Hello everyone and welcome to the 12th session of Theory and Object. I'm going to pass the mic to the course instructor now. Thank you very much Theo, thanks everyone. I realize that we don't have many students online, but it's okay, they can watch it later. So, this is supposedly our last session, but as I promised, we will have four more sessions. I mean, you free, feel free to, it's not mandatory, but feel free to attend because we will talk about the rest of the stuff that we haven't been addressed yet.
I will try to squeeze as much as possible in the next four sessions. Anyway, as always, shoot my friends. Any kind of question, any kind of comment from the previous session that you have would be appreciated. None. No one. This is enraging. Someone should talk something. Justin, do you have any thoughts on the indexing paper since you brought it up earlier?
Hello, can you guys hear me? Yeah, of course I can hear you. I actually have been meaning, and I don't think it's a great way to start because I think it's a bit of a tangent. But in reading your paper and sort of your critique of that value premise of simplicity as solving the inductivist dilemma at the end of the paper. And I actually think what Theo just brought up is an interesting question also. But my question was actually about how, I guess, what alternate values you see, like, guiding that process of creating a robust inductive process. And I guess at the same time, your critique of, like, what unexamined values are operative in Carnap and how you sort of diverge from what you see in his thing.
I guess they're kind of related but they're also different questions. Yes, yes. Well, as I mentioned at the end of that essay is that simplicity is simply what you might call to be a very context sensitive methodological constraint. All methodological constraints are context sensitive. that there are no, what you might call, to be given law for their application, nor do they actually suggest an underlying law. They are indispensable, but their indispensable should not be confused with their nomic or
or nomological lawfulness necessity. And that also happens with simplicity, in the sense that I do believe that simplicity absolutely, particularly the formal account of simplicity as being, you know, portrayed by Solonov, Paul Wittany, and Marcus Hutter actually does play a role in determining which hypotheses are suitable. But to think that there is something, that there is a principle of simplicity or elegance and these two are subtly different by the way,
that there is in fact principles of subtlety and elegance, sorry, simplicity and elegance in the universe, like the way that Paul Whitney talks about in his book, I think that would be just metaphysical dogmatism. It's absolutely, there is, how do you arrive at this? Of course you will arrive, The only way as a computer science you can arrive at such a diktat is by way of the notion of effectiveness, computational effectivity. But then that's a circular discussion. That's what Hume calls a petito petiti, a circular explanation.
There is nothing fundamentally, there is absolutely nothing ontological about the principle of simplicity or elegance. These are absolutely methodological necessities and as such they should be handled in a fundamentally context sensitive and contextual way. So that's one and two with regard to Carnap, I don't want to go that much into this but But all I can say, after writing this paper, the more I read of Carnap and his later works on the theory of probability, I noticed that I was mistaken. Carnap actually is not what you might call to be this kind of naive inductivist.
But I hold on my view until I actually develop some thoughts, some coherent thoughts about it, as how he actually diverges from this naive inductivism. In his earlier works he does, he does, he absolutely coincides with the kind of naive inductivism that I mentioned in the essay, but in his later works he doesn't, he doesn't. And it would be actually, you know, when we are approaching a philosopher, we should always think about the history of that philosopher, which is part of the history of philosophy, in the sense that philosophers are not technicians, they constantly think, they revise their positions,
and if we don't take into account their revisions, then we are simply like medieval commentators, not students of philosophy. And to that extent, and so far as I haven't cohered my thoughts about the exact relation of Karna with inductivism in his later work, I think that I need to postpone my answer on this one. Reza, can I ask one question on the material? Absolutely, absolutely, but it might be really I didn't I really wasn't following last week, so it might be really
You were talking about our labs or something I wasn't sure what it was and I finally like read through that elementary experience I read through the text more LL abs yeah Ellos right but what I don't understand is like like what is an Ella exactly and like an Ella is different to a sensation and that's what confused me because I thought the sensation would be an L.M. but an L.M. is if sensation is some more like hope it what you might call the organized sensible intuition and organized sensible into the pinch of intuition yes as you see of course the no love so So what in a concrete sense is an L-Eb like?
It's just like the flickering of your retina or it's like some pre-sensation or something. It's not even… So you see, it depends on actually the philosopher who uses this term. You know, for Carnar obviously he is not a brute empiricist, he's not a naive empiricist. For someone like John Searle who believes in sense data, ellipse are actually sensations in absolute Kantian sense. And what is that? Simply because I have been affected thus and so by the presence of an item in the environment, presence already gives me some rudimentary experiential data from which I can build you
know a more coherent system but of course for Kent this is not the case this is essentially a human position and also for Carnap this is not the case because he's coming from you know Kantian position so air lips for if for example let's see let's say as if can't have a notion of air lips what would these things be there are something more like what sellers called image models so what are image models they are the perspectival, sensible intuitions of us as a perceptive subject of the items in the environment.
What does this mean? So I see a red dot. You know, obviously there is no such a thing as a red dot there. just a stuff, literally from a Kantian perspective, it's just a stuff. It's what Aristotle calls toditi, dis-suchits, dis-suchits. Materiate individuates things in our immediate presence. So for Kant, an air lab in this sense would be something like a combination of sensations
organized by space and time. Now we know that in Kant, space and time as organizing factors of our sensible intuitions are not given by a reality. fact can does not believe that there is such things as the space and time out there in the real world there are representations which means that they are the categories of what he calls especially of especially also synthesis namely there are representations given by us by our rudimentary understanding so So they are essentially logical representations.
So these are sensation organized by rudimentary logical, by rudimentary logical representations such as space and time. So that for example, we see the trace of this kind of red dot moving and we identify it as a red dot without the diachronic and the synchronic organization afforded by the representation of space and time, we virtually could not make something like a sensible intuition of a red dot hovering at mid-air. So this is what you might call to be air-lip for Kent.
And for Carnap also, this is an air-lip. Carnapian air-lips are absolutely not pure sensations in the way that John Searle would say sense data. Okay, so in order to start the discussion I assume that a number of you have read that
essay, stupid essay of mine that I suggested. So in order for me to actually start, let me first give you a little bit of background story about why the idea of induction versus deduction is important in the philosophy of science. And then those of you who have read essay that pose questions so we can you know get into the nitty-gritty details so we talked in the first if I remember correctly the first few sessions about the idea of explanation in science that one of the tasks of science is to
explain observed phenomenon now I'm not going to get into more details because this is essentially a topic for another session but for now we can say that okay so I see an observed phenomenon like when for example I with a forceful effect my knee impacts the table and there is this glass of liquid on the table and upon
the forceful impact the glass tips over and the liquids you know gets out of the glass so the second half of this scenario is what we call a phenomenon To talk about how the liquid escapes the glass and how the glass falls from the table is to describe a phenomenon. What to explain it is something more akin to the idea that under such and such circumstances
my knee with a forceful impact sufficient enough hit the table and then the glass fell. So essentially the idea of deductivism versus inductivism arises in the domain of what you might call to be scientific explanation. What is explanation and the way that at least in a colloquial sense we see explanation in the history of philosophy is that explanation is something more than mere description of the set phenomena. now
hand held German philosopher and I'm simply using him as an example because there are many many more philosophies of scientific explanation So, Hempel in the 1960s puts forward new theories of what counts as a scientific explanation or as an explanation. First of all, you should know that in the philosophy of science, the word explanation,
The word causation, the word regularity, the word nomological constraints belong to the same class of entities. This doesn't mean that they are synonymous, but simply that they are part of the class of the same family. And what does this mean? It means that we cannot talk about explanation by talking about causation or by talking about causation by the idea of explanation in a modal sense.
In order for that, we need to go a little bit further. We need to go into a different class or family of entities. So explanation, first of all, is not mere description. Essentially explanation is contrasted with description in philosophy of science. To describe a phenomenon is fundamentally different. To explain that which happens that can be said to be responsible for the discussion,
describe for the phenomenon that we are describing. Is this clear? I take it it is. So, now Hempel, I mentioned puts forward a number of explanatory paradigms before I enumerate or list briefly such ad items I would like you to know which of course I know
I'm sure that you all know already that's in the course of an explanation we are dealing with two fundamentally different yet correlated entities or statements explanando which is essentially the description of the phenomenon which we want to know what actually give to this phenomenon first place and why okay the and the explainants. The explainants are the y-components. The stuff happened according
to some constraints that led to this observable phenomenon which we could describe. Now, as you know, in the traditional way that philosophers talk since Aristotle about the idea of explanation, we have something about cause and effect. Nero ordered Seneca to kill himself.
The brake didn't work so the car crashed into a tree. This is what you might call to be an individual cause-effect connection. It is by no means, however, explanatory. Why? Because any number of parameters or factors could lead to the death of Seneca, to the suicide of Seneca or to the car crash like I say that the floor of the kitchen collapsed
and I say that flood or water logging was the cause of it but no this is not a scientific a statement by any means because any factors including columns having feeble foundations or any of such things could lead to the collapse of the floor of the kitchen. So essentially explanation is not about a singular cause-effect connection. It is about something more.
Hemphill wants to argue what is actually more about cause-effect connections that we ordinarily talk in our ordinary descriptions of what happens in the world. According to Hemphill, the explanatory component designates what is more to the cause and effect, the additional component. And of course, the explanatory component itself consists of further elements.
What are these elements? One is the notion of a law or a nomological constraint, a natural law. Every explanation should at least have one nomological component. component is simply what you might call to be a kind of philosophical term for the idea of a uniform regularity or except or exception rest exception less regularity means that it can be applied across the board like Newton's laws of
motion or the idea that penicillin under such and such conditions can get rid of certain kinds of infections okay now here a problem arises you see these two examples are fundamentally different the work of penicillin on infection and Newtonian equations of motion for celestial bodies by way of which we can predict the trajectory of a celestial body such as Mars what is exactly the
difference can someone tell me It's your turn. I guess. is it a problem of how you would come to identify the penicillin is being curative yes and what is that what is that exactly yes that that's absolutely correct for what but what
is exactly that component that distinguish the identification of the work of penicillin on infection different from the equations of motion as dictated by Newton's mechanics I guess this is more confused about I want to say that there is a type of empirical testing needed for medicine that I'm not sure how exactly different from them. Haven't you noticed that the work of penicillin infection is essentially based on the individual
cases? It's essentially a probability, an inductive, a purely inductive notion of law. We have seen such and such medical cases to be cured of their infection once we administer penicillin. But with equations of motion for celestial bodies, the notion of such equations, the system of such equations becomes the absolute explanatory or causal component for the movement of celestial bodies under even if Mars has this mass, Sun
has this mass, Earth has this mass, Venus has such and such mass. Even if the cases are different we as a matter of normal logical law are capable of predicting trajectories with sufficient accuracy Does that, I mean, does that for you mean the prioritization of the kind of laws of celestial movements over experiential...
all it is not really it's not the difference between laws as mathematically configured or structured versus experiential inductive laws it is somehow as hempel says that the nature of explanation the paradigm of explanation pertaining to these two examples fundamentally different and so here temples introduces different paradigms of explanation in order to account for different kinds of explanations that we
do in science, in exact sciences, but also in special sciences like medicine, like economy, like anthropology, so on and so forth. The first paradigm of explanation that Hemphill puts forward is something called the end model, short for deductive pneumological model. nomological model so for something to be counted as an explanation within the DN model it ought to have certain characteristics and what are these
characteristics okay first the explanandum might should not might should be the logical consequence of the explaining of the explainance that which is explained that which is ought to be explained mind should logically follow from that which explains. Explanandum, the logical consequence of explainance. That's why it is called deductive. Deductive only in the technical sense.
The logical consequence follows from its premises. Number one, constraint for something to be called a DN model or a DN explanation, a deductive nomological explanation. Number two, such explanation should at least have one nomological constraint or law. For now, we can think of law as a natural law, like Newtonian equations of motions. We are not yet talking about the law. What actually constitutes as a law of nature is not the topic of today's session.
I'm just simply making a list. So when I have, you know, that I have observed such and such celestial bodies, I have described them, these are my premises. And of course the premises require in the scientific assessment something like initial conditions. know how what they are like the minimum constraints that characterizes such celestial bodies like their mass their speed their momentum so on and so forth
this is what we call initial conditions so we know that we add also a component of a natural law like Newtonian laws of motion then we can say according to such combination of premises, it logically follows, okay? It logically follows that Mars at time T, P1 has such coordination. It follows such trajectory. Now, in order for something to be counted as a deductive equation, essentially all of
your statements about the current state of affairs with regard to such celestial bodies ought to be true logically and scientifically and empirically. So these are the kind of constraints that go into deductive pneumological paradigm. It's deductive in so far as the relation between explanandum and explainance is deductive. It follows from its premises, but also it's pneumological precisely because it is predicated upon the introduction of at least one natural law or pneumological constraint. So this is what Hempel calls the DN model. It's essentially a deductive relation. For
something to be a deductive nomological model doesn't mean that it's purely logical. Of course, in order for us to describe the initial conditions of the system of such celestial bodies we need to have observational evidence that ought to be a structure by way of a theory and then we need to give them a component of a natural law so this is not when i am saying deductive nomological logic the deduct the deductive component you should not mistake it as pure deductive deductive relation that we usually talk in logical terms
as if they were completely deprived of any empirical evidence we only call them deductive in so far as the statement the empirical statement that we make about such and such celestial bodies and combined with a nomological constraint or natural law adduces It's a kind of particular case, unexplanatory.
Mars position at place P1, T1, time T1. Is this clear? Please do ask me questions. yeah you know I was kind of thinking about this class last week and you know you're saying the empirical component would be itself like let's just say like I don't know like a Kantian challenge to Humean induction that the natural law itself is logical but that for Hume is you know unpredictable right? It's not unpredictable essentially for Hume it is also a regularity of course he asked questions about nature of the certainty
of such regularity but that's a fundamentally different matter that we will get to shortly Essentially, another thing that you brought up and I think I forgot to actually address is that you should understand that the Hempellian idea of explanation is an empirical one, but not in a human sense. It is not of the sensedatum kind. beliefs that to derive a nomological natural law or nomological constraint
we cannot by any means use modalities or more modal descriptions or modal vocabularies or basically modal logic so you see as I mentioned that causation explanation nomological constraints so on and so forth all belong to the same class of family of entities now these are all modal okay they're all modal counterfactual explanations once or forth now the thing is that hempel is
also an empiricist but not in a human sense in the sense that he believes that modal vocabularies or modal connections are not primitive or not the basic vocabularies for us to derive a natural law and natural law ultimately requires empirical observation as its main component now i just talked about modal let me tell you like this so you see we have something called counterfactual
theory of causation it roughly goes like this if I had put this glass on the table and if I would had with sufficient force hit the table this glass could fall it's a modal counterfactual scenario precisely because under a different kind of scenario of what if what would have been occurred this would not have been generated the falling of the glass so this is essentially modal modality
essentially a part of what you might call to be rational or logical vocabularies so Hembold does not believe that logical vocabularies or modal vocabulary in general are sufficient for us to generate something like natural laws and that's making an empiricist. Of course there are rationalists like Brando and sellers who believe that every kind of explanation every kind of in fact empirical explanation no matter how much fundamental it is relies on the employment of modal scenarios or modal explanations
counterfactuals for that matter like another example and you should know that modal vocabularies are essentially context sensitive they are particularly actually non-monotonous in terms of reasoning which means that if you change a little bit of the conditions fundamentally different consequences will arise okay so I say that I have a non-wet match the match head consists of sulfur and an
oxidizing agent like a safe match as we have it today and the stick is comprised of a dry wood material I use it against a frictional surface under such conditions if I had done this it would be the case that the match will ignite okay now imagine under a different counterfactual scenario I take my majesty my dry majesty into vacuum with the same exact characteristics of the match that I just described
are used against a frictional system and then it lights again but how so we know that in the absence of oxygen the match shouldn't light up can someone tell me why is the case that this match ignites in the vacuum considering the kind of elements give you and characterize the match with yes absolutely the oxidizing
fucking agent the oxidizing agent make the match head burns but not the stick because for the wood to burn it requires oxygen but the match head precisely because it has already an oxidizing agent in it it will ignite but then imagine a different counterfactual scenario where the entire match is wet No matter whether we ignite it in the vacuum or in the biosphere of our planet, none of this will happen. And from this we come to something what Brandon calls a counterfactual robustness.
From this counterfactual robustness we can conclude the regularity as to the behavior of the match once hit against a frictional surface. And that would be our pneumological law. So this is essentially a modal scenario. Now Hemphill thinks that these modal scenarios can actually be reduced to empirical scenarios and by no means we should take modal scenarios as primitives. However, Sellars and Brandon, being rationalists rather than empiricists,
think that there is absolutely no scenario, whether empirical or otherwise, that can be actually be thought in terms of explanation without modal we have modal primitives. Questions? So they're arguing like a modal primitive would be like the situation where you would just normally light a match and be able to experience it? Yes. Or what is the modal primitive? No, the modal primitive, no. This is not actually Brandon Mansellers. This is what you might call to be the negation of modal parameters in which I simply just
use this stuff versus that stuff and this stuff ignites. This is what a empirical scenario, without any what you might call intervention of what if I have done this, what would have arise in such scenario? It's always context sensitive and defisible. Modal scenarios are always non-monotonic reasoning. As such, they are not purely deductive, actually. Precisely because if we change the context, in the case of our match scenario, the consequence will fundamentally differ.
So can nomological law be identified if you're accepting? No, we can identify them and that's the whole point of Salahs and Brandon, that we absolutely can identify nomological laws, but the identification of such regularities is fundamentally at every step to the bottom is dependent upon modal vocabularies. Essentially the behavior of our match across these examples, we can come across a law of how the match is being ignited in different scenarios using a series of counterfactual modal scenarios and then
Think about how these modal scenarios vary according to the context, hence the idea of counterfactual robustness. Counterfactual robustness simply means that something, what Hempel calls a law, is predicated upon modal counterfactual scenarios and can only be derived from such scenarios and nothing other than that so i mean it seems to me though that you'd have to go to pretty great lengths to produce
these counterfactual scenarios yes absolutely and that's what what sellers thinks that the The derivation of natural laws, in the sense that we move from Newtonian laws of motion to Einsteinian laws of motion, requires essentially for us to find what he calls the diffuser. A diffuser is exactly like what you might call to be the context of the oxidizing agent or the constraint of wetness. so then we can talk about different kinds of scenarios. And hence, we have to revise our laws that we have already derived from the kind of counterfactual robustness
that we have generated from such and such scenarios. So Reza, are you bringing this up as a... rationalist counter to like to hempel or something what I would call the the empiricists account in general it's a it's a it's a yeah I count as the empiricist account so it's a simple in hempel is falls under this logical empiricist falls under this empiricist banner
absolutely yes absolutely yes logical empiricism doesn't really like extend to this concept of no modalities what you might call to be for the longest time in the history of philosophy were underdogs underdog vocabularies in the sense that people thought that okay we are using them to arrive at a law but they are not primitives in the order of derivation of making a law a rationalist like random sellers would say that there are absolutely primitives you can't have a pneumological law without modal
the capitals or modal primitives of one kind or another. So even the empirical, supposedly purely empiricist-based law depends on counterfactuals somehow? Yes. And of course the most extreme of the scenario is David Lewis, you know, his account of counterfactual causation where it takes it to a fundamentally different depth. I'm not sure that I go so much to that extremity but I can appreciate it because it shows how much what you might call to be empirical base vocabulary is dependent upon counterfactual
model scenarios. Okay. Should we have a five-minute break before we come back? Sure, that sounds good to me. We are in the summer, so I think we require more breaks than usual. No problem. It's freezing cold world. Where are you living, Peter? Freezing cold. I live, you know I'd pay very cheap rent that's all I can say and But you are living in Australia! It's in the southern part of Australia, Melbourne, yeah.
To be honest with you, Australia for me is a mystery. I hear this, Australia I said that to me is a mystery. I hear these contradictory reports, it's too cold and it's too hot. can it be too cold and too hot at the same time it's really big yes i know but it gets cold at night there right it's it's 2 a.m yeah that's all right sounds okay let's have let's have a let's have a break all right
Oh, so tired today. You've been doing lots of yard work? Yesterday I dug a few holes, one for myself by the way. But yes, you should come see the garden is actually finally taking shape.
Oh, awesome. Yeah, I'm hoping to come sometime in August. Superb, yeah, that would be good. If you can come in mid-August, because at the end of August we are making a trip, that would be fantastic. I'm still kind of confused how exactly under modal constraints you'd identify a law. Unless you did something like you subjectivized the law that Wei Khan does. No, no, no. You see, modal vocabularies are essentially logical vocabularies. They are essentially a subset of logical vocabularies.
And the thing is that what you might call to be this whole discussion or controversy between the rationalist modulists, those who actually are proponents of the modal revolution, versus the kind of empiricist, whether as sophisticated as Hemphill or as naive as Hume, this whole controversy boils down to the idea that there is no such a thing as a causation or regularity based on pure experience or experiential datum. It requires something else for us to even realize that this is a regularity and to identify
it and to also extract it as such. So essentially in that sense I would say even Kant counts as an objectivist rather than a subjectivist precisely because he already thinks that causation always bound all notions of causation always bound the notion of categories okay and to that extent and in some words categories are intersubjective namely objective in the philosophical sense then is he should be counted as objectivity of course it's not that I would say unlike Hosserl or Hegel that
much observance of what objectivity actually means but that's a whole different question right yeah I would agree with you on that to raise I've been thinking about that lately too that it just seems like you can't subject subjectivity and can't doesn't even happen until you have like robust objectivity yes absolutely absolutely absolutely I mean the whole idea about perceptive subject which is the subject this subject does not even arise this issue doesn't arise unless unless you have objectivity. Absolutely, yes, that is absolutely true.
I mean, it's kind of the same thing though. I mean, it's kind of like this illusion that there was ever subjectivity without objectivity though. Yes. Like Stoicism or Confucianism. Yes, yes. And then you see this in the vein of post-structuralism and all the kind of affect theory and so on. and so on and so forth, where the subject becomes glorified, over-glorified, to the extent that subject is glorified to the detriment of objectivity, and that's when basically I would say is actually what you might call to be the peculiar characteristics of post-structuralism
and also postmodernism. Subjectivity without objectivity doesn't mean anything. You should actually watch this fantastic debate between Ray and, God, my memory, I'm having dementia these days. What is the name of that philosopher who wrote Hegel and Plasticity, Catherine Malibu. So it's a debate between Ray and Catherine Malibu, where Catherine Malibu brings up the Bergsonian idea of subjectivity and experience,
and Ray gets really upset about it. And he goes on and on to talk about that, this is just bullshit, this is really bullshit. There is no subjectivity without objectivity. There is no subject without that which make the object intelligent. Hegel, I think one of the greatest moments of Hegel is to show that why this is so important, why we cannot think of the subject without thinking of it, as an object. With understanding that the object
always requires the labor, the concrete labor of making things intelligible. They interplay in Kant between categories and sensations and in Hegel between the logic for reason and experience. I'm not sure if my question was clear before with the human question of empiricism. Because it seems like, I'm not sure, your feedback was really good though, and it's a lot to think about, but that somehow by imposing
um you know induction on our experience that that that it's sort of like you're limiting the power of logic and it seems to me that there's something like intrinsically carnappian about that that if your experience is logical and your experience is not predictable and you cannot apply induction to it then clearly your logic is flawed something like that yes but carnaps Now, this is what I would, before everyone comes, this I would say corresponds to the third Hempellian paradigm, which is called IS model, short for inductive statistical
model. So he has three major paradigms of explanation. DN, short for deductive numerological model, DS, deductive statistical model, and IS, short for inductive statistical model. Now what are inductive statistical models? It's the idea that what if we could model all such logical relationships in terms of probability theory or a statistical, modern statistical theory. And this is essentially the Carnap's position, at least in the, not in later works, what
you might call to be the middle period. This I would say that to be honest with you, I really don't have any response to this. I think this requires a lot of research for me to actually be capable of with semblance of confidence to answer this question. because you know Valentina and I were talking about this in the previous sessions what if all such logical connections can actually be models a modern a statistical terms this is essentially
what Carnap calls inductive logic. He doesn't say inductivism or induction in the union sense, inductive logic. These are, I think these are open, these are open. I mean, And these are absolutely, I would say, they are susceptible to further research. But then also, like, I was thinking about something else you were saying a couple sessions ago about, like, when you saw tire tracks and, like, you know, about how you could apply that sort of statistical model to that.
But it's also like when you see tire tracks from a linguistic perspective, there's these webs of semantic meaning. Yes, yes. You know what I mean? Absolutely, absolutely. I mean, you should know that this talk is being covered quite extensively in the chapter third of Intelligence and Spirit. And to this extent, right now, I'm still taking side with the rationalists rather than empiricists like Hempo or the kind of logical empiricists like Karna because I think that this really
doesn't make sense to me from a philosophical perspective that you can have a logical probability where you can actually talk about nomological laws because these nomological laws are essentially linguistic not but when I say linguistic I do not mean natural language as such I mean just language as the organon of logical vocabularies of which modal vocabularies is a subset. But then those things would be like modal primitives though, wouldn't they? Like you
have modal primitives of language. I think that absolutely you have modal primitives all the way through yes you have always more of primitives i i i i what i to be honest with you you know you you i mean you you guys probably know me uh you know i'm not set on anything literally i'm not set on anything i can tomorrow morning i said that whatever i said was let's go to the next level but for for the time being no matter how much i have tried to examine the problem of empiricism i think it falls short precisely because the modal issue is fundamental it's absolutely fundamental at least uh to me currently
I just want to say one last thing because I don't want to take up the whole class like I like listening better but that's one of the things I know you like Confucianism and that's one of the really interesting things about Confucianism is it's all in these deep webs of the tradition so it's all really you know so it's like yes and to be honest with you Chagi the thing is that I think the more you look at Confucians and the Stoics, there are what you might call to be no rationalist version 100 AD or 200 BC. This is absolutely true. I mean, they absolutely, in every... when they talk they absolutely are not ignorant of the
empirical constraints. It's just that they think that empirical constraints by themselves don't mean anything, can't tell us anything about what we should do in the world if it is thus and so. We have to tell ourselves. That's pretty much what you're trying to say. I mean, it sounds you know sort of lame but yeah we'll talk about that later I guess questions if you have any questions nothing I think there are a few of you I've told you already who are so silent
i know why you are silent but there's no reason to be silent in such an extended session you know we are all friends there is no reason to feel shy so i know you you know who you are so please do talk and make a question at this point Nothing, please someone, something. God it's so hard to get a word from your mouth these days.
Okay, okay. Let's forget about it. So the second paradigm of Hemphill is what he calls, oh by the way, before I move to that, with regard to DN, deductive nomological model, you can think of it as, so the thing is that deductive pneumological model is, Temple thinks that with regard to at least a pneumological component of this model, we should realize that there is a difference between what he calls accidental regularities and pneumological necessities. Regularities
that are effectuated by way of accidents arbitrarily and regularities, exception regularities that happen by virtue of what we might call a law. A law is always a logical thing even though it feeds upon empirical evidence. So let me give you an example. So I go to this room and I see everyone is a bachelor. Everyone is a bachelor. I say that, you know, this is school at year 1989 when I visited, everyone was a bachelor.
So this is also a regularity, okay? It's a regularity. Everyone's a bachelor. It's always, it implies a uniformity of cases. But then what would be the difference between this and something like a nomological necessity such as Newton's equations of motion? can anyone tell me about what would be the difference between the two between what rather sorry between for example when i go into this room at this college year 1989
and I observed that all individuals sitting in this room are bachelor this and something like pneumological laws of motion as advanced by Newtonian mechanic what would be the difference between the two One thing is called arbitrary and one called nomological necessity. That the Newton's laws are just eternal. Like they're not time specific. But then a question arises.
What would exactly make the first case time specific? the observation of bachelors in the college room at the year 1989 because you couldn't re-observe that same phenomena no but something close you see when I apply the laws of Newton named equations of motion I can explain every single of course except exceptions might arise as Popper said and that's when I
have to go to a different law anomalies so called but imagine we are in Newtonian universe in the year 1800 or 1700 and i see when i apply these newtonian equations of motion i can explain every single phenomenon that i actually observes among the celestial bodies what can i use the law that every bachelor in this X college at the year 1989 is bachelor to explain why
is that this but this person n or X is a bachelor is it that uh the laws would be logically prior to the experience of the celestial bodies whereas in the face of the bachelors the laws formulated after experience no but then the other thing to raise is like the laws you explain the laws but the law itself is also what explains the law i guess it kind of is that kind of yes but well at this point uh chuggies we are not simply talking about the notion of law as we you know students kind of
various students of philosophy know it in the sense that for this example even the notion of law is literally mid-air it still needs to be actually explained to be substantiated the whole point is that the thing about regularity such as newton's equations of motions they can be used to explain that why is that this particular celestial body behaves does answer or can we use the same law that we regular regularity wasn't law
same regularity that we saw in the class 1989 where we observed that all the students were bachelors can we explain can we use this thing this regularity to explain why this specific person sitting at this specific desk is a bachelor yeah because you could just say they're single but that doesn't explain why they're single yes questions i see you are getting pensive now that's dangerous
Joven hasn't talked today, Meredith hasn't talked today, Adam hasn't talked today, and so many of other you. I don't really understand the regularity issue. So isn't like race theory the whole point of saying, oh, we can posit a kind of regularity of understanding how society functions because of these categories? And so I guess I'm not clear on... You see, that's a superb, fantastic, ultimately.
Really, actually, I'm going to rip off you on this. Use it in my new book. But this is absolutely a fantastic point. But this issue has nothing to do with the idea of the arbitrariness versus nomological necessity. It is what you might call to be the conflation, as Hempell would have said, within different paradigms of explanation. So usually what we take as a nomological law or a natural law always happens in the realm of what you might call to be either DN model, deductive nomological model, or what it calls
SD model, a statistical, sorry SN model, a statistical nomological model. But there is also a third alternative and what is this? It is what he calls IS inductive statistical model where you see a number of cases and you attribute to them probability priors. And according to these probability priors you come with a law. But of course this law is not logical, it's statistical.
It's like when I say that people who were suffering from Ebola virus and took this drug were cured according to the number of cases that I have registered on my leisure. Now this is not a logical or categorical in a Kantian sense law. It is simply a statistical induction. Of course, the same thing happens in the race discussion, but even more generally in terms
of the stereotypes. What do you think the stereotypes are? Many people have done of the same characteristics, have done the same thing. And then I bunched them together as essentially under one feature. This is what's called a stereotype. And how it is done is based on inductive statistical model. What is this? Statistical models in the inductive sense are by no means regulations or regularities in the sense of laws of nature, because laws of nature are absolutely logical as applied
to empirical evidences. For inductive statistical cases, such as the one that you mentioned and the one that I just said, we are essentially working with high probability that, for example, There is this Iranian who smuggles opine. There is this Iranian man who doesn't pay his wife because the stereotype in Iran is that men are stingy. And then there is this stereotype of a woman who is detached, another stereotype.
But then, these stereotypes are not laws, they are what you might call to be the products of attributing a certain amount of probability to certain amount of cases. And as such, they can never be understood as laws in the Hemphelian sense or in any sense. There are what you might call to be expectations. The idea of stereotypes encapsulates the idea of probabilistic expectability rather than a law in terms of Newtonian mechanics. Yeah, but Rizzo, sorry, but like, I mean, a lot of, so yeah, we're going to predominantly
like you know race theory like like the department and a lot of their work isn't not actually inductive statistic but i mean going by the hempelian you know categories actually they're trying to do like a deductive nomological explication of for example race and the way they do it is for example uh existential analyses so the the way in which our experiences are uh to be categorized um isn't bound to simply stereotypes simply isn't bound to like local uh determinations of how alterity context sensitivity of the cases yes it's not simply context sensitive yeah but yes but
instead like what is imminent to context sensitivity perhaps like what is imminent to yes but what you should realize joven that uh when you actually go deeper into the into analyzing uh such cases then you realize that even those cases are essentially predicated on a probability analysis and not in it what you might call sophisticated Carnapian inductive logic sense but in the sense that it has been you know the case since the medieval times so I would say that every woman who uses my apologies for being very very unappropriate
but this is just an example from Carl Ginsberg. I'm really, really sorry to make this. Every woman who uses a, what do you call it in English? Sorry, now I forgot. The thing, what do sorcerers or witches write on? Brooms? Brooms, yes. If you use a broom between your legs, then you are a witch and hence you need to be drowned. If you don't drown, then means that you are not witch. But if you actually drown, you are a witch.
This was actually the paradigm back in the medieval age. But this is, you see, the whole point of this thing is that it's just a bad account of probability. It's a bad account of probability. It's what you might call to be a critical account of probability. Precisely because it's just based on individual cases and generalization of such cases. and then generalizing such case once you generalize this enough according to your own cognitive biases to everyone and you think that it's a law to identify a
witch and burn it yeah I think that there's a difference in the sense that that within the Hampellean paradigms or the Hampellean categories, we're still talking about transcendental identities. We are trying to figure out how do these identities relate to ontology, what kind of knowledge is appropriate. But I think within race theory, it's a very different kind of... It uses ontology as, you know, it's ontological unity that comes to… Well, you see, with race theory, it is not just about the idea of a law.
It is about many erroneous assumptions that they should be encountered and tackled with in their own terms. like the very notion that as if cosmological laws could be applied all the way up you say this you get this actually Nick where the Hamiltonian conservation on mechanic can be applied to Darwinian mechanics and Darwinian mechanics can be applied to society so these are all presumptions what is the explanation behind it what what allows you to make such bridges you see what
I'm just trying to say is that yes the core is about this kind of a statistical inductive idea turned into the notion of law but of course it needs for for something like race theory but even more importantly like a rational rational choice theory where you think that our desires are all about competitions and we can make the greatest society simply out of individualistic methodologically individualistic competitions. Such paradigms are not just about laws. They require actually additional erroneous assumptions and those also need to be analyzed
appropriately. Like, for example, rational choice theory, even, I would say that it's even more insidious than race theory, precisely because it creates, from Marx's point of view, the very condition of exploitation. The condition of exploitation becomes nationalized, okay? So in that regard, what do you think is responsible for something like rational choice theory or neoliberalism? Well, it is the idea of what Hempel already calls in one of his books, I can't remember
exactly, the idea of transitiveness. we can simply move the kinds of laws or statistical or deductive inductions that we have observed at a certain scale like at the level of the individual people to the level of market and up to the level of society this idea of transitiveness itself is a presupposition It is not guaranteed by anything, not even by the very means that you as a kind of rational choice theorist employ.
one driven and let me tell you that everything I'm saying is simply under construction I might actually call with different you to write one next Saturday I agree with you I mean the question you know that I think you have to like address is under what are you like reclarifying the problem of race you know are you reclarifying the identity of
yes these problems you know like yes yeah yes the whole point is that you know as long as you see that we should not you know when we are criticized a stereotypes we should also know that a stereotype actually hold a grain of truth a grain this is a really important grain and more of truth not the whole truth and as such this should be taken seriously when there is an anomaly there is an explanation but this explanation cannot be explained or put forward simply by stereotypes when you see that for example let me be quite blatant on
this being a Middle Eastern I don't feel that I should have any kind of fear about making such a stereotype when you see Islamic people or Muslim do the major terrorist attacks in the world then there is obviously a grain of truth yes it's a stereotype and as such it should be constrained to its own scope under its own conditions but then it means that there is actually an explanation and that's how a scientist goes on and shows that okay
you know there are also these people christians white who make terrorist attacks but then why you know we observe muslims as being terrorists that also requires an explanation but the explanation cannot be done by simply a recourse to the various stereotype that we have made So he's saying like the law is, we have to be careful what we're calling a law, but a law would be something logical, but the law itself is subject to conditions beyond the law that dictate how it's to be developed.
develop because because the law is actually and this is really I would say that is absolutely the gift of the transcendence of philosophy Allah current that law is nothing but a species of a logical category applied to particular evidence we can never trump particular evidence as if they were logical categories in order to count them as law for something to be counted the law, we should go beyond empirical evidence as such.
Do you think that the deductive nomological model is pretty critical in that way? In certain sense, but I wouldn't say that in entirety, precisely because it is not fully precritical because it takes the idea of deductive logic very seriously as essentially the chain that allows for us to bridge explainants to explanando and therefore it is not human pre-critical paradigm sure it has pre-critical components but as a whole I wouldn't say that's
pretty critical. Because I guess the question for me is are the laws formulated even logically? Well, I guess it doesn't even make sense to talk about a law being formulated logically somehow being a generalization. Yeah, well generalization, Hempel absolutely doesn't believe in such thing. should always be specified under what exact logical and empirical constraints you make such generalization. So there is no such a thing as a general generalization.
Generalization always comes with specific constraints. What sets apart science from the ordinary talks of witch burning? It seems like at a certain point though, the very laws that would be changed by the empirical evidence are also dictated by laws. So you have to be able to determine that something is something that's challenging your view of the application of the law, right? Yes, yes, and that's something that I thought that Theo actually kind of mentioned in his last comment, if you look at the sidebar, that Newton's laws are non-falsifiable.
No, it's not about non-falsifiability. It is about that this is exactly like the Konyan sentence or even proper really. That this is how we have, you know, we have some logical categories, we have some empirical evidences and we do study the interplay between the two and we extract a generalization or regularity of empirical cases for natural regularities exception exception less regularities in an intelligent sense but then of course
Of course, we might come across a number once we actually extend the sector of experience, of reality. We might come across not just one, not just two, not just many, but numerous instances of what Korn calls anomalies, examples which no longer can be corroborated or abused by our law. And that's when the time that we have to change our law.
That means that change or revision of laws is essentially a theoretical enterprise rather than an empirical thing. Yeah, yeah, because it would never at any point be empirical because it's an anomaly. Yes. Yeah, yeah. Yes, and the whole idea that you, in fact, you try to identify this anomaly and somehow relate it to your existing law like Newtonian mechanics. And then by virtue of that, you notice that upon the observation of such numerous anomalies, then maybe my law no longer apply across the board universally.
Either I should downgrade my law to a different context which is more conservative, like for example the terrestrial gravitational field, or I should expand my law by revising Newtonian mechanics in order to accommodate those anomalies. Such even thoughts requires theory. They are not empirical in any sense. Yeah, it's like, I was kind of wondering that too. Now I'm imagining sort of like a graveyard of laws. But like what happens to these laws? Like if a law is permanent, and it, I mean, it can be challenged by the empirical,
but you're saying the whole idea is that the law isn't permanent. it's that it's this yes and the identification of the impermanency of a law is not something that simply arise because I have observed it's because I can actually relate or correlate the empirical observations I have doesn't so observed by way of a theory by where a structure to the existing law I have and then I see that poses a threat to the existing law maybe I should revise it
so so laws don't just persist as they were but they actually are constantly evolving like you don't just develop a whole new law it's like this absolutely absolutely absolutely absolutely okay one second give me give me one second I'm coming don't go anywhere okay Chagis I was just trying to understand what I thought was your question earlier was which was i'm just going to try and restate it you can correct me if i'm getting it wrong but if our experience is rational to begin with and that rational structure is in some way what
allows for our experience to take place um yeah then how is it even that we need to now then reapply rational structure onto experience which we say that's what that's kind of was my big challenge way earlier in the class was like I mean I was kind of just looking at it from a Kantian perspective like if there are just these categories of experience and you know like at what point can you ever challenge that I think that comes back to me for it's sort of this how would rationalism even admit of paradigm shifts or anomaly if we're saying that my apologies oh Oh, that's interesting. By the way, Chuggies, I'm going to rip up and plagiarize your word.
Graveyard of Laws is going to be my next book. It seems that Joven is already trying to rip it up. That's what you want to mean, sorry. So I have to fight with Joven about the rights. but then i was imagining like these like i don't know like these like just these parallel lines that are just kind of stacking up somewhere yes yeah to to take like the graveyard of law thing a little seriously i mean it is this not just you know in the end a hyper like theorization of concept and so far as your our concepts have to
find some kind of like absolute relation in some way to the object in which we can finally have access to objectivity and is that really like the project? I think it's actually quite different. You see even in Kant You know that in Kant the word concept is concepts with a small c, whereas in Hegel concept is capital c, they are fundamentally different. The difference is that for Kant concepts are such logical entities upon which we exercise
our cognitive faculties. Whereas for Hegel, concepts with capital C is essentially the platform upon which we make our cognitive faculties. There is a difference between Kant and Hegel ultimately with regard to the word concept now in that regard i would say that even in khan's concept is essentially actually a logical species a logical entity and to apply it correctly
to such and such empirical evidences require something like a kind of midway between logic and sensible intuition, between categories and empirical evidence. So even I would say that even in the Kantian sense, we can't just simply think of concepts as this free floating stuff that are basically exploitive because they can actually constitute something like what you said, race realism or social truth theory, so on and so forth.
No, concepts, ultimately the primitives for both of these thinkers. essentially, you should realize, is what sets apart the transcendental turn or modern philosophy from pre-critical philosophy. So Theo was just asking, like, if we have this conceptual Kantian foundation, how can anomalies even appear? And I would say, like, for me, like, one of the things I've been thinking about, I think of monsters, ethical monsters, where your own values are pitted against themselves and you find yourself in a tragedy. Yeah, absolutely. I mean, sure, absolutely. And I think that axiological ethics is also a great example for the discussion that we
had. But let me tell you that, okay, so you are in the business of Newtonian mechanics. then you realize or not even Newton we are we are in the Keplerian mechanics so Kepler does not believe the Sun is at the center of the universe in contrast with Copernicus okay so essentially the real Copernicus evolution is starts with Kepler ascent to the higher heavens you know but then you notice that according to
these formulations or these equations which are couched in my Keplerian theory and of course I am always conceptualizing I then see via a telescope trajectory of a celestial body a bollite which absolutely does not conform to my equations then I see every other boat light and I see that they don't correspond to my equations this non correspondence is not an empirical injunction it is simply something that according to my theory the way I
conceptualize about the celestial bodies I no longer see bollies as following the equations of motion counts in terms of my Keplerian formulas of motion. You're saying bullets right? Bullets? Bolide, bolide, bolide. What's bolide? It's kind of like a shooting star. Oh, okay, okay. You should know that they were extremely important for early astronomers and scientists, precisely because insofar as early astronomers and scientists were still somehow attached to
the medieval, you know, kind of superstitions. boolides represented what in modern terms we call omens. Okay, okay, I see. Okay, okay. So they were really like not only anomaly in a scientific sense, but anomaly in an anthropological sense. Yeah. Reza, do you think we can connect this up to your paper?
Yes. Give me 10 minutes and we will connect it. Plug it. Okay. So, DN, I mentioned it. So other than the N deductive numerological model, Hempel also puts forward two other models. One is called DX and one is called IS, respectively deductive statistical model and inductive statistical model. So the difference between these two is that the deductive statistical model, we essentially try to derive a semblance of a law.
It is not nomological by any sense though. We try to derive a semblance of a law or regularity using as the premises of our deductive argument where we have premises and consequences where at least one of our premises is actually a more generalized statistical law. So I have such and such evidences, I have such a statistical generalized statistical law provable caution in terms of monimprovalent theory then according to the deductive logic
i would be able to derive from it a narrower statistical law applied to a narrower set of empirical evidence or cases so this is what he calls ds or deductive statistical law for model now the third one is called inductive statistical model is in which we no longer actually derive a law it's simply the case that we apply an statistical probability
in terms of a law post-eventus, i.e. after the effect. So certain kind of empirical cases, like, for example, these people in World War I have gangrene. Of course, at that time, they really didn't have any notion of the coherent notion of what gangrene means. Let's say they have infections. So I administered penicillin to them.
Many of these people actually get recovered from their infection. So this is an example of inductive statistical model. It is not a law, it is only a probability. We say, according to this high probability, it is possible that such cases, once applied such kind of stuff, will result in such conditions, according to this specific probability.
This doesn't mean that it's exceptionless. You see, here for Hempel at least, exceptionless is what says support, nomological, deductive nomological model from inductive statistical model. Another thing that I should say, that notion of pneumologic is all about expectability. Like the example I made that I'm kicking the table with a sufficient force and this glass
pulse and the liquids pours out of it. Here explanation also that nomological factor that we were talking about is about expectability. I expect the phenomenon would be once such causal or pneumological factors are in place. However this also brings to the discussion that we had just earlier that this means that if like Hemphill we define laws in terms of pneumological expectability then virtually like Karna we might in parenthesis we might be able
to talk about every law in terms of the probability their statistical probability of our expectation of what phenomenon arises from what explainants so there's never any deductive reasoning about anything for him pull is that what you're saying no no no for him hell actually thinks that deductive normal logical model is
fundamentally different from inductive a statistical model precisely because he is not he does not believe in any kind of probability theory the way the car naps inductive logic things but that's it fundamentally different the story and And that really brings us to this whole discussion about deductivism versus inductivism. So inductivism can be thought in terms of the enumeration of the empirical evidence from which we can, without the intervention of any logical evidence,
derive a law or a law-like regularity. For logical empiricists such as Carnap, however, this is a little bit more sophisticated. Now, we, yes, we do need logical components in order to derive a law-like regularity from the enumeration of empirical evidences or observations. However, such derivation doesn't need to go through the course of deductivist logic or deductive premise versus the consequent logic.
It can be ultimately couched in terms of modern probability theory. This is for Kata. For Popper, however, this is absolutely impossible. It just doesn't work. Now I think this whole paper that I wrote is an assault both on Popper and Carnap in the sense that really the idea of derivation of a law or regularity or uniform exceptionless regularity cannot be derived from either empirical evidences a la Hume or
empirical evidences couch in terms of a probability logic a la Karna or deductive connections between the statesmiths made about the empirical evidences and the state the law like statement which is the consequence of your deductive formula questions okay let's not talk about my paper I'm not going to talk about it I
I take for granted that some of you might have read it. So let's start with questions and we go through them. And hopefully for those who haven't read it, we can make these connections more clear. So I mean, essentially, Hempel is just saying that no matter whatever law you're claiming to have, always just predicated upon statistical probability. So it's never, in the sense that we usually talk about laws, a law. Yes, yes. But you should understand that for Hempel, he essentially believes in the tripatriate kind of paradigm of explanation
in the sense that this is just one paradigm. but in the deductive formula or the DN model deductive numerological formula literally your observations don't mean anything unless they have been already caught up in a theoretical edifice where you actually can make the logical structure within such atomic or airlapsed statements clear or explicit. But like the theoretical structure, how does he characterize the theoretical structure? You see, at least as far as I know from Hempel is that Hempel
comes from a structural exposition but his ultimate what you might call to be allegiance with empiricism so for Hempel thinks absolutely like like Karna or what I actually talk about a stigmular where such logical structures are essentially components of the theory theory as the organ of the structuration object is nothing but the structure or theory so Hample is actually a very different kind
of case that's why i have chosen to use him as an example precisely because he neither follows in your carnapian paradigm nor a stegmullerian paradigm nor human paradigm he's a silly empiricist but an empiricist who tries to take conclusions of empiricism towards the vindication of empiricism by way of the question of logical instruction. So it's interesting it's almost like he's like in a way almost like I don't know if I'm interpreting this correctly but he's
saying like logic itself is empirical almost Yes he doesn't say that ever but he does believe that logical vocabularies can be bottomed out into base empirical vocabularies. Yes. Yeah. In a way it almost like sometimes I'm tempted to interpret Plato in this way. He's making this radical claim that like logic itself is this empirical non-human entity or something like that. And when I read Locke sometimes I get that sort of a sense from him too. luck luck i would say is far more closer to hempel not not plato not plato plato absolutely
thinks that everything ultimately is subordinated to the logical forms literally reality doesn't mean anything being is not existent if there is no reason he's a parmenidian he's a parmenidium in that sense you know parmenidium was the first rationalist but also he he also was the prince of being he talked about being at length but for him being is simply the predication of reason of the logical resources by which we predicate
Reza you said your paper was a an attack or on Khan app and was the other one I failed attack I failed attack I failed but it was Khan app and a car nap particularly but when when you read it basically so I do recognize the objections made against Karna because I thought the paper was more of an attack
on Hada and that that kind of yes yes absolutely but what you should realize that there is a genealogy, a historical genealogy here. So the paper is an attack on Karnak, but then I actually, at least in the version that I use for the book, I recognize that Karnak is not as blatant as Stegmuller or Putnam tried to make him be. However, you You should realize that Solominov attended seminars of Karna on inductive probability
or inductive logic. And Hutter actually read Solominov's paper. So there is this lineage of problems. But I don't think that we can attack Hutter in the same way that we can attack Karnab. Because essentially these two thinkers have fundamentally different contextual constraints on their theories. I would say that Karnab is far more sophisticated, whereas Hutter is kind of like this, you know,
over excited computer scientist trying to use some philosophical dogmas as if they were onto logical facts of reality. And Hempel appears here as like a moderating voice almost or something like moderating logical positivism into something more flexible. Yes, yes, absolutely, absolutely. But nevertheless you should realize that, okay, Hempel also who, you know, really quite extensively revised his position from his earlier works to his
later works, exactly like Karna. But essentially his position can be encapsulated as this kind of critique of logical empiricism is not destructive rather it is productive it is still in the vein of logical empiricism because to the end of his life temple actually defended the idea of a dn model deductive pneumological model and what is deductive
nomological model at the end of the day it's literally the core tenet of it has been given by logical empiricists by the Vienna circle and so what's your problem with Karnat is specifically around the problem of induction. Yes. As being in the sense that... Rationality or something. Yes, yes, absolutely, absolutely. And there are many problems. Let me just give you one. You see, in Carnap's formulation of inductive logic,
so you get C parenthesis H comma R equals whatever measure function now the components of your premises are essentially exactly like him and evidences plus a kind of logical framework it can be couched in terms of anything usually probability theory and from that you conclude that there is such a thing as a normal logical law or regularity now the thing is that a signal eric thing really pins down this problem quite
well first of all you should understand that Karna is by no means human you see when he talks about what he calls evidence or ES statements they are actually logical statements rather empirical evidence is he stands for empirical statements rather than human way of empirical observations in the sense that Carnap does not believe unlike Hume that simply we can derive a law
like regularity from the enumeration of empirical observations the only way that that we can do this is by a way of understanding that these empirical evidences should be taken as a reference reference inside a probability logic so in Carnapian formulation E which denotes the evidence or empirical evidence is not human raw sense that it is rather a reference made to such enumerations of
empirical observations within an inductive logic and inductive logic logic being it's a lot here so this is a very big difference between Carnap and you however a signaler shows that even in this sophisticated theoretical scenario where e-evidence or ES statements are not just a raw empirical observation or since data allow you but actually references made to such observations within the round of an inductive probability logic it's still
something is missing and what is that it is the idea that how can you actually marry the unobserved future probability to the observed past probability that you have observed. From a logical point of view, the relation between that which is observed in the past
And that which is yet unobserved in the future is essentially a disjoint set, meaning that the conjunction between observed sets, how will you interpret it? an unobserved set is an empty set is a disjoint set simply we what does this mean simply we cannot move whether in the realm of brute human empiricism or in the realm of the logical empiricists such as Karna
move from the unobserved sorry, from the observed to the unobserved from past to the future I got two questions on it one, why we do that anyway we do that, yes we've got along pretty fine so why couldn't why couldn't an artificial intelligence for instance or a model of rationality be built on that even if it was well a stakeholder has a really fantastic response to this without knowing anything about AI and HGI of course this is already what we do this is what we call the course of heuristics right yeah it is
heuristics but then if it is heuristics then it is not the principle of general intelligence it is simply an image of our own transcendental subjectivity projected onto the realm of a formal intelligence yeah i mean a formal intelligence as such and hence it cannot be thought as universal intelligence. And ethics too, like if you look at just like Aristotelian ethics or any ethics, you're never just like just because Socrates committed suicide you shouldn't just do that, you know it always requires sort of like a negotiation by the agent. Yes absolutely, yeah absolutely, you will negotiate but I actually prefer to go with the Socratic method of suicide.
literally this world is so obscene that I don't I'm not really willing to negotiate with anything with them at this point yes absolutely this is true yes yes Reza one other thing with that like this is probably this isn't actually that important but I just wanted in your paper like why why do you conceive of Hume's law only temporarily and not say spatially you know we don't know that the laws that we identify here with our scientific instruments will hold elsewhere in the universe in a sense as
well but I just wondered I don't know it's not really important but no that is that is actually a question well let me tell you this that at least the way that that human formulates his idea of regularity is based on time or temporality namely the succession of sensory impressions one after another to that extent we can make a synchronic diachronic regularity impression of such sensory impressions this is exactly what you but as Boltzmann observed this is already actually restricts the idea of other forms of intelligences
other forms of beings where they can might have actually a different conception of time and memory and in that sense then i would say the humane empiricism even though it looks radical is actually quite conservative because it's to a very particular conception of temporality which is our own image of temporality yep yep yeah because it's I mean like that was exactly what I was thinking too you know it's like this whole idea
of the unobserved is somehow in the future or something right like and for us it always is but then it's like there's never even an unobserved then too yes yes absolutely absolutely absolutely yeah and this is this is where actually can't critique comes through a shines through and despite what i have told you about how i feel about kantian critique of aliens i think it actually falls short short we cannot talk about aliens because that would amount on you know armchair speculation about what other kinds of intelligences might be there in the universe such talks should be relegated to theology rather
than serious science or philosophy however there is I think a grain of truth in such things because what if our way of representation of temporality for time or our structure of memory is simply the product of our arbitrary contingent constitution literally in a Darwinian sense the product the results of our own a specific evolutionary course but in so far as in Darwinian
time deep time we can understand different species evolving across across different kinds of evolutionary principles. So one cannot start to think seriously about the possibility of other kinds of intelligence. They might have different, fundamentally different structures of memory and representations and a space and time. Of course, such talks cannot be talked by way of speculation. And that's why I absolutely against peoples like Scott Baker or David Roden disconnection thesis when they see the future intelligence as this kind of unbound disconnected intelligence.
No, no, no, no. This is just pure metaphysical wankery, I would say. This is something that absolutely should be done by way of scientifically informed philosophy. And I think that we have due resources in the history of science to actually challenge our own intuitive biases at this point. Not everything, but at least certain kinds of things. but I mean I wonder if it's possible though like kind of thinking about something like a stoic sage within let's just say we are never going to perform that transcendental jailbreak and we are
stuck in this humean cell and you know like Russell says then there's nothing there's no difference between you know the madman and the sane man so I mean could we have a sage in those conditions within this human confinement? I don't think so. I really don't think so. I think our sage would be just what you might call to be the best exemplar of our own biases. Where our biases finally made explicit. They haven't been revised. They haven't been challenged. In fact, I've been left unchallenged. So Reza, I have a question for you now relating to HEMPL
and that meme that you made about, you know, because you're able, I'm just paraphrasing it, but because you can discriminate stuff that doesn't permit you to say what the nature of that stuff is, so how would that apply to HEMPL in our discussion? Well, I don't think that Hemphill is... you see, Hemphill, even though he changes his views, I don't think that he can... by virtue of his underlying principles behind this philosophy, he never actually questions such things. It doesn't even arise as a problem to him. Okay? This is absolutely, and I would say as an alternative to Hempel, read someone like David
Armstrong. What is a natural law? His book. I thought about that actually. Yeah. It's a good alternative to Hempel. Okay, my friends. I'm afraid that I am called to the garden. You better go out doing your own stuff and let me do my digging. Sounds good. I love the broadcast now. Dig some holes. First, I'm going to dig a big hole for myself. It's really important.