 Okay, so thank you everybody for being to be there for the first session of our seminar in metaphysics, meta-metaphysics and sciences It's my pleasure to reintroduce a good friend of the Cephyses Chars Saint-Germier used to be a post-doc here with Peter Rodzik. He is now a senior researcher in Herkham as a civil university working in the philosophy of music, musicology, and still doing a lot of logic and metaphysics this year, and Here he would talk for us about the humus maxim, the entire nationality, and the strong sense of impossibility Thank you Kevin So I'm very happy to be back here and to have been invited in the meta-metaphysics, in a new meta-metaphysics seminar which is a big change for me because I was more used to to the logic seminar at the Cephyses, but I will try to meet the demands of this meta-metaphysics So at first I was a bit anxious about what I could present because I haven't done much recent work in meta-metaphysics to the extent that I have ever done any serious work in meta-metaphysics, but after I think I opened two masterclasses with some handbook of meta-metaphysics, and I saw that the epistemology of meta of modality was part of meta-metaphysics, so I guess this talk, which is mostly a talk in the epistemology of modality with a slight formal dent, will do the trick and I try to provide examples relevant to metaphysics and the philosophy of science I hope you might be able to discuss some more examples during the Q&A So this is the outline of the talk but this is usually this should be a very natural flow. I will explain what my problem is and how I try to solve it, what the challenges are, how I meet those challenges Let me start with Hume's maxim, so there are many things Hume said, so you might have heard of Hume's dictum which is something about the necessary existence Hume's maxim is about the connection between considerably and positivity So this is the famous quote from the two types of human nature It is an established maxim in metaphysics that whatever the mind clearly consumes includes the idea of possible existence or in other words that nothing we imagine is absolutely impossible So I don't think this idea of reasoning about possibility Values of continuity is completely original to Hume and he doesn't claim it to be original He says that it's an established maxim and we can find it in other authors in early modern philosophy for him, especially Descartes But he was the one at least he was one of those who put it in a very clear formulation that served as a basis for discussion, so I'm going to simplify it a lot and excuses for the scholars of early modern philosophy because I put away very interesting stuff like clear conceivability clear and distinct ideas I would love to be able to say more stuff on that At some point I had ideas about how to cash out formally But for the sake of simplicity, I will leave that away and just talk about conceivability Simplicita And so the kind of things I will be interested I will be discussing today is connection between conceivability and possibility So basically there are two questions here First, what does it mean? What is the meaning of that maxim? Which means which entails answering the question what does conceivability mean? And of course it's true And I will not give like of course like a conclusive answer to both questions, but I will approach both questions First by giving a sort of a formal explication of conceivability Which builds on tools that are not original to me, but I think the The final formalization is original to me at least I haven't seen it anywhere else in the literature And I will have something to say not about whether it's absolutely true, but in which conditions It's true and in which conditions fails The idea being that if we master those conditions, then we can use it in an interesting sense epistemologically So you don't need to say that it's true in order to say that it's epistemologically useful This will be one of the Of the of the main points I will try to make In order to bring some context, I think use maxim so A lot of people hate it. A lot of people have criticized it since the beginning So Read after you after you just criticized humans maxim A lot of recurrent arguments that you can find by authors But still As Javel pointed out in his famous paper it's conceivability guide to possibility There's this kind of schizophrenia here because a lot of philosophers are prone to say, ah, well, of course the fact that it's conceivable does not mean that it's possible But explicitly or implicitly this sort of move is used all the time in philosophy because it's very hard to give a direct argument for a possibility for a non-obvious possibility And continuity is one of the tools at so now the epistemology of modality has developed and By no means conceivability accounts are the dominant account or even the only account. So there are a lot of alternative accounts For the sake of Time I will my talk is only focused about humans maxim is not that I disregard all the other accounts as promising not at all But this turns out that this is what I want what I've been talking about and what I have to say And I think there are still some things to say about humans maxim. So I'll try to say them today Uh, just give examples Of uses of humans maxim in the recent The recent literature of of course the famous example is is a charmos is a zombie argument So here, uh, the phi you can the big phi could be read as the Total description of all the physical facts and the psi could be thought as the big psi as the Complete descriptions of all the psychological facts so, uh conceiving That the big phi holds and the and not the psi holds is like conceiving a zombie world Well, which is exactly physically exactly like ours, but with no phenomenal consciousness at all And then you use the move from possibility of possibility to Say that if it's considered zombie worlds are if a zombie world is considerable and it's possible And if it's possible, of course, it's physicalism is false because physicalism at at least entails the supervenience of the phenomenal property on the physical property. So if it's possible to have all the Physical facts and none of the phenomenal facts Then of course, uh physicalism is false Peter, I see you, uh frowning. So I guess I said something wrong or Not clear enough So price so so far is All physical actual actual physical facts Psi all actual phenomenal facts Uh, maybe I said psychological. That was a slip of the tongue. I meant Phenomenal so facts about what appears to Two people or you can restrict it to yourself if you Yes, because if physicalism is true then, uh, the phenomenal is part of the physical world So or at least is necessitated by it or it's super wins on it So you cannot have a phenomenal difference without physical difference Then it's a sort of eliminate physicalism, maybe but but then the if it's an illusion then the psi is empty Uh, well This this what this was meant as an argument for not not directed at as people who believe it's Uh, uh, yeah exactly Who want yeah those sort of physicalities who believe these phenomenal facts are supervenient on uh physical facts Second sort of uh examples. So in in the in the zombie argument, uh, the move from possibility to possibility is explicit So charmeros makes that move and he's spent a lot of energy Uh giving a theory that uh justifies that that particular move The other case is where the move is merdian piece it So here I I give reconstruct logical reconstruction of the structure of study experiments at least a wide class of study experiments Uh, there's a there's a literature on this So I give the the final reconstruction by moulder and mule Uh, which is a slight improvement on the hackewist original model But the idea is that a lot of thought experiments are used to destroy Theories or to criticize theories by showing that some things are There are some possibilities that are incompatible with the original theory. So that's the Main idea, but in the details what you need in order to to have a thought experiment is a theory So here I would try to grab the example with a Schrodinger's cat I hope I don't say anything uh too inexact about the physics So as is supposed to be a Copenhagen interpretation of uh quantum mechanics The content possibility is a scenario described by Schrodinger Or you have a radioactive substance With a decay within an hour connected to Geiger counter connected to a Flask of poison which explodes if a particle comes out And uh as soon as The poison explodes it intoxicates a cat who dies instantaneously um Then you have a conditional which says that Copenhagen interpretation entails that uh if That possibility were the case Then the cat would be both I mean in a super post state of being both alive and dead Of course in that scenario it's not the case that the cat is uh Both alive and dead have to be one or the other And then you have a bridge that connects the two uh the two counterfactuals which is required for the To have a to have an inconsistency But when you get you get all the pieces of the of the of this Of this not experiment, uh, you have any compatibility between the possibility of C And uh the theory so usually you uh the fact that the scenario is the possible scenario Is used at the end of the day in order to reshoot theory s The obvious question is how do you know that the scenario is possible And then, uh There's a case to be made that simplicity what gives plausibility To the scenario or plausibility to the claim that the scenario is indeed possible Is the fact that you have described it in very clear terms, which means that you have applied the method of continuity Therefore you have provided a clear this indistinct Description of the scenario you see no inconsistency everything is clear Seems that something that could happen even though it's it might be difficult to break about for technical reasons, but in principle, this should be Um, so that's basically Part of the schizofrina so the It's time with we we run solid experiments or we actually Are convinced by solid experiments Seems like we use eumes maximum one way or on the other So we better, uh, so if you want to evaluate solid experiments Uh, we better know if eumes maximum is, uh A good good good methodological principle. I just we need to know when it works and when it doesn't work Then I say, uh, so that's for the for the for the context then, um Just a small note, uh for the aficionados About hyperintentionality It's just an additional constraint that I put on the the explanation of conceivability Uh, which some people find interesting like me, but I guess not everyone find interesting But actually that it's actually important to have a hyperintentional notion of conceivability so first hyperintentionality is a phenomenon that that you observe So hyperintentionality is the fact that contents that are necessarily equivalent Are not always inter substitutable And uh, usually they are not inter substitutable in context with our bad mental representation and conceivability is About a mental representation if anything is so you would expect it to be hyperintentional So what does it mean? So for example when we say that, uh, buddhist contexts are hyperintentional It means that when things are necessarily equivalent It doesn't mean that uh, you have to believe one if you believe the other Uh, and there are other properties, but the first one is a is a basic is a is the most basic property And um, and so there are interactions because between the claim that uh between you maximums and the hyperintentionality of conceivability So if you assume for example that conceivability is intentional, but not hyperintentional And with a very, uh Basic assumption about, uh, the logic of necessity You can derive the fact that, uh, if one necessity is conceivable, then all necessities are conceivable so if you, uh, if you, uh, if Conceivability is not hyperintentional then you will have these sort of bad results for the sake of modeling conceivability. It will be, I think highly, uh, inadequate In case you doubt that, uh Not all necessities are conceivable if one is So here's an example, uh, based on the Galileo's, uh Paradox of, uh It's one of the paradoxes of infinity So apparently, uh, Galileo was not able to, uh, to, uh, to apply notions like equality Uh, having a larger number or a smaller number for infinite collections But he could apply it to, uh Thin collections, so there, so there, but both are mathematical necessities If anything is, so there are case things that, uh, so there are, so Galileo could conceive at least one necessity And could not conceive another one, uh, in order if you want to model this kind of stuff You need a hyperintentional notion of, um Hyperintentionality, um So now there's a lot of interest in, um In hyperintentionality, but it means that, uh, the common, uh, Reflex, which is to use the standard tools of model logic Will not, uh, take you very far. You need to, uh, employ some other resources Uh Now I will, uh Describe two challenges against Jung's Maxim, one coming from Grand Priest, another one coming from Franz Bertow and Tom Schoenen Um, so we start with the Priest challenge Um, so Priest starts with the idea that, uh, he claims to be able to conceive everything That is describable in, uh In understandable terms And actually this is, uh, this is a criticism that you already find in, uh, in read right after Jung And then obviously, uh, a lot of like impossibilities can be described In meaningful terms, you don't need to be like, uh, always confused in your language when you describe impossibilities It trains now, it trains now in Nouvelle Anneur, and it doesn't right now in Nouvelle Anneur Understand perfectly what is to read, understand perfectly negation, understand perfectly conjunction And I just like asserted an impossibility Um And so there are, I gave like a very mundane examples, but there are a lot of like more like seriously interesting examples like mathematical falsehood Uh, so you should take for example, good code box, conjecture and negation Seems like you can understand both You conceive both in that like quick sense, but If he wants to and the other is it's not At least they are both not false Um When you conceive, I mean if you are like a logical monist, you believe there's one true logic, but you can still Reason according to alternative logics. So it seems like, uh, in those cases you Um describe impossible reasonings Uh, yet if the logics are well defined, uh, that's nothing you don't understand there It's possible to describe inconsistency, but first that's what I did with the rain example Um You're having consistent fictions, and of course, Chris is very happy to cite his own example of, uh The the short story about Sylvan's box, I was about to say the root list start and now it's the other thing is Sylvan's box Um, so with the box, which is both empty and uh, and not empty I remember frankly to start this And uh, for all these reasons, so if you can consider the impossible straight away like that, then of course, he was maximum is false Of course, this is just like the beginning of a of a dielectric Uh, because of course the proponents of humans maximum will, uh, will respond the number of things to this Um So one thing that you find in literature is the idea that at least the possibilities are really Possibilities, so each time you have any possibility you can re-describe it in a certain way so that it's a possibility and I will give So it's not applied for all the counter examples to use maximum, but it applies to a Significant one and I will See in a significant class of contra examples and I will discuss it a bit later. So please, uh, stay tuned Uh, another, uh rejoinder is the idea that conception is confused with supposition So we can't suppose impossible stuff and that's things we do when it's time. We do a reduction like absurdum Uh, but conception is something, uh, different It's a stronger attitude that requires more than just, uh, supposition So that's one possible way But then the owners of someone who says that is to give like a clear characterization of what additional ingredients did in conception In addition to supposition and I mean the the second point directly or like links toward the third point, which is that, uh The alert conceivable conceived impossibilities are not conceived in the right sense Uh, required by Jung's maxim, uh, and if you say that then it's on you to provide that clear sense So I guess it's on me today, uh, because that's the kind of challenge I will try to, uh To meet So regarding this idea of misdescribed, um Impossibilities so it's related to, uh, of course the the work of Kripke. So I just have one slide here We can we could talk about it for like a whole A whole semester but the idea is that, uh In the reasonings that move from consumability to possibility We when it's done with certain Things or certain a certain vocabulary Certain sort of concepts then we are prone to make confusions So here I wanted to spare my audience with the example of water and H2O because I know I have Philosopher of chemistry here that will like cry if I if I use that example So I used instead an example about proper names of jazz musicians because I like jazz Um, so as you may know, Abdullah Ibrahim, uh, it's just uh the same Jazz musician jazz pianist as dollar brand. So the South African jazz pianist just like became famous as dollar brand and then converted to Islam And how he became known as Abdullah Ibrahim and it's actually if you don't know that you Can actually at some point learn. Oh Abdullah Ibrahim is dollar brand A dollar brand just like you learn that as per se Except that you don't really learn it because you read about it When you really freely except this can happen in real life. Let's say it happened to me actually Uh, so then, uh This is this can be the basis of a counter example to the to Jung's maxim because Before I had this formation. It seems like I could conceive. Actually, I believe that Abdullah Ibrahim is not dollar brand Unfortunately, I might think I I could conceive it But of course, it's impossible because they are the same person Uh Of course, this all has to do with the with the semantics of proper names. The fact they are that they are Rigid designators And what another way to describe what is happening or to describe the contents of my conception is that It's a different sort of argument. So the first argument is like is is uh, is it valid? To produce false conclusion, but the second argument Um Says that I can conceive that the author of african marketplace Which is one famous record of Abdullah Ibrahim Recorded and then the name of Abdullah Ibrahim is not the author of children of african, which is a famous record of dollar brand So, of course, you could imagine a situation where you have two transitions who basically First of one has exactly the the life and the output of dollar brand and the second one that's exactly the life and the output of of Abdullah Ibrahim and you can imagine that there's they are The same and uh And the result for the history of jazz is exactly the same You have the same performances and the same records under the same name at the same time Uh, and of course, it's possible that to have this this association something that could have happened Uh, and so this argument is valid but what So what The mistake that people do when they, uh, do the argument from one to two is either that they could they they they confuse They may describe The possibility they really conceived which is the possibility that the author of african marketplace is not the author of children of african But then they describe it as the fact as As the fact that Abdullah Ibrahim is not a brown Which is not what you're considering considering something different And it's because of this misdescription that you get into error But if you describe things correctly, uh, you either not concede that Ibrahim is not a brown and then Don't feel the pressure to conclude that it's possible that Ibrahim is not a brown Or what you concede is just that the author of african marketplace Is not the author of children of african and then it's okay for you To conclude it's possible that the author of african marketplace is not the author of uh children of Africa But then you have not concluded that it's possible that Abdullah Ibrahim is not a brown So each time you have an inference to a new possibility and you place it with an inference to a possibility Described in slightly different terms and the reason why there's a possibility of of confusion Is because of some semantic features of proper names Uh, which is that they are, uh, rigid designators. So they refer to the same individual in all possible worlds But we when we use them we associate to them, uh Descriptive content which is contingent chosen for the actual world That's and when we confuse the two, uh, we can make those model errors This is a this is a very broad topic. Uh, I gave a very schematic presentation of it. Uh, there are there are a lot of elaborations of these ideas Of the of these models. So basically two dimensional semantics Is a way, uh, to clarify this picture by having two dimensions of meaning primary intentions secondary intentions Uh, I won't go into those details. I don't need to What I want to point out is that this sort of, uh, counter examples to use maxim which have attracted a lot of a lot of attention They are directed to Semantic explanation Uh, and, uh, this will be Uh, uh, interesting for later because In the picture that I'm going to propose semantics is not as central As it is there in the sense that those kind of cases can be accounted by model model principles, which are Only contingent contingently associated with semantic, uh, because it's not What is happening here is not prime primarily semantic Uh, which gives it I think a different picture of, uh, model error than the one that is Uh, Greatly discussed in the literature Uh, so here we come to, uh, the conclusion of, uh, of priest, which is that we have a dilemma Uh, either we understand the notion of conciliability as, uh As he understands it, which is something that I can describe In terms of I understand and then use maxim is false Or you need more to conceive than just, uh Uh, a clear understanding Clearing basic understanding, but then The thing you need to add to get to use maxim amounts to I mean there's a risk that what you add is just being possible or being logically possible and If that's what you need in order to conceive something to know that it's already possible then right in the risk of, uh Turning the maxim into a tautology and it basically means that if something is possible then it's possible And something that which is useless because you don't gain anything from moving from conciliability to Possibility if possibility is required for conciliability So perhaps there's some other notion of a conciliability that satisfies you maxim, but There's a know how to articulate it So there's a challenge here. There's a dilemma and a challenge and I take the challenge Uh, I will briefly mention another, uh dilemma which is a bit different and a bit more focused on the constraints Uh, or the tools that or the ingredients that you we might use in order to characterize conciliability So it's a bit more focused on the analysis of conciliability So it's from a paper, uh, by Bertrand, actually it's not forthcoming I'm sorry. I just like, uh I just copy pasted the The beep tech file on field papers and I guess the file was generated while the paper was forthcoming But I think it was published Several years ago probably around 2016 I would say So apologies to Hans and Tom For this paper exists. You can read it. I mean, it's a good paper. You should read it So interesting news maxim So basically they they have, uh, actually it's not really a dilemma. It's more like a tetralema Uh So basically they they consider all the so they made an argument based on the the psychologism and cognitive theory you might have of conciliability It's based on the idea, uh from the literature on mental representations that they are Uh, basically two sorts of formats That are used by, uh, Philosophers of psychology and psychologies when they want to theorize about imagination so we have, uh, link so linguistic Linguistic code, which is the the symbolic code that is familiar form the Uh, so for those language of self hypothesis. So the idea that information is encoded is coded in the mind Uh, symbolically We have the hypothesis that uh information is encoded in the mind pictorially So this is another scientific hypothesis that's been decided by Kostlin Uh in the image in the imagery debate Of course, you could have a mixture of the two or something different or something else altogether different. So we have four options and they We consider each option and uh, it's time. They say that it's bad for our humans maximum So, uh, if it's linguistic only then basically we're So that they claim and here I disagree and explain later why They claim that basically we're in similar situation as uh in the priest argument, which is that, uh We can present linguistically almost anything also including possibilities So, uh, if it's linguistic only then, uh, the most likely would be false This is a very schematic, uh, presentation doesn't do justice to all the details of the argumentation of the authors But that's basically the bottom line. Yes What do they mean by that? Well, everything is uh, no, that's that's a good. No, that's what I mean. I don't think I took that from them. That's what I mean is that, um No, uh Yeah, this sentence doesn't make much sense. Thanks Peter for pointing that out What is meant here is that, um Because I was about to say everything linguistically so everything linguistically representable will be Conceivable, so that's what I meant. That's not very good. That's what I meant So including impossibilities because they are linguistically representable or Representable like encodable by symbolic means in the mind So if you believe that thought if you're fordore and you believe that thought is uh, it's productive You have uh, you have in the language of thought you have a sentence that corresponds to trains Here in gogan and earth now Uh, you have you have the sentence that corresponds to its negation and you have the sentence that corresponds to conjunction of sentence and its negation and so this is something that is encoded a linguistically in the mind and another view Of course, it's impossible. So that's that's the idea. It's a rough idea Uh, so if it's a magicic on it's it's not completely useless But it's a very limited use because it's sort of content that can be represented purely Imagistically are very limited. It's about basically shapes in space To be very quick and you will not get a very far If you want to apply that with your philosophical arguments, for example Maybe you can use you must maximum if you want to simulate in your mind How to move your piano from your leading room to your dining room But nothing much more than that Uh, you have the you have the possibility of a combination of the two and then you have to look at the way The linguistic and the magic representation interact And then there are issues that have to do with the fact that basically either so the magicic presentation not completely constrained the kind of linguists more informative linguistic representation you can Apply on top of it, which means there's a there's a mental arbitrariness and if there's arbitrariness it's easy to Conceive impossibilities in this way Uh For a sake of time. I won't go into the details, but so the example here is you can imagine one medalli or someone who looks like my medalli Doing a boxing fight against One medalli and say I have imagined that one medalli fights again cash is played And this is something that is impossible because they are the same person and they can't They cannot you cannot do a boxing fight with yourself So according to one at least one prominent series of the way linguistic and imagistic Representations interact something can easily get And so this is bad for use maximum and then you can say well some no possibility something Different from those two options, but then it becomes mysterious. You don't know how to make it work So here again, you have a like a tetralema If you will And the challenge to provide the the right explanation of Yes But what does it mean for conceivability to be linguistic or imagistic like what is even the The the the subject of conceivable what what are the things that might be conceivable or the black states or And if they are like wouldn't that I wouldn't you have another notion of conceivable if it's imagistic like applying to different kind of objects Like what are we talking about? So we are talking about so the idea is to have like a Uh, so we all agree that conceivability is a sort of mental act We have psychotic Constability is something that involves some sort of mental representation Okay There are some like psychological theories about how Mental representations work informed by theoretical arguments given by philosophers of mind and informed by empirical research Uh given by a psychologist and So there's kind of agreement that there are two main candidates for the sort of code or medium by means of which the mind and the brain to a certain extent Represent information either it's symbolic which means that there's an element of our awareness and you can have sort of like Syntactic like structure Or it's imagistic And then uh, it has a Uh, it's based on resemblance And it has a part whole structure that is isomorphic to the target of the representation, which is not the case with uh linguistics like representation So those are like options that are in the theoretical options when you want to give a theory of mental representation and if conceivability is a sort of Mental presentation evolves mental representation You you would have to situate it. Yeah, so if I promise what it involves So its ability to conceive and the conception itself might be majestic or just exactly But there's this more motor operator element to it. Yeah, which makes me doubt like It's not because conception might be a very majestic thing And that means that that that's the thing that I'm able to conceive Well, the the is just majestic No, no, no, well I think the No, the the I think the The format of the representation constraints The content to a certain extent. It's not completely orthogonal I mean it's very hard to To present majestically some very abstract contents At least that's the idea behind point number two We can describe by linguistic means we can make conceptual distinctions between things that are Totally identical in terms of how they appear to the eye for example or to the senses something that So you would require linguistic like Code to make those distinctions And you will probably not be able to make them. You should have surely a majestic code With that you might be able to make them if I have a majestic plus linguistic So there's a there's an interaction between the content And the format to a certain extent Okay So This brings me to my problem which is to find an explication of conceivability which fits It is at least True to the intended meaning of humans maximum Which is highly intentional because that's one constraint I gave on formal modeling And So I don't want to put the concept that it's true because I don't I think that's use maxim is like a universal truth and this explains How it can be used and how it can be used at least sometimes profitably In philosophical and scientific practice and So the first step is to Put a filter on the available candidates in order to rule out The priest can the priest starting point which is conceivability just is understanding things crazy In in understandable terms and the um Basically that's one refinement of uh Of priest if I can understand all the logical parts of the part of five then I can consider five This is what automatically takes you to the fact that you can conceive contradictions Uh, if each member of the contradiction is easy to conceive and just involves negation and conjunction Uh, and so and then he says well when something is considered you may not even appear to be possible But the problem is that this appearance of to be possible is what conceivability is supposed to convey And uh in the discussion of among the philosophers who have tried to give an explanation of conceivability and say well if you want to to distinguish the the uses of the the the The verb to conceive or the adjective conceivable which which are relevant to use maxim and though that are not Need to focus on the on those that are Uh accompanied by an appearance of possibility. So this is the point that made by yablo uh in the in his uh seminal paper And I think this is completely right so if uh conceivability does not even uh Provide an appearance of possibility whatever that is Then it's not conceivability in a way that is intended by you and the humans here so Any explanation of conceivability should should be able to at least uh account for that so mere uh Linguistic understanding would not be sufficient and for the same reason mere The mere fact that we have a like a symbolic or linguistic like representation in our minds will not space um, so then yablo provides his own uh Explanation which equals physical conceivability and says peace conceivable for me if I can imagine a world I take to verify p It's not very clear. I mean where I mean it's like this to be a possible world or an impossible world Which is unfortunate because but either way it's problematic because if it's uh, it has to be it has to be a possible world then We are back into a free cinema. We need to know it's I mean Being possible. It's part of our history conceivable. So you might say this is this And if it can be any world then Then it might be false because it means not generally true So this does not really help So what What might help so the thing I propose which is ironically based on some things that uh, france vertu proposed as logic of imagination So the idea is to twist it and to use it in order to construct an ocean of conceivability that under certain condition uh meets uh the meaning and uh The explanatory not the explanatory but the dialectic uses of eons maxi So this is where the formal part arrives and I can see that I don't have much time left. So I will try to Be quick, but my goal here is to convey the main idea If you're curious about the details, maybe we can discuss them in the q and a So the important idea and I think this is a very important insight from uh from france vertu is the idea that uh We can one reason I mean there are many ways to To formally explicate the notion of imagination or conceiving, but one for two ways is to Approach it as a binary thing As something that involves a starting point or an explicit input and an enrichment of that input that gives an output that we showed So for example And the input is what is under voluntary control is something that can be initiated Something that can be started and you output it everything that goes with it It's not that it's not something that goes it by the geological necessity Uh It's sort of sort of enrichment that is that is constrained in some way But not by a purely logical necessitation of like purely logical consequences Uh, I give an example right away. So this is the way to be formalized So the thing to bracket is the input and the the thing afterwards is the output so the On the left is the explicit part which is voluntarily started and the thing that is in the output is I think that follows implicitly for me So here's an example Other friend Axel is not a hipster at all But I can conceive that So, okay, let's let's imagine or let's try to conceive what Axel would be as a hipster Then so what so the input is axel is a hipster And the output is something like it's richer than that. It's The fact that he Has a beard. He has tattoos. He eats bacon Blah blah blah blah and then put all the stereotypes My stereotypes of uh being Of what what a hipster is It's important that the These stereotypes can vary from people to people. So this is a Uh, this is highly contextual But this is basically the basic construction that I'm going to use and this has been proposed by france bartholth It's nothing new there And you can give him semantics for this construction and cons actually has given has given various types of semantics For the purpose of this talk, I will be content with actually the first one he gave which is in terms of impossible worlds So there are two reasons for that. Well, there's one Well, at least one reason for that Which is that we interested in making differences between the possible and impossible. So it's very nice to have In our models the possible worlds and the impossible worlds being separated also because the The kind of tools that we use to explain in which case use maxim works in case doesn't work so well have to do with The relative closeness between impossible and possible worlds. So the kind of Ideas around in mind are best presented in the uh with the impossible worlds framework Uh Well, the only thing you need to know is that there are two kinds of impossible worlds. You have the impossible worlds that behave logically Well, and those that don't behave well at all um so In the worlds that behave well, you can find anything through there But everything that is through there will be closed under logical operations There are also Misbehaved worlds where anything can happen And those worlds have no logic. So I mean in those worlds There are worlds that violate any considerable law of logic and those words are necessary If you want to conceive alternative logic, so if you want to conceive failures of uh loss of logic So that's why they are there. Uh, I will not make use of the wild possible worlds much but uh So basically in a model you have worlds possible among which have possible Uh, the intentional impossible so that are the well gave ones and the extension impossible. There are the non-well behave ones Uh, the art is a The actual world is part of the possible worlds You need to single out in your language a set of uh sentences that Will be able to give you impulse for imagination So the assumption here is that the agents that we are modeling are not able to start Uh, uh, conception with any input So for example, if I ask you uh to conceive what it's like to be a bat Usually you you are lost Cannot do that. So not everything can be started And probably if I have asked you, uh in 1962 I think or 1961, uh, whether Uh, you can conceive uh through the just if I believe not to be knowledge You would probably have some trouble Starting an imagination like that. So I think it's important to to achieve It's important for the story. I'm going to tell that the set of admissible inputs Uh, uh, is in general subset of the of the set of the sentences Um And the the the semantic machinery that will explain the the logic of the enrichment of the enrichment is done by a selection function So this is something that is Uh, machinery that it has been used uh in the In the logic for counterfactuals Uh, here the it's uh It's a section function that outputs Set of possible worlds which are supposed to be um The closest words among those where a certain sentence is true. So the key construct here is f The binary function takes a sentence and world And it gathered a set of worlds worlds possible impossible That are closest to how I conceived the world w2d Among those where phi is true. So you take a sentence axel is a Heapster Has become hipster Consider all the possible worlds Uh, where it's true that axel, uh Uh, is a hipster And you rank them according to how they correspond to my conception of the hipster And and in all the those that rank as the closest that that have that are the The closest to my conception of the hipster, you will see excel with the pair to explain Basically the the way it works and then uh You so what I define is uh is basically the the template of the model and you want to have full fledged models They need to impose conditions and you need to impose rules about how logical vocabulary behaves at possible worlds and at Well-behaved in possible worlds so This will hold. Uh, those are there's nothing like surprising here for conjunction to be true Uh, both outcomes have to be true. There's nothing like, um surprising You have some clauses for a necessity operator, which is here, uh, the global necessity operator Which means that something is possible if it's true at all possible worlds We don't care about that actually it's a relation here because we want the most encompassing a notion of necessity to be possible just to be true at one possible world And here we have the clause for For verifying an act of imagination. So it's true at the world w that Uh, if I start we imagine xgcc that's five i imagine episcity that's i So first has to indicate that five belongs to k and that visible input And then for all the worlds that are In that set here So all the worlds that are closest among the five worlds From the way i present these things All those worlds will verify it Yes That's true, it's a it's a loose talk, but it's can be justified by the further conditions on models that I haven't split that spell at yet so So this is a like gloss to give you the intuition But oh, that's uh, that's a very good observation And actually in order to to express those conditions that justify the fact that it's a characterizes the sort of form of closeness I will use this notation of true sets So sort of brackets With the first means the set of worlds Where five is true and with a minus a set of worlds when it's not true And then you have those conditions here So Fee is true everywhere. So when I when I select The world based on on an input the input is is true in all the worlds. It's conserved basically You have some some rules for the behavior of Conjunction that's supposed to account for the melody of imagination if I imagine a conjunction I have to imagine each the contract separately if I imagine Conjunction separately then I'll say imagine the conjunction. This is perhaps a little more Controversial, but uh, I To I think there are good reasons I mean It's not bad to have it. I say like that and then you have a condition of uniformity That really justifies the idea that it can be seen as a as a form of uh, as imposing an order um Because it's uh, it creates a sort of a notion of equivalence between uh, between inputs at worlds So the inputs are equivalent If uh, when I started one I get the other and I start in the other I get the other one This is something that is a sort of condition that is uh in the Taken from the star knacker A theory of a Possible it's a it's a very standard thing in the in the counterfactual in semantics for counterfactuals Uh And then you define consequence and validity in the usual way Uh, something is a consequence of a set of uh premises if you should take any model and any world In that model where all the premises are true then that for all the construction will be true too and so automatically given these conditions of models you either you get a some validity That are supposed to express Some some principles about conception the fact that when you consider conjunction you can see which conjuncts kind of stuff Um Also, uh, in validity that are provable which show that it's hyper intentional so that we are happy Uh I will not spend too much time on this What is more important is how to move from this binary conception of uh Imagining of conceiving To a notion of conceivability because conceivability is a unary thing. It's uh, you don't ask Whether this connection about Something implicitly conceived or and so So here the idea is to follow actually uh, something a move that has been made in the literature on Need to show the epistemic logic on public announcements It's actually this is tactically is expressed in the same way. So we have a public announcement and Show it out publicly that I can say Uh, then people can make inferences and they can reduce That uh that file But you can also uh quantify over Possible means this has been done in literature on public announcement We can quantify over all possible public announcements and here it is to quantify over all Addisable inputs So something is conceivable if there is an admissible input Which is such that if I voluntarily Start with that input then implicitly I will consume that stuff And that's basically the meaning of that If there is a good side Uh, then taking that as an input to a different side And then you can recover a unary notion of conceivability from that binary stuff Which uh, nevertheless capitalizes on this fine-grained analysis of conception having An explicit part and an implicit part a voluntary part and an inverter part This will be crucial to the kind of application I want to do later Um, so I guess I'm almost done. So I will have to be very quick So I will just make announcements And uh, not give arguments from now on because I don't have time. Sorry about that So the first thing to say is that under this modeling humans maxing fails And Hold on. This is not the end of the world All that means is that uh humans maxing is not a logical truth About conception, but I don't think it has to be a logical truth about conception to be an interesting principle So that's the first thing I want to say Uh, the second thing is that there are some additional principles or constraints One might want to put on the selection function that governs the rules for the Implicit enrichment of the specifics That might actually gets us closer to The intended behavior of humans maxing and here the idea is Condition called the strangeness of the possibility which is the idea that the Very roughly but only very roughly the idea that impossible worlds are always further away than possible worlds where they want to evaluate possibilities Actually, there are various ways to cash out this Principle there's a strong way which is I think too strong too strong and there's a weak way which I think is exactly what we need So basically for a possibility. So if I want to evaluate the possibility I have to look at the closest five worlds And there was always there will always be one that is possible So it's not possible that when I want to evaluate Possibility all the closest worlds are impossible And this seems right because it seems that some impossibility Among the five worlds something impossible would be Closer to something more resemblance something possible, which is something a bit weird and And the the main the main form of the I mean there's nothing deep about it But if you put Strangeness the weak version of the strangest of impossibility Then you have this relation Or not something completely new. I think it's I've seen that in the literature I don't claim originality for this But if you start with an impossible input and you use that principle then you get a possible output so and So this is the basic basic block for me to construct a notion of Consistibility that will work according to humans maxing I and so just to give the The main view because I don't have time to So I think I have a slide where the view is explained I think it's there basically so There's a notion I would I would call correct conceivability Which is which applies to p if there is an input q Such that it's known that q is possible And the output is obtained in accordance with the strangest of impossibility constraint The idea is that correct conceivability is a regulative notion. It's not supposed to be a descriptive notion It's supposed to be what the standard We need to look at when we evaluate the implications of the method of conceivability Basically when I see a conceivability argument I have to ask well is the Is the input really possible or do I have good reason to think it's possible? And when I look at the way it's arranged, it's organized. I said do you have good reason to doubt that it has been Enriched by putting Impossible worlds before all possible worlds And if I check those two boxes, then I'm kind of confident that The argument should work And in the cases where it doesn't work it's because one of those two reasons So it is hyper intentional In case you care about it as much as I do But also, uh, it can explain how thought experiments expand model knowledge and this is something that Is not something that will be clear to you because I haven't explained it, but The idea is that There's a distinction between easy model knowledge and hard model knowledge So unless you're completely model skeptic, I think it's fair to say that there are things that we have At least justified belief About whether they are possible things about the arrangement of objects in space Like easy stuff things you are familiar with It would be it would need to be I think a very hardcore model skeptic to deny that we have knowledge of this kind of stuff Uh, and sometimes this sort of model knowledge is necessary to trigger Uh, to describe an input that would then give me some interesting goods And the best example I have of that is the get your cases Because so that's where I took my example Earlier about 1961 before the publication of the get here the get here a paper People I mean it was very hard for people to uh to conceive of through the stratified belief without knowledge unless they have read Plato Because there's an example I think in the menu In the data story Thank you But for all people who didn't read Plato, it was very hard to consider this kind of stuff but what what get here, uh Did was to provide a scenario the possibility of which is very hard to deny Because it's about people like being in very mundane relation. They have information The one would say that it's I mean it's very Unlikely to be in such situation, but there's nothing nothing like impossible about that but once you uh Accept that as an input for an act of imagination And you enrich it then you come very naturally to the conclusion that there is knowledge without So you start from easy model knowledge and you get hard model knowledge by the method of possibility And the way and and when you do that it's You're not in a situation where you are at risk of putting an impossible world Before all the possible worlds where the scenario Is true So in that way you have like an example of the method that works well And what I've been trying to propose is a model about the best case the case the kind of case where it works And to eliminate the kind of case where it doesn't work So I have a model basically so you can see you can see it as a as a as a policy of good practice for a conceivity argument You can also see it as a model of model error So a model of like what goes how can things go wrong? And I think what is interesting about this picture To the extent that it's uh, it's completely condensing because I haven't argued all about the details Is that there's nothing specifically semantic about it? So it subsumes the the prep key and cases because there are cases where you put any possible world before all the possible worlds But just by accident that there's a semantic story. I don't think the the story about model error needs to be semantic like by necessity or And So I don't have I haven't given all the arguments, but I think that's understood in this way It's not an empty tautology But it's it's not it's not epistemological use of this because you have this kind of thing that is presented And also I think it also meets uh Beto's challenge Uh, I mean there's a there's a sense in which it meets it at hominem because I've articulated the notion that Beto's understands He created it. So I just add a small twist to it so it should not be mysterious and to the extent I mean if then there's a question about how I can make it psychologically plausible then I would have to To go into one of the words on a dilemma, but I have I have argued I mean there they are places in in Beto and Thompson's argument where I think there's a place to disagree or to to avoid some of their conclusions I'm happy to discuss that in the q&a if you want, uh, and I will stop here Thank you very much Um, I wish you want to take five minutes of both or if I just I don't need five minutes Both Impression Thanks to all this So It's maybe not essential but in the beginning when we talked about the shooting this gets um Um So it seemed to me that the conceivability there Um Right to right to be in any way Makes sense but leads to some sort of possibility Um, not just that it's conceivable. It seems that it's It's conceivable as Experiment that can be set up or something. It's not just a mental Uh state that we can imagine It's like we can actually imagine to realize it which used to be going a bit further, right? Um, like we have it's not just that the ideas are up here Is that we always also have a at least Idea of how to set up the experiment and it's procedural or something and that seems to be very important to get to the right kind of conclusions Um, it cannot be just mental gymnastics To go to the possibility conclusion It's conceivability plus something else but conceivable to realize or something. Okay. Yeah Yeah, that's uh, that's an interesting point uh I don't think I have ready made answer So just to be clear. So the idea here because conceivability is not there. So the possibility we use to uh To justify premise two Um So if I understand your question a bit more like a I know it's not an objection but as a challenge or something the idea would be that Uh conceivability If not, it's not stringent enough To give me the right kind of possibility What is why does not seem to maxim but some some uh Well, I guess it depends on how you you how specific the content is If the content is uh, if the idea of like uh executability as an experiment is part of the content The target of what you need to conceive Uh, then, uh I don't see why conceivability would not be a good guide for that Um Because even if you think so When you say it's not just mental gymnastics, I guess the implication or the caricature there is that mental gymnastics would over generate Yeah, but that would be like the usual sort of arguments against means maximum right that we can Sure, but uh, but uh My response to that is that uh The the right way to understand it is to understand it as constrained in a certain way And I gave the uh The basic principle governing the constraint which is the principle of stringence of impossibility Uh So it's easy to see when the clear contra example and the clear good examples Uh retrospectively speak that the constraint is respected or violated Yeah, this objection is how do you apply it in the first place when you have to Discuss a case that isn't clear And here I think the answer is not better neither better nor worse than when you use the uh relative uh Comparative possibility or comparative closeness as a heuristic to evaluate counterfactuals It's not it's not the perfect heuristics because there's an element of weakness So if there's nothing I think considerably different uh here than in the case of counterfactual, but it's not completely useless in the sense that If there's a clear violation you will be able to detect it and uh So I don't think it's completely uh purely post hoc rational reconstruction Slightly I think it answers a slightly different question on the one we started with but you seem happy, so I'm happy I'm the wrong question Um, thank you very much for the talk. Could you explain a little bit, uh There's called the range of the city of spanish class. Oh, yeah, because after that I thought I understood after that you explained it's highly contextual Blah blah blah and now I'm confused Oh, you're right. I didn't actually I didn't say anything about city of spanish values I don't think I even pronounce this even though it's written so Thanks for the media opportunity to correct that uh that mistake Um, so the city of spanish boost case expresses, um Um, so all things equal equal, so The all things equal is, uh, my beliefs assumption in a certain context So the selection function Is something that so this is a mechanism that governs what will be enriched and how Uh, this is supposed to be specific to a certain context where I uh, and so In I mean here i'm giving like physical explanation explanation of the framework. I don't know if Franz Bertot will agree with everything I say, but in the way I picture things, uh, there are some aspects of this which are like, uh Internal there are like beliefs There are things that are more or less high high higher by the way I see the world And there may be elements that are based on like, uh assumptions I make, uh in the In a certain situation in a certain conversation because I want to interact with people or So those beliefs plus those, uh Those context contextual assumptions To the extent that they are kept fixed There will be regularity in the way things are implicitly enriched So that's the sense in which it is a terrorist virus If I change my mind about what hipsters are if I get a like a different view, uh, no, it's this is so 2010 9 2020 hipsters are completely different Uh, they're they happen. No, they shave themselves Uh, they have removed their tattoos and uh, they eat meat I don't know So if I make that that update in my conception of hipsters, I will Change the way I conceive my friend Axel when he becomes hipster So that all things equal it's the context which is More objective than just your belief your presumed belief and together Yep Strange use of I think it it's comes from It's not it's not the same It's not continuous Or at least I don't see clearly the continuity with the use of satirist paribus and discussion of scientific laws. That's what you are after But this use of satirist paribus was used by I think priests And Berto about conditionals for the talk about satirist paribus conditionals Because there's and it's for conditionals that are non-monotonic So in the sense that everything's equal That will follow from that But if you add further information or change the context, this might change. So that's that's a basic idea No questions So I have to question myself, uh, so it's a third to process. Well, I'm not sure if it's real or even so one question would be So this kind of idea that This reaction to the OCD and OCD is Contestable as you say. So can we do it and go away? in in case of Scientific communities testing who belongs to the part of the community by saying well, yeah Who will recognize the first experiment as the best way to use it? So if you are like in the Korean scheme, you could define the Different mathematics as the people who recognize the set of the first experiment as being extremely useful Is Yeah, so I thanks a lot for uh for bringing this to the table because Uh When I was super ambitious about these ideas and I wanted to make like real Applications because basically the the thought experiments. They are just like a source of examples, but they're Ideally, I would like this framework to be able to say stuff about Issues in the interpretation of thought experiments So one thing that I'm interested in I still don't know exactly how to how to apply this framework to this kind of issue But those those are the issues of counter thought experiments So there's one thought experiment and people see how the nutrition about what would happen there And then someone else uh arrives and say no no in that situation. I think a completely different should happen So uh historical example of this are the newton bouquet and the interpretation by uton and the interpretation by mar I suppose the the case of uh, um, so This is the first of physics would correct me if I say something wrong But in my understanding of the use of Schrodinger cathode experiment so at first it was used as a as an antagonistic thought experiment against the Copenhagen interpretation But then it became like a tool to To distinguish very solution to the measurement problem And so and of course depending on what your assumptions are about the foundations of quantum mechanics What what the way you enrich the situation described by Schrodinger in the first place will go might go in different directions Uh, so I mean This is something you can express with within that framework. I mean, it's cheap in the sense that Framework is context dependent. So there's context dependency. So I don't know if you gain a lot of insight into the specific Dynamics of Schrodinger cathode experiments, but at least you have a framework to discuss The sort of phenomena where people agree on the on the input But disagree on the output Uh, something you can Represent I don't know if you can explain it you can provide a good explanation of but at least you can represent it You can use that framework to describe it formally, which I think is a good thing for the framework Surprisingly not Surprisingly not but perhaps I should Yeah, I mean this is um, no, I mean This research is really about framework for reasoning generally about The validity of humans maximum when it works when it doesn't work And usually when I do experimental philosophy, it's really to uh, to do the anatomy of a specific intuition So it's very specific And uh, I haven't used so far Actually in Denmark, we did that with an assumption there. We used also experimental philosophy to discuss the general model of of thought experiments Uh But that was more like sounds. I mean, I of course I'm a class. That's why I was in the I was on board But I think it was some's approach to uh, to try to tackle this kind of general issues with thought experiments Whereas I'm perhaps a bit more Uh I would say not conservative, but um in The most recent work I've done in experimental philosophy was about like really Uh dissecting sort of intuition to see what are the factors that write it Whether it's reliable How far generalize is this kind of stuff? Thanks Uh, for my annual question, if you see one that says questions Oh, there's a big other that just pushes it So It's it's a very interesting that it's Like James Maxim would not be tautology, but something with content, I guess in a bit Hispanic sense The content about what? Uh About the world about the physical nature of the world about our ability to conceive Um Yeah, so that's a very good question and I think the answer I want to give Uh, it's it's about what is required For considerably to be a guide to possibly So it's it's a total it's so this answer is a bit total logical But it's not completely tautological because it means that This connection is not for free Okay, it's not a tautology. So it's not something you get for free Uh, and I guess this accounts for part of the The feeling of people who say well conceivability, I mean Who are you to deny that what I I'm not able to conceive what I can understand theory based on just like the I understanding of of the english language or french language by reading french How dare you say that I'm confused or that I don't really consider it And of course the idea is that well, uh, it's about a certain way conceivability is used to track Possibility and this idea is also uh Very uh, very much accepted by people in the philosophy of imagination Who distinguish creative uses of imagination from cognitive use of imagination where the idea that we constrain imagination in a certain way So the content of this you maxi in the way I really like the I never thought about this consequence of the claim that it's My claim of being happy that it's not a logical truth Uh, so it has a content and that content is uh So you can see it as a difference between mere representation based on understanding the sentence and uh moderately guiding Presentation And what does the work is? And possible so there are those two things that are responsible for for to work properly So I haven't said much about Strangest of impossibility. So it's something that came out as a As a constraint on counter possible reasoning But it's a controversial principle in Well, it's controversial. I mean a lot of people I mean the citation I gave by no land works. It's intuitive immediately says that well, they are counter example. So the View which I think is the received view, but I'm not sure actually I think it's a the received view is that it's uh, it's a reasonable principle, but it's not True all across the board At least there are some there are some examples So they are pressure pressure for me to uh to say something about The fight that is not the universal principle because a very straight story which say well, this is a this is a This is a stable. I mean Well-founded principles for the evaluation of counterfactuals in the presence of impossibilities In presence of impossible worlds So in order to track Possibility when we conceive In the possible presence of the representation of impossible states of affair This is the answer to the problem The story has to be a bit more complicated because of that and so As you can see it's a potential the qna to Feel some part of my talk But then I feel pressure to say things about the counter examples And to explain why there are exceptions I think the exceptions become the occurring in some very specific contexts That are reduction like contexts So there are contexts where you Assume something impossible So it's not a surprise that impossible worlds will come before possible worlds even though the The antecedent is itself possible. So that's the story I want to take so So I don't want to deny the fact that Those things are assortable And I don't want to cook up a story to say no no to be like revolutionary about I can accept the idea that in some contexts But I would say that those contexts are deviant with respect to the purpose to the to the purpose of a certain Possibilities as opposed to ascertaining impossibilities Which is what we do with the reduction reasoning This is probably the most controversial Part of the of the argument something which for which I expect most resistance from the reviewers But we'll see I mean that's what I I need to I need to to to find the right balance and and the right arguments the right considerations to To To convince people who are skeptical about In this the strengths and the weak strengths of impossibility constraint for those reasons That's not stupid reason. Those are Important reasons You know So One of which should be There are some impossible situations that we did It's impossible for people that we I would need to you to remind me something about the What do you mean by possible here? If you take your word that you freeze the part of it you would have The other part changing and the The part that was frozen if they deflose they don't see the change happening or something But that's curious reason that is not very likely possible Yeah, so it's not physically possible. Yeah, okay No, it's completely within the preview of the framework Because the notion the model space is a what is Well, let me put it that way the framework is supposed to Is designed such a way that those cases should be representable in it There's also there's however difficulty about the application to a third experiments in physics and I would explain why is because in physics Uh What matters is Probably a restricted notion of possibility which is Then problem that then you have like a lot of theoretical assumptions that need to bring in and I don't see Those have to be to be I mean, I don't see like Decisive reasons to go One way or the other when you design the framework. So one big question is whether you accept a metaphysical Possibly But physically impossible scenarios Whether if you accept that your your model space will have a certain shape And if you don't accept that you have a different shape and the analysis of these kind of examples will depend on that So for example, if you want to discuss the possibility for laws of nature to change Or the possibilities of the laws of nature to have been otherwise You will have to make decisions about the The fine-tuning of the framework whether how you conceive the relationship between metaphysical modality how you express it Because what we have so far? It's just like a set of wells And the the constraints that are put on those worlds are mostly logical So it says little about how those words Metaphysically if they could so could you have a metaphysical difference between worlds that are Allegeically the same So those are things that need to be to be Added or to be sought through when you want to apply the the framework for certain case Yeah, it's something that that The framework allows you to to to describe this kind of stuff, but doesn't force you to do it one way or the other and and The results you will give in the analysis of specifics of experiments will depend a lot on the choices you make So we should use accessibility relation to model physical necessity physical possibility What properties do you give to this? And the framework does you well, you can do it this way you can do it that way Then you have to find the good arguments to be Probably as general as possible to be as unprofessional as possible if you want to design framework But you wanted to represent something that everyone will accept So So that's so in that sense the the framework doesn't say anything very specific about about this kind of stuff hopefully provides a framework to to say more One thing I was interested in and I I don't know enough the literature about idealization to have something like to know if my intuition about this is like Sensible, but it seems to me that some some thought experiments that rely on Idealizations when you remove certain type of forces or you remove some sort of physical effects They take you to physically possible worlds and so Then If you if the whole reasoning remains or a comment or at least include this kind of like Physically possible also how can you use that to say something about physics? So yeah, so I guess I don't know if this particular framework will be useful for this kind of because I think this is a issue of like the context sensitivity of Representation for it does not Matter the most here unless you believe that idealization is Context experiment and maybe yeah, I think it's mostly a question for people who analyze some experiments by means of counterfactuals But in the case the problems are similar Because you want to Question is which part of logical space or model space Are you in and which part you need to be? Where you start where you get at the end and where you need to be to have a result that matters For the theories you want to discuss. So, um, so the issue of idealization Is something that I still need to get a good grip on I think I was thinking about it in the other way around which is to I don't think I mean I think many start experiments involve idealizations in the sense of a sense of minimal idealization, which is to to bring in as few Uh material as you need to make your argument But because you leave things out like frictions, for example, or You're in a not, I mean, it's not really fictional in the sense that it's uh, it's not in the real world But it's also fictional in the sense that it's not It's not In if you think that if you think of loads of physics as loads of nature Outside, uh, you you remove some some parts of the governance of the nature Of course, if you you mean that's a different story But it's not just fictional in the sense of uh out of this world. It's just out of physically possible worlds Yeah That could be interesting in your framework the difference with an approximation Okay Is idealization or like exactly like you said, we are impossible Well, physically that you used to represent the world, but some Some Scientists say no, no, I don't do that. I need an approximation. So it's incorrect description of the actual But most of the time It looks exactly the same It's really the interpretation. I know that some philosophers say that There's special case where idealization approximation would be distinguished But from your point of view, it's very different because you will build The possible world in a different way and the argument for different way I suppose yeah Oh, I mean, yeah, there are many options, uh But how to cash this out. So thanks. I mean, I was thinking about Things about it, but it's too messy in my mind to communicate them because if you do maybe after a few days later No questions So I was thinking about how central It's the idea of closeness was and similarity and so on. Yeah, I think that it doesn't figure in the models themselves It's like Um, so well And I have a thought But it's based on memory. So it could be a major remembrance And when I see the shape of the models, I'm starting to doubt about it But there's a so there's a sort of like There's a formal fact Was about the mathematical faculty. So it's not very trusting mathematics. So it's not but There's a question whether you can define from the selection function a partial older Or or a relation of comparative possibilities So give me a give me a world Give me a sentence Which is admissible Can I describe or can I define? An older relation The partial older relation On the rest on the world on the set of words And if formally this can be done. So if I can define the relation that has the like the reply of t reflexive Symmetric and transitive Then there's a clear answer to your question and I think I remember working on that But I don't remember if I and I got this result, but I don't remember if I got it from this Sorry From this definition of model or a slightly more A slightly stronger one And now it seems a bit a bit Maybe too weak to get to that Rather weak? Yeah, exactly. You really want your impossible worlds to be All further away than I mean, this is a different question I think this is a different question. So there is a question whether the The selection function imposes an order on worlds given a world and And the statement So whether this ordering is defined So that so this is what is required, you know, which would say like Literally and minutes that F of phi and w Gathers the closest worlds To my conception So in order to be able to say that I have to be able to show that this This relation this order relation is definable And this is the first step And then comes the second question, which is how does like the model status of the worlds Interacts With this relation. So is it the case? So is there is the case that For all let's say possible Impots Is it the case that The closest worlds are always just possible And if that's the case, you satisfy the strong version of Transient to the possibility I'm pretty sure that this doesn't hold just by so that's why you need to Discuss additional constraints like transient to the possibilities. So here I mean You will not get Things like that just from the sort of stuff because it's not sensitive to to model status That's so you will not get it, you know, I don't see how you could could emerge I mean the but the first question I think is an important question and I will have to solve it When I go home because it's uh, it's whether this is Enough to give an order in Because is that simple to your story? I think it's central. I mean, I think it should be because I really would like to have to be able to take this closeness Uh Think seriously Because otherwise I mean if there's no relation of closeness then even the notion of Transient to the possibility is uh Is a It's not what found it. It's in my imagination And if you want to talk about words being closer than others you need to have a to have a real relation of Comparative possibility out there So if not if I'm not able to define it Then it's just we'll talk and uh, no, it's not good But you could just use Another framework. No, I mean it's like just the the impossible world. Yeah, no, sure. No then so it's not the end of the world It's just that this particular Modeling needs to be a bit enriched um I think in a case that it should be enriched for the logical part If I want to have an interesting logic of conceivability That is definitely it is too weak. So for example, this is too weak in the sense that The conceivability of P or q will not be equivalent to the possibility of qrp With because there's no nothing that governs the behavior of disjunction here So if you want, uh, this commit activity of Of disjunction under devotee of nature and other like principles like the Morgan principles kind of stuff you need to To put them by hand in the definition of the models in a very Uh, so in the case of disjunction you have to be very careful because you don't want to To have additivity for example showing up If you care about not having additivity have to be careful So I have an idea about how to do that. I haven't like uh checked all the details, but I'm Uh, I I think there are ways to do it. So at least I'm working on it in order to Be able to deliver a logic of conceivability that is well behaved and well behaved at the end because if the idea is to formalize the notion of letting to to have a formal approach, uh, and then end up with the logic where conceivability cool is hyper intentional, but well Disjunction doesn't does not commute I mean what are the I mean I would say that the the benefits are formalization are not very Very good because the it's clean adequate. So, uh, Yeah like This has to do probably with the fact that your artistic impossible worlds are completely unrestrained Yeah, you have to just dare to enable the weirdest thoughts, right? Or You open to that case, but if you like make them Help me girls or something like that you'll get those free, right? Yes, but I will get I will get addition and I don't want addition. So, uh Yeah No, but I think so There's a question about why To have all these impossible worlds and you can shut me up when it's time because I can talk about this for hours But um, I think the only motivation for having these impossible worlds, uh, as far as I can see Is to deal with counter logicals Because those are the things that would give you, uh, worlds where any logical or fails And if you want to take up the challenge of priests, which is a I can conceive There are logics. I understand. I know that they are like, uh Not the right logic, but I can understand I can reason about them So I think that that's a good reason to include them in a logic of conceiving Then you should also be open to it not being I can conceive of a logic and I know Well, okay So Here we have to be very careful now. This is a very good point, but I uh, I thought hard about that And so there are different viewpoints are viewpoint of the of the of the of the ascriber of the so the logician regulates the way we talk about Uh acts of conceiving Uh, and if so if the modeling is based from the standard of classical logic, for example Um, the question is whether, uh, conceiving that a or p Um Um Will uh, it's the same thing as conceiving that that p or q and it's something, uh, different of considering that It's always the case that for all p and q See what I mean? So there's a So in in those cases of, uh, uh Alternative logics, uh The weirdness comes from the fact that we say something, I mean, uh about, uh So you can see counter examples Um, but they would not have the form just a or b. So what we'll have is a or b but not so Conjunction So this would hold everyone and the question is why Are you happy not to have that? So, uh, basically if you question both then, uh, it means that there's no Uh And uh, what we need to be open is to be, uh, to the fact that if I can see a and b plus something else which is that Simplication of construction doesn't hold I might not get one of the contracts at the end This should have a model But it's something much Those acts of conceiving have much more structure than Just in here Um So I think the way we represent those like, uh, weird logical contents that, uh, the way we express them within that basic vocabulary Which is very the vocabulary of propositional vocabulary It's this vocabulary is very weak to express the sort of ideas we manipulate when we discuss actual logistics Usually we do not care about whether the conjunction is true. We care whether the conjunction follows From the context of taking separately. It's not something we can express just with the word alien language So that's why I think those things are Too harmful If the goal is to include actual logistics, that would be my my retreat and so I would like to impose special conditions to have this for this reason but I'm open, I mean, uh I think the the end of perhaps you will convince me that Even this has to go but you have to convince me you need to give me some some complete examples See how it first describe them in that very limited language and And if I need to remove the principle that covers that Then I think from the end of the framework, that's why it's too weak to be interesting So that's why the direction I'm going is to One last question Friends, I was very fit to even give you an example of what was wrong That's irrelevant How much is the idea of underrepresentation necessary for your idea of principle and conceptivity? And then I actually like the rest of you know about activism and the idea that you have Perception or consistent without any good representation. Will it still hold this kind of idea for today? That's a very good question. Uh, so, um So yes, I think, uh I think we we we sit in the last seminar a few years ago Eric nine Who's one of the co-author with Dan Uto of this kind of uh of these kind of ideas Uh, so I'm sympathetic with uh an activism and it's kind of stuff on a number of counts Especially when they want to explain, uh adaptive behavior Which needs to uh where we need to coordinate our our perception and action for this is very good and uh in the points they make against uh The computational theory of mind have very well taken The problem I have is with the claims, uh, the generalization to uh We kind of count for all cognition without representation And of course, uh, the obvious examples you have I mean, but then it's a when I discuss with these people with these people and with these colleagues Very often there is a there is a Um We have I can I mean one of the reasons I got interested in in quality science was to account For the hard stuff like linguistic understanding Uh, like Chest calculation, for example, how do you account for that? This is something that people do. How do they do it? What are the strategies to prove complex stuff in math? And for this sort of thing How do we understand the jurisdiction for this sort of thing when you when you deal with uh Decoupled representation I think it's hard to I really don't know how to do without representation. So But um But I guess the the the inactive approach would be very useful For discussing for example the insight you could get about physical systems By playing with physical toy models For this I guess they would have interesting stuff to say and to the extent that those It's kind of manipulation of physical toy models could be seen as limit cases of thought experiments For this kind of thing. I think this approach has a interesting stuff to do But it does not really communicate with what I offer and I guess my approach or the kind of approach I'm after Is mostly uh interesting When these kind of stuff have nothing else to say Uh, I don't know what the inactive story about about Showing showing earth cat is I know they will have stuff to say about the Perceptual motor Perceptual motor grounding of physical concepts of even of mathematical concept, but it will be super general And they will say that the notion of the political space space of basic intuition to have about transformation like the deformation of objects for enough but when when it comes to uh, like This distinction between Explanatory proofs versus non-explanatory proofs or The details of every abstract reasoning. I don't think They they can do a lot without my whole presentation and uh, so that's why I don't feel like It's bad for me to your presentation is in this discussion That's really interesting here without further ado And for the case for this kind of stuff can be applied for within activist perception The the Possibilities and the possibility will be almost the same concept there for for 90 years Because the perception is already an action Think for our truth, we already have simple possibilities of actions So that would be less to get between possibility and possibilities. Maybe will you maximum be less of a concern? Uh That no that that no, that's a very good. That's a very good question. Uh, I would say that uh, this The the the part of model space that you will reach with this kind of consideration are Very close So can I sit on that chair? Can I jump on that on that table? Uh, so I guess they probably have a good story about what I called easy possibilities Which is something I take for granted in the account So probably they Here they have they have they have interesting stuff to say Uh, there there's a bit of a literature on the uh, on the notion of affordance From this point of model epistemology. So the sense in which we perceive Versus infer possibilities and uh So there's a book by a colleague called Guinard du Clac who I think is belgium is just based on that It's a bit, I mean it's from has several years old, but uh So there are people working in model epistemology in that area But this is not the sort of phenomena I'm primarily interested in with With this discussion This question can do everything at once He's very welcome