Ray Brassier - How to Train an Animal that Makes Inferences Sellars on Rules and Regularities

We should start. Good afternoon. Welcome everybody to the second day of the Human, Animal, and Politics, Science, and Psychonomicist workshop. I will introduce our first speaker for this afternoon, Ray Bursier. Ray Bursier is Associate Professor of Philosophy at the American University of Bayreuth, Lebanon. Ray's research interests are in the area of fractionalism, naturalism, historical and eliminative materialism. Philosophy of mind and cognitive sciences, and I think Ray has been working for a number of years in between analytic and continental philosophy, and along what we said yesterday about the necessity to bridge that gap, I think Ray has been one

first-wise, put a lot of efforts into carrying out his tax. Grace, the author of a monograph titled Nylon, Unbound, Enlightenment and Extinction, which came out with Albert in 2007, is also the translator of Paddy's St. Paul into English, and the editor and translator of Theoretical Writings together with Valverde Toscano. He also translated, and we don't know whether we should be grateful or less than grateful, he also translated into English the quite well-known book by Ketameyasu, After Finitude. The title of Ray's paper today is How to Train an Animal That Makes Inferences, Centers on Rules and Regularities.

Thanks, Lauren. Okay, so this paper will have three basic parts. The first part in which I simply want to introduce the context, the basic and theoretical framework in which Sellers' work makes sense. and it's by way of trying to introduce Selser's project because he's still a relatively kind of underappreciated false god even though I think he is beginning to emerge as arguably the most important American false god of the 20th century. And this is something that is...

I mean, Sellers' great contemporary champion is Robert Brandon, who is exerting a significant influence over contemporary Anglo-American philosophy. But it's important to recognize Sellers, Brandon's debt to Sellers. Now, Sellers is classified as an unhealthy philosopher, but one of the things I want to do today is to explain why this is irrelevant, this kind of sociological characterization of this work is irrelevant and you can actually some of those principal concerns can be also be recognized and taken up by any contemporary philosopher who is interested in the problematic of

post-Cantian critical materialism, so to speak. Critical materialism as opposed to dogmatic or metaphysical materialism. So, I want to distinguish between the basic philosophical context of Sartre's work can be characterized in terms of three kind of categories. Naturalism, rationalism, and nominalism. okay? Sellers' naturalism is Darwinian rather than Aristotelian. This means, very simply, that for Sellers the propriety of function is context specific. The contexts that explain

function or purposefulness are contingent selections rather than necessary designs. So this is the key, it's important. All these terms, naturalism, rationalism and normalism, have to be kind of qualified in order to be relevant to contemporary philosophy. There are lots of competing versions of philosophical naturalism, but unfortunately one of the symptoms of the research of contemporary naturalism is a kind of neo-Aristotelianism. Serajit Sera is a Darwinian naturalist because he believes that Darwin has successfully demolished the Aristotelian injection of purposefulness into the domain of the physical realm.

So Darwin explains purposefulness as a contingent byproduct of purposeless evolutionary mechanisms. First part. Secondly, rationalism. Sellers is a Kantian rather than a Cartesian nationalist. Concepts are not representations but rules for combining representations. so one or four consequences of this is that for sellers for sellers of rationalism our ability to do things with concepts requires that concepts be able to do things to us and the brunt of sellers' account is to explain how humans as concept

mongers can be motivated by concept and this is what I want to try to explain today, sellers has an extraordinary kind of sophisticated account of our conceptual competence. Conceptual competence being what distinguishes humans from other animals. So that in this scenario, subjectivity is no longer, but for sales difference from Kant, that subjectivity is not a condition of representation, but a function relaying at least two concepts. And this relay between concepts is inferential. So the propriety of conceptual function, according to Sellers, is inferential. And inference is what relays two concepts.

And rationality, therefore, is an abstract context generic function. Not a context specific function, but a meta-function, a generic meta-function. And I'll try to explain what this means. Hopefully this will become, I realize this is very abstract, but hopefully this will become clear. So, in this regard, I believe Sellers fleshes out a suggestion, a claim that we've heard a couple of times before, is that the generosity of the human consists in transcending biological specificity. This is a claim that's often made by kind of post-agaming philosophers, that it's a kind of bad view it makes when you characterise it as a human as a void and animal.

The animal that's capable of is this kind of, the capacity to transcend any kind of biophysical determination. But unless you explain the mechanism, unless you give a concrete explanation for how this capacity is acquired, you have something that is like gesture, hand-wake. and this is, you know, if one accepts that to be a consistent rationalist is also to be committed to a kind of naturalism, one cannot allow unexplained discontinuities in nature. There are discontinuities in nature, but they can be explained. They're not miraculous emergencies out of ego. So there's a continuity between

cells of rationalism and its naturalism because in both instances it's about function. So rational capacities are a kind of higher order development of biophysical capacity. I'd like to explain it. Finally, normalism. Sellers is a normalist, so that concepts are linguistically instantiated functions. Thinking for Sellers, or conceptual rationality, cannot be understood independently of language. So he embraces the linguistic terms. He's a philosopher who thinks he cannot give a credible account of

our conceptual capacities, independently of the structure and functioning of language. This has two consequences. First, the sociality of language implies an intimate link between sociality and rationality. And second, Selv's normalism is compatible, paradoxically, because normalism is almost taken to be an anti-realist position. Selv's normalism is compatible not only with realism but actually with a transcendental realism, precisely insofar as he refuses to conceive of the reality of concepts in terms of a relation between concepts and extra-conceptual realities. So, for Sellers, Sellers does not believe that language, he

doesn't believe in concepts or representations, and that they are to be understood in terms of a relationship between an ideal entity and some kind of real instance. By the same token, this is because he doesn't believe that meaning is relational. He doesn't believe that meaning can be understood in terms of relations between words and things. Meaning is functional. It's a very conceptual role. So that meaning can only be understood in terms of a conceptual or symbolic economy of functions. So that Sellers affirms the reality of conceptual function as something whose propriety requires no ratification in terms of a correspondence relation with the non-conceptual order.

But at the same time, he is able to affirm the reality of an experience transcendent physical domain because he construes experience itself as a form of sapience. Sapience being contrasted to sentience, another key distinction for Sellers, and sapience being a conceptual achievement through and through. This is Sellers' Hegelianism. The limits of experience can be conceptually transcended because those limits are conceptually delineated. However, Sellers' ultimate metaphysical, the currency for his realism is processes, not things.

What Sellers is a realist about are processes. In fact, this is one of the most difficult parts of the system. Sellers is a metaphysician. He's an unrepentant metaphysician. But he ultimately endorses the metaphysics of what he calls absolute processes. But a stratified... The domain of processes is stratified. Not all processes can be straightforwardly translated into one another. They're not univocal. So just as life forms are patterns of organic function,

thought forms are patterns of linguistic function. So Sellers' metaphysics is anchored in a transcendental functionism. Identical functions can be variably instantiated in different material media, and the variability of organic form is bounded by the propriety of adaptive functioning. In other words, what constrains the metamorphicity of organic functioning is adaptation. Every, all functioning must be adapted. A maladaptive function results in the elimination of the organism. By the same token, the variability of linguistic form is bounded by the propriety of conceptual function.

Serves believes that linguistic forms are the possible kinds of... Given that languages are symbolic economies and they are autonomous or self-regulating symbolic economies that are not anchored in any kind of correspondence relationship with you know, linguistic reality. This usually is taken to license a kind of relativism, a kind of full-blown relativism. You have a proliferation of language games, none of which can be, some of which may resist translation into one another because there is no kind of, you know, there's no transcendental language game

that would allow you to convert and identify in variances of function. Seirot says that there is, there are invariances of function across language games. And the task of a transcendental metaphysics is to identify these invariances of function, of conceptual function. Okay, so the two basic distinctions that organize Seirot's work is a sapient sentence distinction, the distinction between knowing something and feeling something.

He had said, this is the distinction characteristic of intellectual modernity. This is the kind of the Cartesian Galilean break between knowing and feeling. And Salazar is unequivocally committed to this distinction against any kind of postmodern backsliding that would try to kind of erase it and reduce thinking to feeling, for instance, as is very popular nowadays. This is underwritten by distinction between reasons and causes. There's a domain of reasons or conceptual functioning and the domain of mechanical or efficient causation. And it's very important to insist on the separation between those two.

you fuse them together, if you fuse them together, you've got Aristotle. You're back to Aristotle. And this is what Sellers-Illinois-Liver fuses. But these are formal distinctions, not substantive dualism. Dualism is an unarticulated distinction, a ray-fied difference. Humans, what Sellers does is he articulates this distinction between sapiens and sentience, reasons and causes. Humans are concepts, one verse. This is the essential difference between humans and other animals. But this difference marks an imminent discontinuity within the natural order, not a transcendent exception. We can make sense of the discontinuity between humans and other animals.

So self then gives a positive account of the intrication of the conceptual and the causal, or reasons and causes, without presumptively collapsing one into the other. So nature is not reasonable, and reason is not natural. Yet nature's unreasonableness is not unintelligible, just as reason's unnaturalness is not supernatural. The problem that Sellers' post-Arwinian rationalism confronts is this, is the following. How do blind evolutionary contingencies generate purposeful rule-governed activity?

the kind of rule-government activity that Salish takes to be constitutive of conceptual rationality. So this is what Salish sets out to explain. by trying to explain how human animals learn to speak, learn to use language and therefore learn to think. If concepts are rules and rationality is the ability to follow rules, but concepts are also linguistically instantiated functions

and the key to understanding the specificity of the human which is to say the emergence of the human as this rational capacity lies in understanding how an animal can learn to follow rules how could an animal learn to speak Sallers rehearses a well-known and apparently fatal objection to any such account In other words, any accounts of language learning in terms of rules following are susceptible to an allegedly devastating objection. And the objection is very simple. Learning to use a language, L, is learning to obey the rules of L. But a rule which enjoins the doing of an action, A, is a sentence in a language which contains an expression for A.

Therefore, a rule which enjoins the using of a linguistic expression, e, is a sentence in a language which contains an expression for e. In other words, it's a sentence in a meta-language. Consequently, learning to obey the rules for language L presupposes the ability to use a meta-language in which the rules for this language are formulated. So that learning to use a language presupposes having learned to use another language. And by the same token having learned to use a meta-language presupposes having learned to use a meta-meta-language and so on. Here we have an infinite regress.

So, because of this regress, the claim that learning to use a language implies the ability to follow this is absurd and must be rejected. Sellers wants to diffuse this, he wants to kind of rebut this objection. And he wants to explain how the redress can be terminated by explaining the ability to obey rules in terms of the ability to conform to rules. These are not the same abilities, but the latter, the ability to conform to rules, will provide the precondition for the former. the ability to obey rules. This distinction, in Salazar's account, is closely related to the distinction between

doing something for a reason and doing something because of a reason. Clearly, animals do things for reasons. Their behavior can be explained as satisfying an adaptive person. But they do not act because of those reasons. Their behavior is caused by purposes of which they remain ignorant. I mean, the parent, you know, Dennis, you know, excellent example is a kukwu. The mother kukwu deposits her egg in the other bird's nest and expels one of the eggs so that the other bird will not notice that there is an extra egg in the nest.

Now, this is what explains, this is the kind of evolutionary rationality that explains the cuckoo's behaviour. Now, obviously, cuckoos can't count. It doesn't know how to count the number of eggs. So, this capacity is simply a form of rule, it's a kind of pattern-governed behaviour. It's a purposeful behavior where the purpose is not represented by the organism that is effectuating that purpose. So animals, in other words, are impelled by purposes but not motivated by them. In this regard, animal behavior can exemplify complex pattern-governed activity,

even though the animal has no intention of realizing any such activity. and Sellers illustrates this idea in the following passage which is quote one on the handout which I hope people know. So this is a quote from Sellers' By the way, all these quotes are taken from a single essay called Some Remarks on Language Games which is a paper written in 1954 and it's a truly astonishing paper. Sellers writes, What would it mean to say of a bee returning from a clover field that its turnings and wigglings occur because they are part of a complex dance? Would this commit us to the idea that the bee envisages the dance and acts as it does by virtue of intending to realize the dance?

If we reject this idea, must we refuse to say that the dance pattern as a whole is involved in the occurrence of each wiggle and turn? Clearly not. it is open to us to give an evolutionary account to the phenomena of the dance, and hence to interpret the statement that this wiggle occurred because of the complex dance to which it belongs, which appears as before to attribute causal force to an abstraction, and hence tends to draw upon the mentalistic language of intention and purpose, which is obviously irrelevant, which is obviously inappropriate for organisms, for mindless organisms. in terms of the survival value to groups of bees of these forms of behavior.

In this interpretation, the dance pattern comes in not as an abstraction, but as exemplified by the behavior of particular bees. So, Sellers draws their attention to the following problem. How is it that purposeful structure, which seems to require a mind to recognize it, can play a causally determining role in the explanation of what we know to be mindless animal behavior? It can, so long as we bear in mind two things. First, abstract patterns can be realized by a multiplicity of concrete particulars. Second, the particular acts which realise the pattern do not occur because of it.

These acts occur because of a genetic wiring diagram, in the case of the bees, which expresses itself in this pattern and has an adaptive value. And as a good Darwinian, some believe that language learning must be similarly understood as an adaptive phenomenon. Though he will of course, it's true recourse to genetic determination. He does not accept the claim that he's a language faculty, intrinsic hard-wire language faculty. Language is a social and cultural phenomenon. The particular acts through which syntactical and semantic regularities find expression need not intend those regularities. To see how this might be so, we need to distinguish between pattern-governed and rule-beying behaviour. This is the fundamental distinction that's at issue in Selig's account.

Pattern-governed behavior is exemplified when one performs an act A in a circumstance C, or an act D in another circumstance F, for a reason, but without intending the reason, i.e. the complex pattern that these acts realize. And rule-based behavior is exemplified when one performs an act, A in a circumstance, C, while intending to fulfill the demands of a system of rules, which is to say, when one does this because of those rules. language learning therefore can be understood as learning to do A and C or learning to do D and F

because of the system of moves to which these acts belong but without intending to realize that systematic pattern so the decisive component is that pattern governed behavior stands to rule obeying activity as a game stands to a metagame, where the latter, the metagame, is understood as a second-order game in which the rules obey in playing the game, the first-order game, are formulated. Human animals learn pattern-governed behavior by being conditioned to arrange perceptible elements into patterns, and these patterns in turn

into complex sequences of patterns. And all that is required for such conditioning is a system of reinforcements rewarding appropriate pattern-recognizing responses and sanctioning inappropriate pattern-violating responses. In learning to discriminate between appropriate and inappropriate responses, one learns the distinction between legitimate and illegitimate positions, or proper and improper moves in a language game. The positions and moves in the game are specified by formation and transformation rules in the metagame. so that crucially for sellers to make a move in a game is to perform an inference.

But the execution of the inference, understood as pattern-governed activity, does not presuppose mastery of the corresponding rule of inference. And this is the second quote on the handout. Thus, learning to infer where this is purely a pattern-governed phenomenon would be a matter of learning to respond to a pattern of one kind by forming another pattern related to it in one of the characteristic ways specified at the level of the rule obeying use of language by a transformation rule, that is, a formally stated rule of inference. Now, the key thing is that the basic idea is very simple, is that one can acquire competence in executing material influences without mastering the formal rules of influence.

And this is what infants do. This is how infants learn language. They learn appropriate correspondence, they learn to respond appropriately to objects, circumstances, situations by issuing a verbal signal. And this transition is explained in terms of a straightforward competence, an inferential competence, in relaying stimulus and response. So, in this regard, inferential competence is a necessary but not sufficient condition for rationality.

So then, how does one progress from inferential competence, the kind that is displayed by infants, to full-blown conceptual rationality understood as the capacity to be bound by rules? In other words, to obey a rule. Not just to conform to a rule, but to obey a rule. And the answer lies in understanding the connection between game and metagame. So one shifts from inferential competence, which for some reason is constitutive of linguistic competence, to conceptual rationality by learning to identify the metalinguistic formation and transformation rules,

governing positions and moves in the language game. but this involves learning to use meta game concepts in precisely the same way as one learns to use game concepts so in other words one makes moves in the meta game on the basis of one's ability to make moves in the game but in order to make sense of this in order to kind of understand what's really going on we need to introduce some further distinctions to clarify the connection between these different levels between the game and the metagame. So CERRS characterises linguistic transactions, linguistic operations in terms of three fundamental types. Language entry transitions, intralinguistic transitions,

which are inferences, material inferences for CERRS, and language exit transitions. Now, interlinguistic transitions are moves from one position in the game to another. And the paradigm here is simply kind of like a move in chess. A move from one configuration to another configuration. It's a move in the game. All inferences are moves of this sort. But not every position in the game is arrived at from a previous position in the game. in the game. So what sellers call a language entry transition corresponds to a position in the game arrived at from outside the game. They are points of entry into the linguistic

register and perception is the paradigmatic language entry transition for sellers. By the same token, not every move is from one position in the game to another. One may make move that issues in a position outside the game and these are what he calls language exit transitions and action is a paradigmatic language exit transition so perception is the paradigmatic entry into language you respond to a kind of and you enter into a conceptual or linguistic register. And then you can carry on with the legacies of moves in the game, transitions understood as material inferences.

But at some point you will, the consequence or the implication of a practical syllogism will be an injunction not just to say something else or think something else, but to do something else. This is the language exit transition. Action is what you do as a result of a process of reasoning, you're compelled to do something.

Now, the key to understanding the connection between game and metagame, or in other words, between pattern-governed and rule-obeying behaviour, lies in appreciating how entry and exit transitions are not only operative horizontally from the extra-linguistic domain into language, and from language into the extra-linguistic domain. this is perception and action but it's also these transitions are also operative vertically from a language into its meta language and out of a meta language into a language so that game

and meta game stand to one another as language and meta language even though, this is very important not all games are linguistic Not every meta game is a meta language. So it's important, although you can take the there's a powerful, there's a strong kind of formal analogy, it's important not to say that all games are linguistic. There are lots of kind of forms of conceptual activity which are, even if they're kind of linguistically instantiated, they don't involve operations operations over linguistic tokens. And this leads us to the third quote on the handout, which is another letter, I thought it was another kind of quite lengthy quote. Now,

since the games in which rules occur are language games, it occurs to us that categories of language entry and language exit transitions may throw light on the nature of rule-obeying behavior. Thus, we might start by trying to follow in formulations. Words which mention the positions of a game, position words, are, we might say, the observation words of a rule language. In addition to their syntactical role in the rule language, they occur in sentences which come to be occupied as a result of a language entry transition into the rule language. The rule language being the meta-language, in which the rules of the game are formulated. In which transition from language to meta-language,

the stimulus is a situation of the kind met by the position words. So, action in joining context, on the other hand, are the motivating expressions of the rule language. These are what generate language, meta-language exit transitions. And in addition to their syntactical role in the rule language, they occur in sentences, the occupying of which is the stimulus for a language exit transition out of the rule language to a response which is an action of the kind mentioned in the motivating compiling. So, this is an absolutely crucial moment in terms of understanding the articulation between these vertical and horizontal dimensions of language entry, language exit transition.

One shifts from pattern-government competence to rule-obeying rationality, when a position in a language game becomes a stimulus for a move in the metalinguistic game. One perceives positions, in other words, what does this mean? This means that one perceives positions in a game just as one perceives physical objects. And one can be motivated, one learns to be motivated by reasons or by concepts, one learns to do things for a reason by learning to perceive semantic configurations, by identifying a kind of a meta-linguistic transformation that conforms to a pattern, a rule governed.

sorry, a pattern-governed structure. So this is why for settlers, settlers makes kind of, this is a naturalization of the normative in it. It centers naturalizes creativity. And there is no essential difference in kind between the capacity to engage in the material inferences that are constitutive of linguistic competence with the capacity to be motivated by rationality, to be motivated by reasons.

One becomes capable of being motivated by reason, by perceiving a syntactical configuration. In other words, one perceives meanings, patterns in the world, just as one learns to perceive physical objects, determinate configurations. And the extraordinary beauty of Selvus' account is that these, he explains, you get conceptual normativity out of, you know, the account of rule government kind of conceptual normativity falls out of this account of pattern government competence. This is why Selass is able to naturalize. The distinction between reasons and causes does not correspond to the distinction between the normative and the natural.

Selim naturalizes normative compulsion, or normative compunction, by saying that we can be affected by reasons, just as we can be affected by physical objects. and because in both cases it's conceptually saturated. This is another straightforwardly Hegelian trope. This is why and this leads us to the final quote on the commander. The motivating role of ought has often been misconstrued. In the first place, those who recognize that the role is essential to the meaning of ought sometimes can include that ought as motivational

rather than conceptual meaning. This is of course a radical mistake which has primary source in the matrimonial or bow and arrow theory of meaning criticised about what Fels means by the matrimonial theory of meaning is that meaning is a relationship between a word and a thing or a concept and an object. Once you abandon this relational conception of meaning and adopts the conceptual role semantics that sellers enjoy, then you see that there is no essential difference between the meaning of ought and the meaning of any other kind of semantic token. Soll means ought is exactly as legitimate as brooch means

red and und means and. There is something which so means is exactly as legitimate as there is something which got means and ought means a prescriptive property of states of affairs of the form X does A in circumstances C is exactly as legitimate as red means a descriptive, indeed observable property of physical objects. And just as necessary means the modal property of states of affairs of the form, if x is A, then x is B. So, the power of this inferential role semantics is that it makes normativity the key operator

in linguistic functioning. So, therefore, we don't have to, so the distinction between the prescriptive and the descriptive is no longer a differential fact. And you can, this is why Sels is a realist about normative claims. There are things which you ought to do because these are facts about the world just as there are physical facts. But you need a certain conceptual competence to be able to recognise and respond to these facts. So, I want to conclude now by trying to draw some of the implications of this account, of this Szilagin account.

One, there's five, but Frank and I will hopefully discuss these in the conversation. One is the distinction between reasons and causes is irreducible from the viewpoint of the conceptual, but reducible from the viewpoint of causes. This is why this is not, there's no methodical dualism of the normative and the natural reasons and causes, because although this is a necessary distinction, it's an asymmetrical distinction, it's a distinction that's only operative from the normative side. However, in the domain of causes, you can explain the continuity from causes to reasons.

There's a discontinuity from within the conceptual order, but a continuity from the physical order. There is nothing spooky about normativity or rationality. You don't need any kind of gaps or radical breaks. Second consequence is that the kind of conceptual capacities that are characterized, that can characterize in terms of like, you know, apperception or self-consciousness is a reflexive ascent which is a function of pattern compression. Because what you do when you learn to respond to reasons is that you identify syntactical configurations. And this requires an ability

for pattern compression. There's an analogy here with Dennett's discussion of real patterns, the metaphysics of real patterns. The difference between a real pattern is any pattern for which you can identify compression algorithms. In other words, you can identify a straightforward kind of algorithmic sequence that generates the pattern. And the distinction between noise and information is simply the distinction between compressible and incompressible information. Anything, information is simply kind of, you can recognize information in terms of a series of redundancies that allow you to compress a signal. And noise is simply incompressible.

It's actually an excess of information because it's an incompressible signal. Rationality consists in pattern recognition functions. And conceptual rationality is a capacity to identify to reliably identify structural functional configurations third the framing no there's a what says rules in terms of the time says the time I think is spoken to an objection. This is a bit the... Although... And there's a kind of straightforward objection to ourselves

which says that, look, you've introduced another kind of regress. So if your account presupposes that, who trains an infant to use a language? Who regulates the rewards and sanctions that are necessary in order to induct a human animal into symbolic normative order. In other words, you need someone who's already kind of linguistically competent. So it seems that, Sellers says that you learn to recognize animals or trained, human animals are trained to convert ought to be, what he calls ought to be's, into ought to do's.

You recognize the state of affairs as something which ought to be the case, and then you are trained to make it, to bring it about that this is in fact the case. So there is a distinction between the trainers and the trainee. But this, what stemars, unfortunately, one weakness in self-disappointment is that it does not give any kind of positive or concrete account of the different kinds of contexts, and different social contexts which might influence or frame the generating of the transition between ought to be and ought to do rules.

In other words, what is recognised as the norms instantiated by any linguistic community are obviously historically specific. And here Sellers needs an account of social structure. Because in a way all the work is kind of farmed out to a pre-existing social structure. But what Seros does not give is an account of the basic kind of the operative dynamisms that may give one social structure a specific kind of, may lead one social structure to recognize one series of patterns or one configuration of patterns at the expense of another.

So fourthly, the social nature of rational capacity rebuts the recourse for rational miraculous characterization of rational subjective capacity. Causal exceptionalism about rationality is irrational. And this is, I think, what I take to be the power of the Szilagyi mechanism. Because Severs makes the most sophisticated forms of conceptual rationality emerge, or he makes them supervene upon these straightforward competences, there is nothing you can explain the rational capacity as a straightforward social capacity.

And you can give a positive explanation of it. Finally, this rational capacity is not metaphysically conceived as an absolute or unconditional potentiality. So this is why the difference between a Szilagyi account and a kind of bad uniform in terms of man as a voided animal. the plasticity of our conceptual capacities and the way in which we can keep on revising, criticising our conceptual obligations is not entirely open-ended. It's almost framed

in a specific, I think, ultimately kind of social and historical context. And this is, I think, where the Szilagyi and the Krenn need to be supplemented by a kind of a Marxist effect. So in other words, Szilagyi gives you a kind of a hierarchy. He explains the infrastructure and superstructure of conceptual rationality, but he doesn't explain the material base of the infrastructure and superstructure. So in other words, what you need is this large enough fund who could, I think, be usefully supplemented by a more just a kind of socio-economic structure and determinations. OK, I'll stop. Frank Ruga is going to respond to Ray.

Frank is a research associate in the department of philosophy here at the University of Gallien, he is also co-editor of the Moral Provisoire book series at Mende and one of the initiators of the materialism and dialectics collected. Frank is the author of Hegel's Püve, which came out this year in German, published by Constance, in University Press and in English, published by Iber Continuum. And you publish also extensively on continental philosophy and Marxism and translated books by, among others, Badiu and Arancier into Germany. Thanks. Thanks a lot for this beautiful paper. I will ask two questions, and they're more or less improvised a bit, but...

And to not prefer any one of the two of them, one is addressed or referring back to the conception of the animal and the other one to the human. The first one, I enjoyed the idea that the human animal is something like an irrational incorporation of rational. There is no reason why there is a human animal somehow, at least from the perspective of the natural sphere. But there is an emergence of reason. So the break between, let's say, nature and reason can only occur for reason. It's not a break within nature, if I got it correctly. But in a certain sense, and this is what I was asking myself listening to you, wouldn't that mean somehow that all animals are potentially, at least potentially, but maybe more, rational animals already?

Because they do things for a reason, maybe not because of the reason, not a reason they intend to realize, but this means they're proto-human animals, proto-conceptually. Because, if I get it right, doing things for a reason means to implicitly somehow, at least from the perspective of the rational, see something conceptually implicit in animal-like behavior, in pet-like behavior. the dog just likes his bone or whatever. To be able to describe that in the following way, he does that because he likes it or whatever, means to... and he does not realize that he is somehow, I don't know, relying...

all of his actions are relying on certain conceptual schemas. He is nonetheless following in close sense conceptual schemas. So, in some sense, if I get it right, with the emergence of human language, there is a change in nature. Because all nature then will be somehow, forever be proto, or potentially human, or any animal will then be proto, or potentially irrational animal, an implicitly conceptual animal. Although one can only say that from after the emergence of language, in some sense, or one can say, all animals are retroactively rational animals.

Retroactively, this can only be said after the emergence of language, but with the emergence of language, retroactively, all animals will be potentially rational or conceptual animals. I mean, this is more or less, which somehow means the animal, to already anticipate, it's not just a title, the animal does not exist. because the only thing that exists is an already changed animal here. This would be the first question if you agreed to that reconstruction. And the second one is rather, I think, somewhat relaying on the third point on the objection to Solars' account, and maybe a bit unsurprisingly, it depends on what you put up. Because in some sense, if I got it right here, the Solaresian account functions, if one claims, it is the initiation of a rational animal, of a child, into language.

But it cannot take into account or can't the emergence of languages. Or to put it in different terms, maybe closer to the view, it can explain how an individual, an individual child, is initiated into an already established language community following rules, having established certain rules. But can it also, I don't know, take into account what it does mean that there is a new rule emerging. a new set of groups. So a new language community, to put it this way. One implication, or one argument which you put up, which I read as entailing germs,

germs cells of an answer to that question of how one can conceive of the emergence of language from a salaritian position is the following. saying if affection by reasons functions in a similar way than affections by physical objects, one somehow also retroactively extends how reasons functions, how rational behavior functions, onto the causal sphere, onto the sphere of physical objects. So somehow it seems as if there is not homogeneity of a language community, but a homogeneity of the causal sphere itself. Because reasons, the concepts, are not essentially different from the physical objects, from the affection, from physical objects.

So one replaces the, let's say, homogeneity of one language community into which a child is initiated with the homogeneity of the causal sphere as such. This is how I read, at least to some degree, what you call so as natural. So to speak from a realist position about nature means, after the emergence of rationality, means to say the caustal sphere functions in a homogeneic way or homogeneously. And I was wondering, I mean, to put up a guy whom I also find a bit unsexy, but Davidson once said, the language that language philosophers talk about, one homogeneous language community, never existed.

So I was wondering if my reconstruction of what you said would be correct, then somehow maybe this causal sphere in its homogeneous, let's say, structure wouldn't exist. And one way of, let's say, twisting it a bit, maybe to give some hints of slightly Salernzian, but using reading maybe even, one could say that within the moment within which someone perceives meaning in the world, where there maybe is none, there can be the emergence of a new rule.

Somehow, by the way, it calls this element an essential element of an idea. So there has to be some sort of perception of something which is valid, I don't know, taking what it stands for, or whatsoever, no matter, and mostly it does not exist, no matter if it exists or not. So, I'll leave that in. I will just jump in, because I think it links quite well to what France said before. if i don't trust point right the first point it was an objection to sellers saying basically sellers could be accused of a kind of crypto anthropocentrism because after the emergence of language potentially all animals which are our animals only because of the emergence of language

could be uh rational could become rational but going back to ray's independence of ray and sellers previous work, I think this whole projection takes up a new meaning if you think also about the disappearance of language, not just the emergence, it is to say if you think of language as an event which changes nature but is time limited. I just leave there and see what, right? These are very difficult. Thanks. These are really tough questions. First of all, are all animals potentially rational? Look, the emergence of rationality is contended. There's nothing inevitable about the emergence of rationality,

Nor is there anything kind of necessary about the determinate structure. In a way, you could say that the kind of rational capacity. I think it's very important to distinguish between two levels in silence. There's a kind of a rational capacity which consists in giving and asking for reasons. to be simply to kind of... The reason why humans are... to be minded is to be congenitally rational is because they're always engaging inferences. The material inference is the elementary

unit of thought. And then, you know, random analysis analysis into a full-grown research program so that the content, propositional content, is constituted by inferential relations. It falls out of these network of inferential relations. Now, what's important in this account is that the relation is not the... In a way, what's important is not the role players, but the roles. It's the conceptual role that's important, rather than any determined concept that occupies that role. This is why we can always revise the content of our concepts.

We can come to abandon one set of beliefs about the world. Or we can abandon one set of beliefs because they're unreasonable and replace them with better beliefs, reasons that are superior to justification. So, are all animals potentially rational, it depends what you mean by potentiality, not in an Aristotelian sense, in a sense that some other kind of biological lineage could have happened on the kind of circumstances that generated the development of language. And this is really open-ended.

Like, no one really knows. There is a kind of significant discontinuity here. There are empirical aspects. But I would say that this discontinuity is empirical, not metaphysical. We don't yet know how all of a sudden our closest homeland ancestors, you know, were obviously kind of, you know, did not have our cognitive capacities. and then there seems to be this astonishing kind of accelerated kind of development. No one knows what happens. There are kind of plausible empirical hypotheses, but they are just that. They are empirical hypotheses. The claim is that certain material structures need to be in place in order for an animal to develop a man.

So it's not an open-ended... So the kind of all animals are potentially natural. No. No, no, crabs and mosques are not potentially rational. At least not until they've changed quite a lot. This change is not already informed. We don't have this kind of, you know, innate openness. But I do understand the claim of a retroactively yes, because we see that it's a complete citizen that we develop minds and we can think and reason. It's a kind of extraordinary, it's a blessing and a curse. For a long time people thought it was a blessing.

Many people nowadays seem to take a very new approach. I think it is a very, very good step. But yes, the retroactive dimension would be that we can understand the contingency of the protests that led us to become minded. But we can understand that as a series of causal steps. Once again, the claim is that not all causes are potentially reasons. Reasons supervene on causes, but not every cause has a potential reason. So, again, you have to be careful. The claim is not that the reasons are implicit in nature and they become explicit in the

human mind. Because we know that already kind of reasons, which is to say, purposes presuppose natural selection and organic functions. There are no purposes in the physical domain. This is kind of straightforward. It's just efficient, blind mechanical efficientization. So it's not... So this does mean, however, that there's a continuity between biological and psychological. But there's also a discontinuity. And the key thing, I think, this is what's really important about sellers, sellers acknowledges the continuity. You need to have a set of organic capacities to be able to think. But none of those organic capacities

necessitate the kinds of cognitive capacities you have. There is a kind of a strong discontinuity here. Your second question about the American supply No, Sannes cannot come for this, but no one else can. The other point about the emergence now, the emergence of new sets of rules, I take this to be kind of a huge agenda, I'm very kind of, I appreciate this point. point. Yes, there are kind of... The key thing is that when you acquire the ability to reason,

to give and ask for reasons, you can change your mind. You can discriminate between good and bad reasons. And in this regard, I said... So the thing would be that rationalism consists constantly criticizing and revising the basic semantic content, but you need to abide by a set of rules to change the content of your beliefs. In other words, you need a basic structure of revisability to revise your substantive beliefs. And what I don't think you can do, and this is a the Sylogen point is that any subset of your beliefs about the world can be called into

question and abandoned, but not all of the knowledge. You need to hold on to some convictions in order to change others. And I can't make sense of the idea that you could suddenly have, you know, there could be an absolute discontinuity from one conceptual structure to another. And this is why the Storajan account is not, in a way, kind of implicitly refutes this kind of cooing and talking about kind of paradigms. No, there are rational criteria that allow you to adjudicate between good and bad reasons for believing things, and good and bad justifications. And scientific, liberal science exemplifies this kind of critical

critical kind of activity. Maybe just one point. This means that the concept of the rule comprises its own revisability as a potential, I don't know, as a potentiality. So to speak of rules means that rules could be different, but this does not change at all the concept of rule as certain. Yes. So it's still rule following. Yes, exactly. Yes. Yes. you can revise certain rules only by following other rules. Just in the same way as you can revise you know, there are, I don't know you can play chess, you can play a game of chess in three dimensions I think. I don't know how you do it. Rules.

In other words, you amplify supplements, this rules, and another set of rules. And you can do, and this is what we do all the time. But you still need certain rules to generate others. Well, this is Neurath boat. Yes. In the analytic tradition, this is what they refer to the image of, yes, a ship one time, you have to be standing on others, and you can't just repair all of it at once or switch out the entire line. Yes, basically, yes. Shall we get a mic? Is that Peter? Any other questions? Yeah, I see. Can we get a little bit? Oh, no. I'm sorry. Well, Ray, I wanted to ask you about what strikes me as the very randomized version of sellers that you work with.

In your talk, there was an oblique reference to McDowell, which used to be actually spooky when talking about supernaturalism. And that sort of language is very much a part of McDowell's mind and world. And so you have Brandon and McDowell is, of course, Sellers' two main descendants representative of what we now call Pittsburgh neo-Legalianism. And one of the key differences, it seems, between Brandom on the one hand, who's now been joined by Hegel scholar Robert Pippen, and then McDowell on the other, is that with both Brandom and Pippen, you have this insistence on social linguistic rationality as transcendental very much in a Kantian sense, where there is an argument, there's a denial that you could come up with a rationalistic, natural explanation for this, where you just

instead have to take it as always already given. And of course, I think there's some justification for this in Sellers, and then it comes out in your talk precisely in that, to put it in psychoanalytic evolutionary terms, all of the examples that you're giving are of ontogenetic language acquisition, but of course the emergence of languages such at the phylogenetic level just isn't accounted for and have any indicated that Sellers doesn't have an answer for this. Now, it seems as though I can brand him with his inferentialism and Pippin following along and trying to turn Hegel into an accomplice with all of this, that they actually are reading Sellers very consequently. And that given how Sellers frames, for instance, the causes versus reasons, distinction, et cetera, that there are grounds for construing Sellers

as very much Kantian in this sense, and not Hegelian as he sometimes flirts with claiming. I mean, his Hegelianism is very much limited to the first three chapters of the phenomenology and reading that as a critique of, you know, the decisive demolition of the myth of the given, a la, you know, what he's after in the philosophy of mind. So it seems to me that, and by the way, you know, Frank, you suggested that Badiou has an account of this, but I don't think he does have an account of the phylogenetic emergence of language as such. And here he's like Lacan, and Lacan, Similarly, like Brandon and Pippin denies this, Lacan shamelessly has recourse to the Bible and says, in the beginning was the word, and then says, you know, any such queries about the origins of language at the collective historical level are just epistemologically

out of bounds, off limits, etc. So what I'm wondering is you favor what seems like a Brandonian inferentialist interpretation of sellers, and yet you want to have a sellers who would be open to the possibility of outsourcing the explanation for the emergence of language as such to evolutionary theory and to the empirical life sciences. But McDowell, too, complains that Sellers can't really allow for that, even though McDowell also doesn't just dance his foot about second nature being nature, too, and that's it. So I'm wondering, then, why you opt for a Brandonian reading of Sellers, and it seems to bring with it Brandom's very anti-naturalist and anti-metaphysical, deflationary inferentialism that then has contained within it prohibitions, arguments prohibiting,

I think what you want in terms of what I too want, which is this rationalistic naturalism that can explain these things. I don't think that you sellers, Brandon, Pippen, any of them can explain, and many refuse to. So I'm just curious as to why you made this decision to put your eggs in that sort of interpretive basket. First of all, ontogeny versus phylogeny. Now, the emergence of... I mean, sellers would straightforwardly say, well, look, how language emerged is going to be a topic for empirical investigation.

There are all sorts of speculative hypothesis about the emergence of language, but he believes, as a kind of scientific naturalist, You know, we have to refer to the finding of evolutionary biology and cognitive science to explain how this might, this transition might not occur. In other words, it's not that he... And the claim is that this is still conceptual. The point is that our understanding of the emergence of linguistic capacity as a natural phenomenon, using the resources of natural science, is entirely conceptual.

So he's not, in other words, the distinction between, he's not kind of reneging on his commitment to the consistency of the conceptual, because science is conceptual through and through. And he says, but science has the appropriate concept for understanding the emergence of life, not speculative metaphysics. So, in other words, he would also claim the unboundedness for Sáenz gives us from Macdowell in that Sáenz believes in, I don't think this is the unboundedness of the conceptual. Sellers doesn't, because Sellers believes that the distinction between the conceptual and the extra conceptual,

between reasons and causes, is itself intrinsic and constitutive of the conceptual. And that if you simply swallow what causes, if you say that all causes are potentially reasons, that all objects are potentially concepts, then you regress into a kind of a metaphysical idealism or a metaphysical naturalism. But are you not troubled by the fact that, you know, for McDowell, he frames this as the problem of, all right, if you get rid of the myth of the given, and he agrees with Sellers about that, how do you avoid tipping over into, you know, Donald Davidson's coherence when he uses Davidson as the case in point? Whereas, as he, McDowell repeated, it looks at the frictionless spinning in a void, because given some of your realist proclivities,

that is, of course, a genuine problem when you appeal to the intraconceptual status of both scientific discourse about the causal and non-scientific discourse about the rational that's not causal. Okay, very simply, Seller has an answer. And this is a bit of Seller that both Randall and MacDowell ignore. This is the distinction between linguistic significance and picturing, and the distinction between truth and maximally warranted assertibility, which is simply about consistency, and truth as correspondence between what he calls natural linguistic objects, utterances on the nature of natural linguistic objects and physical phenomena. Sellers is a two-tier approach.

Sellers wants to bind together coherence and correspondence, unlike Davidson. And this is one of the bits of the most difficult bits of Sarah's. That's why everyone seems to ignore it. But the picturing relation is simply about what he calls information, it's about a kind of a translation of physical information from one medium into another. The human mind processes information from, is a part of the physical domain, and processes physical information. And this information processing yields a kind of picturing relationship between a linguistic token understood as a natural linguistic object and another part of the world.

This then licenses a semantic assertion, truth understood as a maximally warranted assertion. When you are appropriately related to the word causally, you are then justified in believing things about the word. He wants to articulate, once again, reasons and causes. And this is why it's not just a frictional spinning in the world, because the reason you are connected to the world, but the connection is opaque, it's not transparent. But what he wants to... Again, it's difficult to kind of... to minimize what an extraordinary achievement this is.

the claim is to insist on the absolute conceptual mediation. But, no, conceptual mediation is itself objectively mediated. So in other words, the subject's mediation of the object is itself objectively mediated. And this, it seems to me, is what a properly kind of dialectical material should insist upon. Thank you. Slavoj, then, thanks. I'd like to be short. I was also enthusiastic about this intervention, and there are so many basic points that I'd like to start with the debate. We don't have time for it, of course.

for example, and I'm not alone here, in many inquiries into maybe the origins of language. I don't know why, but for maybe some personal traumatic reasons, I like those like Stephen J. Gould, who insists that for conceptual reasons, that precisely, although language obviously does provide some adaptive advantages, but that its emergence as such cannot be diffused in this way. That's Gould's point of honor, as it were, that the rise of language is a kind of a total contingent byproduct of something else,

which has in itself no clear adaptive purpose. all adaptive games, like you know all these angels bullshit as you, you know this is the lowest group of angels, when he says angels that, when people started to collaborate they discovered that they, you know when the collaboration between people became complex, they discovered that they have many things to say to each other so they invented language. of the struggle. Know that all this is to use today's Baghdad or American terminology, that all the adaptive advantages are more collateral damage. That we use, and again, language has uses, but to put it in more philosophical terms, all these adaptive uses

can only be explained already if you are within the perspective of language. You cannot step out and say oh now i compare oh my god if i have language i can do this easier so again this doesn't mean any obscurantism it's just i think it's an even more perhinean solution it brings to the end this wonderful goods conception which i really like of ex-adaptation the adaptation is not i want to do this i develop an organist i use as in a textbook we call us something okay but a more fundamental point where I'm like now intervening to this debate phylogenesis, ontogenesis and so on. The only tiny point where maybe I don't totally agree with you is I think you tried maybe too

quickly to flirt with Marxism there. You know we have a formal structure but then empirical how we need social analysis. But I think that something is missing with CELAS, already at this formal ontogenetic level. What? If you allow me only two, three minutes, I cannot resist, it's my nature. I will illustrate it, it will not seem so. If I think it's perfect to give an example of your, this language entry, then. Within language, coded, exchange, and then exit. Of course, sex. Let's say language entry, a perception of a naked woman, man, whatever. Then language works, exchange, seduction, OK.

Out, act, we know which act. Where do things get here complicated? I will give you an example. I thought you saw the movie. I used this example recently. Often the British one brushed off with Ewan McGregor before he became a Jedi when he was still working last year on. You remember from that movie there is a tiny debate which I think is a perfect example of what you were referring to as this compression, different level of codes, sorry, rules, meta-rules, and so on. The scene is the following one. Elon McGregor and a girl, of course there is an attraction between the two of them, so he accompanies her home in front of the apartment building. she tells him, why don't you come up to my place for a coffee?

Then she answers, okay, glad that there is only one problem. I don't like, I don't drink coffee. And then she answers, no problem, I don't have any. You see, why? Because at the level, if we take this conversation literally, it's in a way meaningless, not in chairs. Come for a coffee, I don't have coffee, okay. so fuck off or whatever. But where does this immense erotic power come? Her answer, I don't have any coffee, basically says, come up and fuck my brains out. It's the most obscene erotic invitation you can imagine. And from this now, so I will go on into what?

What happens here is that if we take this now, I'm not just telling jokes, but not besides, if we take this conversation literally, It's in a way meaningless. The result is zero. I offer you something you said I don't like it, I say oh it doesn't matter I don't care it and so on. But the result is not zero but an extremely intense message. The rule, this is how maybe, now I come to the conclusion, is missing a much more precise point now in cells. To get what you refer to as this compression of rules and meta-rules, you have to introduce already at the most elementary level the notion of an exception to the rule, not in some boring pseudo-Lachemian sense,

but excepting in the sense that not following the rule becomes regulated. You know, the only way language becomes conceptual language when, at least at a certain structural point, the only way you prove that you really, to use Scheller's language, are really speaking a language, I mean, doing it, not only obeying this objective causality, but following reasons, it must be this kind of empty short circuit, where at the immediate level, the result is zero,

like miscommunication, but nonetheless, the message gets through, precisely through this miscommunication, which is why, as I know I repeat it here myself all the time, for me, these apparently exceptional situations, like, sorry to tell all jokes, like, I invite you to dinner, I have money, we both know you don't, which is not true, but let's say. Okay, then we both know I will pay. There is a ritual that when the bill comes, you insist a little bit that you will pay. But we both know that this is an empty game that at a certain point I will say no, no, but we have to go through this empty ritual. You make a gesture which we both know is an empty gesture. I am supposed to say no, but you know what I mean.

What I am trying to say is that something has to happen at this level of, you know, rule which is obeyed precisely through its cancelling or empty gestures and so on. I claim that maybe it's something, a paradox like this is needed. In a very empirical way. to pass at the level that you describe of really understanding reasons. And I must emphasize another point I hope you have read here. It's not simply, this would be the worst misunderstanding, a question of metaphorical meaning. No, the point is not that the girl in Brashoff, when she says, come, drink a coffee, means it metaphorically, like, I don't know,

my wet pussy is like a coffee, I don't know. I cannot even imagine the vulgarities. It's totally wrong to... I'm not talking about mechathor at this level. I'm talking precisely about what you mean for this kind of conflation of rules and mecharules, where something at a zero level appears as a meaningless. Like, okay, you offer me coffee, no coffee, fuck off each of us, you know. This, there must be some kind of conflation of rules so that at the level where, at the level of life, I'll put it in this way. If bees were communicating with those guys, then this one dance means, then the only way they can communicate is either

the message means come and fuck me and I do it or not, or the message means come for a coffee and if the other says I don't drink coffee, it means, okay, both bees go their own way. Only with a man you can get a human person asking another person, do you want to come for some coffee? And when the answer is no and no, it means the ultimate density success communication. So, I'm sorry if I'm too long. All I claim is that this analysis can be led a little bit further in specifying the precise mechanisms of conflation of level which make, if I may sound anthropocentric, communication specifically human.

There is still, you know, my point is that from what scholars did, there is still some things that are to be done at still abstract level prior to bringing in all that Marxist bullshit of concrete analysis and so on. Sorry if I lost it. Okay, obviously the, the counter-represented number, you know, this is very, it's not, but I think it's very important to insist that rules can be extraordinary complex. I mean, we're not just talking about, it sounds, you know, when you talk about rule, people think of something, you know, fairly kind of, very broad, right?

The problem is that there's this kind of concatenation of rules that are nestled in other systems of rules. So we're talking about a level of extraordinary complexity. But in the examples you gave, I mean, it still seems to me that there is... I don't see... I mean, I take it, you know, it's always that you still... Something else can you do, you understand perfectly what's happened. In other words, you know, a misunderstanding is still... If you have the capacity to recognize a misunderstanding, then even if not instantaneously, then I don't see how this is an exception.

I don't think that misunderstanding is an exception to this semantic point. Because we can always misunderstand one another. But understanding what a misunderstanding is involves a kind of competence. So I'm not sure I see the force of... In other words, I wouldn't be happy with the claim that this is some exception. these are, if you have an account, and once you take these account, the kind of the ramifying social context, then there's, you know, this kind of a, you know, potentially limitless complexity to the ways in which we can, you know, understand an upture, depending on,

you know, the context. But this sensitivity to context can be explained as a kind of, as a rule government competence. So I'm not, so I accept the claim that it gets much, much more complicated and, the Solarshi account is very formalistic, very formalistic and it would probably need to be supplemented with some, it would need to be supplemented but I don't think that it's structurally deficient, or at least I'm not quite convinced that, is that the examples you've given would reveal that it is structurally deficient. I think it can be augmented and fleshed out.

I think Brandon does it. Brandon has no problem accommodating. It's all about context. It's all about the meaning of your options depends on the context and this complex negotiation between you know, speakers' meaning and literal means, etc. But I take your point that I was a bit too casual in the flirtation of Martin. But that's something I... It's not meant to be throwaway, but I take it very seriously. I do think that the Sarajan account does need to be supplemented, but I think it's kind of compatible with psychoanalytic and Marxist.

You have to watch about language adaptability. Oh, yes, yes. Exaltation. Here, right there. Yes. Yes. Functioning. There is a discontinuity between linguistic function and biological function. So in that regard, and Simon insists upon this, is that the proprieties of conceptual or linguistic functioning are not straightforward, can't be read off natural functioning, so that they are exaptation, in a Gouldian sense, because they satisfy purposes which are not, which have nothing to do with kind of satisfying biological needs. So, again, I think that exaptation, sellers would straightforwardly endorse an exaptive account of linguistic function.

We have a last question for the students. Yeah, it might be in some sense a question related to what you just addressed already. I mean, it's obvious and I'm sure you thought about this too when you have this differentiation between material inference that leads to doing something for a reason, you gave the example example of the cuckoox bird, as opposed to reason which does something because of reason that of course has a certain history of the philosophy and one thinks of Kant and his differentiation between doing something out of duty and something that accords duty, flicht gemessen autobflicht. And the problem that Kant introduces with that of course is that

only the latter doing something out of because of duty is actually a moral action but as Greg brings up to your differentiation, it is impossible empirically to decide which is the case if I look at an action, I cannot say out of the way in which it is being done or out of the account that I can give of it either as observer or as actor if it is done because of or if it is done out of you. Isn't your account kind of inviting the same problem? Aren't you introducing a kind of an indeterminacy when you try to describe the difference between causes and reasons? And you say it in a certain sense in your account

when you say that it is an asymmetrical relationship. From this point of view of causes, you can reduce reason to causes from the point of reason, if you're not. and so if you are an observer in that difference how can you actually determine at a given moment whether or not there is something that is happening out of reason or because of reason is it something for reason or because of reason I think is the language that you use that is a question I would like to right now from the papers there was this experiment I think it was done in Chicago where they very happily addressed again human animal, namely a rat, that chose to first save its fellow rat before it went and ate

the chocolate. So it had a choice to open two doors, one behind which another rat was imprisoned or another door behind which the rat could eat a chocolate. And the rat first chose to free the other rat. And of course that is being celebrated in the papers because it's supposed to show that nature itself, animal nature, the human animal therefore, So was there another red? A female, beautiful, young girl? No, I think actually if you're going to go in that direction, one could say the red is actually the true proletariat because it precisely opens the door every single time. It doesn't look if it's blonde or long-legged or brown. No, the red, no. There's another red. So the question I think is how can you really determine the difference between the two, unless you already have predetermined that certain kinds of actions or certain kinds of inferences are reasons and other kinds are causes.

Okay, another tough question. So it says that you can, you have to insist in principle, and this might make him, he is a critical candidate, he's not a straightforward candidate, but he is obviously a candidate. You can in principle always tell the difference between doing something because of a reason and doing something for a reason. And you ought to always strive to make the distinction, even if you don't know. At least you are compelled to try to identify the difference. The difference, this means that this, you know, this means that kind of, you know, look, sorry, absolute rationality is, it is an ideal.

It's not something that is, it's an ideal that we are compelled to strive towards, regardless of, you know, whether we succeed. So in other words, so yes, I mean, I think Seller's offense is, because it is, it says once you are engaged, in a way we are condemned to be reasonable, and even if we fail to be reasonable, this is the curse of rationality, you are condemned to be reasonable and to fail to be exhaustively rational. This is the human prediction. But there's no way out. And Seller's use of this metaphor is that conceptual rationality is like original sin. Once you've taken a bite of the apple of reason, you have to eat the apple to the core. There's no going back to the state of innocence.

You have to reason, to push reason to the limits. And if you die before you're finished, that's too bad, but you have no choice. I want to distinguish between being impelled when you do something for a reason and being compelled when you act because of a reason. And the point is actually that, you know, and I think Perth's point is that, you know, the opposition between freedom and, you know, freedom is just being compelled by reason and not being able to do otherwise, no matter how much you want to do otherwise.

So in a way, kind of rational, you know, the kind of potentiation of rational capacity would be the extinguishing of the capacity for choice. So that you only, in other words, the ideal of rationality is a kind of pure, you know, computational kind of ethical, which is why machines are freer than animals. And I think this is the ideal that drives the Szilagyi. The ideal of optimal rationality is that we achieve, we liberate ourselves, we start intervening into our own kind of material constitution to eliminate pathological inclinations. This is the project of human self-fashioning, which I think underwrites, I think, the kind of, the Szilagy kind of agenda, or at least this is the way in which I think it ought to be kind of developed.

And we have to do things for ourselves to optimize our rational capacities. And that presupposes the differentiation. I mean, if you're acting in order to eliminate pathology, you need to be able to empirically decide where you act out of pathology or where you act out of reason. Yeah, and I would say the fact that you don't know doesn't let you off the hook. Okay, that I accept. Okay, we stop here and let me thank once again Ray and Jack for his self-presentation. We'll be right back.