerge, new functions associated with these varied structures. And then these emergence of new functions basically create a de-stabilizing effect for its undergirding or substrate levels of organization. That makes basically the organization of the complex system in that sense unstable. But the thing is that the complexity really, as we talked about, is really the understanding of how things basically build on top of one another.

So it's a question of statistical stability in the sense that what comes on the top, on the uppermost level of a structure and function is already entrenched within its deep structure of functional organization. And hence, really, in the realist account of complexity, as complexity increases, the variation of a structure decreases. The variation of a structure decreases. The variation of a structure decreases because if the variation of a structure increases,

it creates instability in the organization, in the entrenched organization of the system. So in that sense, basically, as complexity, statistical complexity increases, it does not destabilize the system. It basically, it is in response to generative entrenchment of a structural functional hierarchies hierarchies that undergeared and support, for example, a new property. And so, would a less higher up in the hierarchy and a less simplified structure, would that

them be the motion song to gates that you mentioned earlier no it's not really the structure I'm in mind this is something that what I want to talk about that mind is a functional designation mind shouldn't be understood as being of structure but being of a function being of a function is not is different as has a different kind of evolution this is what also you know one of the single most useful insights of Hegel that a spirit has functional designation and the function of a spirit is different from the evolution of basically that

what we basically attribute to biological evolution or physical evolution. Now the thing is that as the complexity increases, the variation of the structure decreases. But the decrease in variation of a structure doesn't essentially mean decrease in the complexity of the emerging function. So this is because the relation between a structure and function is not isomorphic, is not one-to-one. And this is something that I will get back to later.

Reza, I have a question. Can you hear me? Bring me down. Sure, yes. Okay, the epsilon machine, is it, the epsilon machine is sort of like a complex, complex, is it like the same as existing computational machines or epsilon machine is a new form of a more complex computational machine made up of smaller. It's not, you see, an epsilon machine is not a complex machine.

It's, in fact, a very basic machine that shows how complex it can be, can generate. Oh, okay. Sorry. I'm so sorry. Okay, let me reconnect. Josh is saying that everyone is robotified. Okay. Chimerization, that's so good. Everybody's up on their, like, Reza literature, right? like immediately, chimerization. The pink ice cube.

Love it. Okay. Okay. So, yeah. Okay. So, you see, epsilon machine is not a complex machine. Epsilon machine is a computational reconstruction that works the statistical metrics of complexities. Basically, the epsilon machine is what allows for the construction of complexities and detection of complexities. It's an intrinsic model in which the architecture of the machine itself, of the computational machine itself, is reflected, encodes the structure of information processing in a physical system.

Now, Epsilon machine is a machine that works by measures of statistics, measure of statistical classes. Now, it represents the emergence of complexities and new structures by way of how different or similar equivalent classes of the same probabilities, namely statistical groups can be composed, can be combined and randomized so as to basically lead to the emergence of new structures and basically new complexities.

So in this sense, it's a different kind of computational machine in the sense that its metrics is a statistical metrics, it's a measure of statistics, and also in the sense that it doesn't is not complex as such it is really and represents a how complexity constructed in physical systems out of basically its statistical of a statistics of causal states for example what we talked about like you know you have different pathways leading to the present

I it then tries to read basically by way of a a a a method of self-referential it tries to no a approach them we are recursive processes but as also is would approach them we are these recursive processes it doesn't produce the same of what has already come before it namely doesn't simply reproduce the transition from past to the present but by way of randomizing it by way of recombinatorial dynamics, it also negates this recursion to the point that it negates redundant patterns, patterns that were associated from its transition from past to the present. So it basically yields a novel state from the present to the future.

I kind of have a question and this is maybe more from my own particular research than anything but it's I've been trying to do some research into some of the deep learning algorithms and stuff like that and I see Boltzmann machines coming up quite a bit and I guess I'm wondering if this is in some way related to the particular form of structure you're talking about or? No, simply just a structure in general. A structure being simply the way that we can structure, as I said, is really basically the intrinsic choice, intrinsic, not as I said, not the external choice, not the observer choice.

I guess I'm not looking at... So it's between randomness and order. Yeah, what I'm looking at right now is a simple definition straight up from Wikipedia. Restricted Boltzmann machine is a generative stochastic artificial neural network that can learn a probability distribution over its sets of inputs. So I guess that's kind of why I'm seeing this. Yes, I mean, it's similar. But I mean, OK, a good, I will put some stuff in the classroom to like a really elaborated discussion as how basically the measure of a statistical complexity and also the reconstruction of epsilon machines as intrinsic models of detecting and quantifying

a structure are different from different measures of complexities and models built around them. Like, for example, Shannon model of entropy, Boltzmann, logical depth, so on and so forth. That would be very helpful. Thank you. Reza, regardless of these particular types, I was more thinking about their application. The machine... Application, there are models. No, no, for sure. But I was thinking about... Intrinsic models. What is used in high-speed trading, for instance, is very similar to Epsilon Machine, right? I don't know high-frequency trading algorithms really function.

From what I know, it's basically similar to what you're describing with Epsilon Machine because it relies on the data from the past, but it's able to randomize and so be a step ahead of what's actually unfolding and be able, with some certain degree of success, predict predict what then what the next what the next move will be and act on it given all the like sort of other stuff are are are there which means like high speed access to how prices are developing you know what I mean like a the material substrate has to be there which means you have to pay thousands of dollars to get that yes but I mean the way that I understand high frequency

trading is more on the sense of randomization. Yeah, that's why you said random. And that's what made me think about it. Yes, but this is self-referentiality plus negation or recursion plus randomization in order to create self-dissimilarity. Now, on the one hand, and also to explain that how structures can be built on top of one another. Basically, the present situation built on top of the past, present to the future built on top of that, with it being dissimilar to the past to the present.

So it actually creates a hierarchical form of complexity in which, as I just talked about, as the hierarchy expands and as how as complexity increases it stabilizes it stabilizes the nature of complexity its its its in fact its I restricts randomization at the level of a structure even though ran restriction at the level of randomization of a structure doesn't directly translate to restriction of emergence of new functions this is important

so the Epsilon machine is really that simultaneously allows for detection and representation of radically novel patterns but also it represents the entrenchment, the generative entrenchment of complexity, the hierarchical nature of complexity as such, that moves toward basically localist stabilization. Hence, complexity is always local. Thank you. So we're 2.09.

We can go on, but we're already nine minutes over the class time. Give or take a couple of minutes here or there. So if people want to ask more questions, or like Reza has some final remarks to end the class, this is the time to do it. I absolutely have no final remarks. So any more questions, guys and girls? Reza, you should have noticed how the sound went crazy. We had like a tumorization immediately, because it's so much more like tumorization than it sounds like robotification. So I had one quick question, Reza.

Regarding the manifest image or interpreting the manifest world, per se, and the two levels of observables and conceptual activities, or as you said, the move from exertion to inference? Yes, the moves between exertions and inferences. So is this realm of conceptual activity I think you mentioned that it does not include metaphysical discursive frameworks. So I was just asking as to whether those sorts of constraints or delimitations are superfluous or are necessarily overlooked by manifest, or should be overlooked by the transition from exertion to inference.

No, they are not overlooked. Basically, metaphysical assumptions can be explained in terms of our moves between assertions and inferences. Basically, metaphysical claims are adjudicated by way of these rules, intralinguistic rules. Okay. And how we map these rules to the physical world. Right. Yeah, I was just following Pete's recent work, you know, I think it's very interesting to look back on the metaphysical continuum, and I was thinking that this was a very productive,

delimiting vision of conceptual activity, but was wondering if it negated or if it encapsulated metaphysics as well. It encapsulates it, yes. Great. Cool. Thank you. Welcome. But the scientific image then would move beyond those claims or overtake. No, the scientific image, you see, the thing, we had this question on the classroom that But you see, science always yields metaphysical claims. Scientists always come up with metaphysical claims, even when they don't claim that—even

when claimed they are not doing that. And metaphysical assumptions basically need to be traced back to their epistemic assumptions. Again, those epistemic assumptions need to be traced back to, you know, the moves between assertions and inferences. And that's, you know, part of what philosophy of science does, to show that not only science yields, comes up with metaphysical claims, and these metaphysical claims need to be, you know, robustly appraised and analysed, but also the metaphysical claims and the metaphysically

contaminated scientific approaches can also, if applied incorrectly, can lead to biases. And this is really what really happens, as we talked about very briefly, in modeling, in modeling where basically science tries to render structures and regularities intelligible so as to explain the complex nature of reality. And models are tinged with biases, in fact.

And that's, you know, philosophy of science both, you know, tries to work on these metaphysical assumptions and the epistemic assumptions behind them, but also how these biases or metaphysical, basically, assumptions influence the actual scientific practice at the level of modeling, for example, or, you know, articulation of intelligibility. Cool. Thank you. Welcome. Hey, so if we are no more questions, maybe we can close the class and move the rest of

the conversation to the classroom on Google Classroom. Is that the word? Sure. Yes. Okay, so thank you everyone and thanks Razal for this awesome class. I actually would have liked to sort of like hear more, especially on the manifest image side, but I guess I just have to wait for next week. Now, my question is, are you participating in the after finitude, the aesthetics after finitude? Yes, we are on Skype. Yes. We are on Skype, but you're not coming, right? No, thank you. Okay, so I guess that's not next Sunday, right? It doesn't conflict with the class time, right? Oh, is it next Sunday? I'm not sure what they say. I'm going to look and make sure that it doesn't conflict with the class time.

I thought that, I checked it. I thought it wasn't conflicting with, let me see. Mine is on the 5th as far as I know. Yeah, I think it doesn't conflict because next Sunday will be, I was still like 31st or the 1st of February, actually. Yeah, yeah. I don't think the mind is comfortable. But I will recheck with Amy. Yes. So thank you, everyone. Thank you, everyone. Thanks, Reza. Thank you. Bye. Bye. Thanks. Bye. My camera wasn't working today. It's not like I'm hiding. My camera wasn't working. Bye-bye. Bye-bye. Take care.

Thank you.