 in the seminar on Metaphysical Science. Alexander is... No, no, no, Steph is a seminar. Yeah, Steph is a seminar on the Metaphysical Science, the last meeting. Okay, my pleasure to... Alexander doesn't need a proper introduction because you already know it, but pleasure for me to be a seminar against this last talk. So Alexander, can you justify the interaction principle if loss of nature are changed? Metaphysical answer, for you. Okay. So since I'm replacing someone, I've put that together recently, and yeah, it's still a work in progress, and I've been sick all week, so I'm a little bit lightheaded. So if you are confused by a sentence or I'm confusing, please stop me and ask for a specific question. So the title is, Can we justify the induction principle if loss of nature are changing? What I will mostly answer is not a justification, but in what way the principle of induction could still be compatible or defendable if loss of nature are changing? So it's not as strong as justification, it's more a compatibility discussion. Okay, so the plan will go like that. So first I will discuss a little bit how, why it's a metaphysical perspective, there's many ways to do that, but I will just give an example of how Newton metaphysically justified inductive reasoning based on a metaphysical conservation of the nature of the world, just to give you an example, a historical example. After that I will go in the three basic metaphysical theory of loss to show how day to day they try to make the kind of claim that Newton's was doing more clear, more solid, so a good conception of, a good metaphysical conception of loss should be compatible with the principle of induction, it's like in the Cayet Charles that what you should do. Here after that I will discuss how this theory could be altered to have changing laws of nature, not in the very strong metaphysical sense, but in a way that makes sense of what the scientists, when the scientists says laws are changing. I'm afraid here I present this result, all the results of the section three two years ago in the paper published, so I will go fast, except if there's a lot of people in the room that were not here two years ago. Okay, and then I will show that most two out of my theory of metaphysical laws are incompatible with the induction principle, except one. So one survives, but I will not tell you before the end. You will guess, drink the roses. And I was surprised about that. I thought the tree of them would die. Okay, so just to, to flex our philosophical muscles, how do an example of how a famous scientist, famous natural philosopher, justifies production reasoning based on a certain conception of nature. So in the third book of the pre-kitya, it's the only part where Newton is discussing a little bit of mythology. So how to do inductive reasoning, how to use experiments, blah blah blah. It's two pages, it's not really long, but it's still. And the rule one, the first rule of these rules of reasoning, no more causes of natural things should be admitted than are both true and sufficient to explain their phenomena. So it is a rule of inductive reasoning. When you find an explanation to a phenomena, you should not multiply causes. Why? And here is the part about conception of nature, the conception of nature. As the philosophers say, nature do nothing in vain and more causes are in vain when fewer suffice, suffice, suffice, suffice. For nature is simple and does not indulge in the luxury of superfluous causes. So it says you're justified to do that kind of inductive reasoning about the cause behind the phenomena because nature is in a certain way. Nature is simple, nature is nice. Nature does not do excess. Another famous case, and this one is even more known. In the rule three, those qualities of body that cannot be addicted and limited, blah blah blah, and that belong to all bodies and on which experiments can be made should be taken as qualities of all bodies, universally. So what is saying that in this rule is that the qualities, the properties you identify in experiments could be extrapolated to very far away, very big and very small on all scale. And why are you justified to do that? Why could you, are you justified to extrapolate, to end up, to make an induction based on what you measure in the laboratory? Because ideal fantasies are not to be fabricated recklessly against the evidence or experiment. Okay, that's epistemological. Nor should we depart from the analogy of nature since nature is always simple and never a consonant bit itself. The analogy of nature is an expression that was really studied by historian of science and philosopher of science. There's a very famous paper of McGuire about that, that died two years ago. One year ago. And so it's because Newton believed that there's effectively an analogy of nature. So the properties here are not exactly the properties at small scale, but they are analogous. The properties here are also analogous to the property over there. So you are justified to do that kind of induction. You could be wrong, but you are mostly justified because nature is built in a certain way. So what you notice here is that it's not that clear. It's difficult to really point what kind of property, what kind of nature, justify what kind of inductive reasoning. After that, all that was put under the rug and called the principle of uniformity. So in Mill and all these guys it's all inductive reasoning. It's justified because of principle of uniformity. Whatever that means. Nature is the way it is and it's why we're justified to do science inductive. Okay, so now I will present three classical. It's always the three same that I present in every of my talk in the last two years. So I'm very sorry about people that already know that part. But I will show explicitly how these three theory of metaphysics of laws try to have a better way to say exactly the same thing as Newton. They will put things explicitly in their theory to guarantee or at least to mitigate the skepticism about inductive reasoning. So when a metaphysician built a theory of metaphysical theory of laws, of course they want to justify counterfactual reasoning, they want to explain science, they want to do, but also they want to be compatible with the fact that inductive reasoning function in the world. So they would like to have something like a metaphysical explanation of inductive reasoning. So let's look at the regularism at Al-Aliwis. So I know that most of you are familiar with this theory, but yes, so I can go fast, slow. Okay, Julien does not. So I will, no, but that's okay, that's okay. It's why there's slides. So it's just, I don't want to bore people, but if you don't know, I will explain. So what is a law according to the Al-Aliwis version? So a contingent generalization is a law of nature, okay? It is a law of nature, it's a equal, it's a very strong claim. If and only if it appears as a theorem or axiom in each of the true deductive system that achieved the best combination of simplicity and strength. So the best, the best system to describe the world, the best deductive system to describe the mosaic of facts, so all the facts, past, present, future. The best system according to simplicity and strength, all the axiom of the system in the theorems are laws of nature. Of course, it's not enough just to say that to well define the system of law of nature. So to be able to justify counterfactuals and all the things we want natural laws to do, Lewis had to add at least, and it depends on the list, it depends on the people, the philosophers. We'll satisfy other metaphysical hypothesis and we'll look at them because you will see how most of them favor inductive reasoning. So first, Lewis presumed that there's a mosaic of facts. So you can, there's a way to represent all the facts, past, present, future, and it's called the mosaic of facts. I will not discuss the details of what our facts move for him is mostly instantiated natural properties, so if we can discuss about the notion of facts, there's maybe we should not discuss that in the list. But there's a mosaic of facts, and these regularities supervise on the mosaic of facts. First, metaphysical hypothesis. Fundamentalism, perfectly natural property. Properties form a unique set of properties whose instantiation constitute the bedrock ontology of the human world, the mosaic of facts. For such a bedrock, a modern derivative existence obtained through supervision needs. So we have a number, a finite number of natural, kind of natural properties, and they are the thing instantiated in the mosaic of facts because it's finite, because everything derived from them, if we know them, if it's possible to know them, you're able to do science. Physicalism, perfectly natural properties are physical and should differ from the identifier and identifiable as such by physics. So it's an explicit hypothesis that the properties of the world are knowable. So that's good for the principle of induction. In fact, the property of the world are knowable. Why? Because they are. Physics is good, physics is the way. Why physics, not biology? Because of the moment he was writing, I'm quite sure, and the people in you. Three. Meteorological reductionism. Non-perfectly natural properties, for example, biological one, for example, reduce to perfectly natural properties, such a reductionism immediately follows from a strong interpretation of supervenience to the effect that emergent properties are if so factored to be excluded. We exclude, a priori, synchronic emergence. Because that would put in danger the inductive principle. If suddenly the mosaic, if you appear, problems, does not exist. Yes? That seems more like a counter-exampling is di-prolic emergence, is that synchronic? Yeah, but here the meteorological reductionism is really a synchronic. It's to say that every properties, you know, higher-level biological properties are just supervene on physical properties. So the appearance you're talking about is synchronic? Yeah, in a certain sense, but if you don't have that, if I see perfectly natural properties now, in future, the perfectly natural property will not have a non-supervenient new properties over them. So I'm justified to do science now. I have, it's not perfect, because maybe I didn't see anything, some part of the mosaic, but at least the properties we're measuring now are the one that will exist in the future, necessarily, according to that. Internalism, past, present, and future facts are ontologically on par to the effect that there is no genuine becoming in typical Umean worlds. This usually translates as a black universe, but not always, not always, depends of the kind of Lewis you're defending, but it guaranteed that the future facts exist in the same way that the past facts and the actual facts, so the mosaic, the best system is really describing all the mosaic, even right now, at least in principle, even if us, we are in the mosaic at a certain position, right? We definitely, future facts have a status, past facts have a similar status. And finally, universalism, the best system, there's one of them. It's the same everywhere, everywhere when there's one. And the regularities that are identified in the best system, spend on all the mosaic. Now, oh, there's one just, at the end of the mosaic, there's the regularities that suddenly appear at the end of the mosaic, it's impossible to predict from anywhere in the mosaic that it's excluded. And you see how these five principles guarantee inductive reasoning. They don't make it necessarily efficient, but at least they are compatible with some inductive principle. Because they say that there's no privilege position in the mosaic, the properties that are measurable, properties that we know are the properties that will still be the properties in the future, all kind of nice stuff to protect the principle of induction. To the account for intellectuals? Yeah, according to Lewis, with that, you can justify all the intellectuals reason. Even though there are no... Even though there's no real modality, nowhere in the system. How does that have an impact on all our possible worlds? Some metaphysician would say, it's impossible to justify intellectual in the Lewis tradition. You know, how did he go when our past collaborator was defending? Lewis built that explicitly to be able to justify counterfactual. How? Because he says this best system and this general structure of the mosaic allow you to justify a lot of counterfactual conditions. If I had did that, I would have get blah, blah, blah. Why? Because it fits with the best system. It's a deduction of the best system, for example. So it's a very epistemological way to see the thing, but that's on purpose in this. It was built like that. After that, you can ask yourself, how the same person that was realist about possible world decided to exclude modality completely of his real metaphysics? This is something we should ask to specialists of Lewis, because it's the same guy that did exactly that. It's explicit, there's no model, explicit model properties in the world. In the actual world. The actual world is a mosaic of facts where every facts is unpowered. But these principles are maybe necessary? True, yeah. These principles, according to Lewis, are necessary to be able to justify counterfactuals that have all the nice properties, model properties you want to have. Model, not model properties. So while the laws themselves are general, what are these principles you can get? Model content. Yeah. Yes. So based on the best system, I could say, if in the future that I don't have access, I would have blah, blah, blah. I would obtain blah, blah, blah. Why? Because the best system exists and justifies this thing. Of course, my knowledge today of the best system is partial because we will add access to the best system only if we had all the same, of course. But there's all kind of nice properties that guarantee that we are in a way to get to the best system. Okay. Now, Armstrong conception of metaphysics of laws. Same questions. So the idea of Armstrong is really to find a way for the intuition that the law governs their instance. For a Lewis, the laws are not governing their instance. The laws aren't just there to describe instances. Armstrong wants to get to something strong, the laws govern their instance, and he wants to have the idea that there could be other world with the same natural properties but not the same laws. So he wants to have both, this both intuition. So he defines laws that way. Sentences like, it is a law that Fs are Gs should be understood that it is physically necessary that Fs are Gs. So he will define physically necessary for some special kind of insistation. That will be represented like that. There's an insistation relation that if F and G should be understood. Something being F necessitates that the same something being G in virtue of the universal F and Gs. So it's in the virtue of the nature of properties that they are governed in a specific way. Nature of the properties in our world, of course. Because in another world the same properties could maybe do something else. And that's it. Since relation among the universals are timeless, necessary connection, this should protect me, protect his position against skepticism of the principle of induction. Because the properties described by the universals are always related in a certain way necessary that produce the same kind of instance regularly. So causal relation in the world or in fact just the effect of these connection between these universals. If there's enough of them, not very clear how much, but if there's enough of them, the world is regular. And that's it. It's enough to protect inductive wisdom, at least to be compatible with inductive wisdom. The world's completely regular or it's sufficiently regular? We can discuss in Armstrong, it's sufficiently regular. That may be completely regular. Armstrong does not want to exclude some discovery of science that the world is, this part of the world is not regular. That's possible. But it's sufficiently regular because there's sufficiently necessitation relation to guarantee that we can do science. And of course there's a privilege of the universal describing natural properties, of course. That I forgot to put in the slide, but there could be all kind of universal related but I kind of necessities. But the one describing natural properties have this very nice, okay? Disposition, essentialism. So if properties have dispositional essences, does that, that, that bird, then certain relation of necessity will hold between the relevant universals. These relationships can be identified with the laws of nature. So it is necessary that if you have a certain disposition that if this stimulus blah, blah, blah, or occur automatically this manifestation on this disposition on the object X, if and only if, it guarantee a certain conditional counterfactuals. So it seems we see laws, regularities, but in fact these regularities are a by-product of the action of disposition and the world. And these dispositions are the basic. There could be other stuff. There are so many kinds of dispositions as today, but most of the dynamics, what is important to predict, to do an inductive reasoning, isn't the action of the disposition. It's not always clear why it guarantee the principle of induction. I think in many dispositional lists, I think often they don't even try to argue for it. It's so obvious for them that if the world is governed by disposition, they don't need to argue for a principle of induction. So my guess is that they defend something very similar to, I'm strong, but in I'm strong it's explicit. In the disposition, it's less explicit. I think they believe that there's enough disposition at play. They are not too diverse. They are knowable. They make the world advancing somewhere. And if we know them or we learn from them, we can do inductive reasoning. Okay. So you see how these three position about loss of nature, trying to put a little bit more precise than just to say nature is simple, equal to itself, blah, blah, blah, nature is uniform. You are trying to give you good reason to believe that if the laws are described by these theories, induction is not due. It's not automatically valid, but it's efficient, but it's not due. Excellent, just one comment. I mean, it's clear to me that in the Lewis case, you have to justify some how the principle of induction because you don't have necessities in the world. So to account for the recognition, you need to say something else. I'm gonna say something about that probably at the top, but in these two views, the spatialized view and the governing view, since you have necessities in the ontology, that's just automatically, I mean, you're automatically performing the, is what I think that should mean. I mean, the principle of induction could be just something very close to our way to know things, but it's obvious that we don't have. So what, it's much easier to argue. I agree with you, but you still have to say that these necessities must be knowable. Okay, yeah. Okay. And they have to be numerous enough or strong enough or the same now and in the future. So there's not new necessity. This is very clear for the disposition, you know, there's no new disposition that will appear in the middle of nowhere in the future. And for, it's explicit in that song that there's no uninstanciation or very instantiation in very late and very far in the future. New properties, not probable, or there won't be a lot of them to guarantee. So yeah, you're right. It's much easier if you have only necessities in the world. You just have to argue that these necessities are knowable and strong enough. Yeah, yeah, yeah. And strong enough. To guarantee inductive principle. You're right. No, it's true. As you see, Lewis works much more. It's very difficult to justify it. Mosaic of facts could be anything. Yeah, something new that always could come up at some point just to try what it would be. So, in this case now, I mean, what's necessary, yesterday is gonna be necessary in the future as well. Exactly, it's why there's all these. Yeah, yeah. All the purchases. Okay. Now, the possibility of changing laws. It will be just a part of this paper. This was published last year. I will not present all the papers, just a few parts. I just have to say before that the goal of this paper was not to discuss changing laws for a metaphysician, but how we can change minimally the metaphysical theories I presented to accommodate something like changing laws for scientists. So, at the end, you will see that some of them are not really simulating changing laws. It's making the thing looks like changing laws, but not really metaphysically. But still, they will put our arguments about the principle of induction in strong difficulties. Okay, first comments, first remark. We don't just want different laws at different locations. We want changing laws and dynamic change, changing a lot, becoming something else. If I say there's law governing electromagnetism here and there's law governing nuclear blah blah blah here, okay, doesn't change anything, not interesting, not really challenging for the principle of induction. There's a notion of local change and global change. Not really useful for this talk. Okay, so there's two conception of what is changing, but it won't be important for this talk. So let's look at Lewis. So to have something that looks like changing laws, we will work only, we will modify minimally the principle of five, and only the principle of five. We'll try to keep the four order on changing. So of course, we will not say there's this two best system because that will just kill the terrorists. The cost is just really good. We'll keep the best system, but we will weaken the second part that regularities necessarily spend supervenom all the mosaic. We will allow that some regularities could be rigid. So let's keep the unity of the best system, but allows for possibilities that regularities do not spend on their whole supervenom base. Okay, is it incompatible with the best system? No, so at least for a limit case, we can prove that it's not incompatible with the best system. This is the argument for the limited case, which is not the interesting case, but it's interesting just to show it's not incompatible. So imagine that W is a best system in the logical world where the list of axioms could be organized and simplified here between a part that is more factual and a part that is more dynamical, but it's just a way to organize the action. Just imagine now another world that is exactly like the first world except in a negligible space-time zone, and this notion of negligible could be formalized. So it's of the big measure of zero compared to space-time. So there's two options for this new system. If this is the best system for W, or we keep this one for W1, we say, okay, this zone is so small that it's not worth it to ax an axiom. It's still the best system of expression, blah, blah, blah. Or we had an explicit description of the properties in this negligible zone. Okay? We can do the same reasoning if we multiply this negligible region for infinity. If there's an infinite number of them, it's not a problem. If the measure of the sum of these regions is still negligible compared to space-time. So at least it's not impossible. It's not reasonable to keep up, to still have a notion of best system even if the regularities in your best system do not span on all the Muslim. There could be a lot of place where it does not. Alexander, would you say that W1 is like W? So it's the same Mosaic, except in a very small region. But it has the, there's a small variation, for example, in the observational thing that you can have in one word. No, no, no, it's not observational. It's in a region, it's not the same natural, perfectly natural properties. So there's something that should be captured by the other, different regularity in this particular region. Okay, it's because I was imagining something like, okay, like the whole argument in the sense that you have two different manifolds and, but in one there is an atom hole and the other one is not a hole. But this hole doesn't produce any observational change. And so here, up here, yes, I mean, you just see different things in the hole. There's different things in the hole, right? No, yeah, exactly. Okay, so this is an argument to say that it's nothing compatible. On the other hand, it's not a very interesting case. The interesting case would be something like that. The Mosaic is separated in two regions. And at least part of the actual do not translate from here to here or the readers. So, or the Mosaic cannot, can be represented by a disjunctive best system or not. And this, we just say we don't know, okay? There's no proof our world is our world more this one or this, we don't know, it will depend on the Mosaic. But at least there's a possibility it's possible that we have a Mosaic in such structure that the best system will be a disjunctive one. So there's some facts that will be shared. The dynamical laws here will not be the dynamical law. But there will be a regime change in the world. Yes. What's the case where it's, can I be disjunctive? So there's two possibilities. If you have a Mosaic like that or you're able to find laws that cover all the Mosaic but will be in the very specific way in the different zones or it's a non-potentic disjunctive system. Something applies here, something different applies here. We don't know that. Maybe there's a very clever way to do the Mosaic in a certain, to do the best system not to have this. And there's no changing laws. In that case there would be no change laws. But still, in this one, in this one, the best system is the same. It's not changing metaphysically. It's changing that way. So what would be changing the laws in the best system? It will be there's a best system. There's one, because it's not yours. And this best system is disjunctive. It cannot be reduced in another way. So at one point you have a change of dynamical laws over the same properties. Supervarying of the same kind of properties. And that would be from the Mosaic point of view changing laws. On the other hand, I don't know if our world is like that. Maybe our world, everything that looks like changing laws or in fact you could subsume them under a better best system. That's possible. I don't know. But the best way to make sense of changing laws for scientists that would be compatible with the best system would be something like that. It's a disjunctive best system. And I don't know if our world is like that at all. Disjunctive one is not going to be predictive enough, right? Yeah, there would be problem. You can see already that there will be a lot of problem for the principle of induction. If you would know either S or S star, that you wanna know under the red line. Yeah, you already see that if our world is like that, principle of induction is doomed. You already have my conclusion on this one. One more question. I mean, I get the concept, but it's hard to me to see a more concrete example. I mean, you already say something like, okay, we live in a Newtonian world and there was a system or something in Newtonian mechanics and at some point, we discovered that in a small region we have a scale, we have quantum mechanics and then you already say, okay, is that just jump to between quantum mechanics and let's see. Okay, let me give you an example, okay? Let's imagine we have certain law of electromagnetism, certain electromagnetic phenomena and they work well, so, and they are predictive and we're quite confident that they are related to the best system, right? Yeah, in a good way, they are. They work very well, they will be in the best system. Suddenly, the same properties behave very differently. Same properties, charge in certain contexts, certain conditions, do something weird. Okay. Okay, so the electromagnetism will, it's still electromagnetism, it's still the same properties, but it's different. We can subsume these two things or there are more general best system. It's not the case at all. Yeah. If we cannot, and we still want to say, no, it's still electromagnetism, it's still the same properties, it could be a disjunctive system. Okay, okay, yeah. So at the end of the mosaic, when we have the complete mosaic, if the best system is really this disjunctive thing, it's like, we would say changing laws. If, oh, no, no, no, we discovered that in fact, we have more general way to build a mosaic, it won't be changing laws in that sense. Okay. Is it better? Yeah, it's better. But it must be the same properties. You know, it's not like electromagnetism and, oh, something completely new. It must be the same properties. Okay. Yes. But I was wondering why you speak of disjunctive case while I guess there's good reason to exclude some kind of group-like solution. And not disjunctive, but say conjunctive, like before the real life is as after one is as. If the best system is something like, oh, it was not electromagnetic property, it was group. Yeah, that, uh-huh. And it's the best system? Ah, it's the best system. It's not changing laws. It was just bytes that, Lewis bytes that bullet explicitly, right? Is that some point in introducing? Lewis explicitly says that that could happen, right? Oh, no, that could happen. With the group example. That could actually, that could happen. The only thing is that these things should supervene on the natural properties. That's possible. If at the end the best system, the more compact, best expression is to go to these metac properties, no changing laws. It's just that we didn't know. Why would that be included? No, no, it's your next student. We don't know. I don't know what would be the best system. Maybe the best system is not to add too many of these metaproperties. Maybe the best system is to have simply juxtaposition of partial best system. I don't know. No, but I mean, if you can have a disjunctive one. Yeah. It seems you could also have a... Yes, but I'm not. Yeah. I'm just saying that if you have, if you want a notion of changing laws, okay, in Lewis, that's without killing Lewis, without killing the best system, it's something like that that you should aim for. On the other hand, maybe it's in our world we will never have that, that I don't know. I don't really see the contingency of that, but that's what I'm saying here. That's the problem with the best system is that we have no idea what is the real best system and what it would look like. Except it will be a deductive system with a finite number of axioms that will express a lot of stuff, hopefully. Maybe not everything, but a lot of... But one of the worst things of accepting stuff like the group case is that you have to formulate such a way that you are referring to a specific time and a specific angle. And that's something... It will be some kind of contextual, maybe... It depends how you define it. Yeah, that's what I was hoping to expect, that you exclude these solutions by saying you have never something in Mexico or something going on. You cannot refer to T.E.L. But the red line, if... I would say that it's true that you might come to something like this, but before you get into something like this, you are going to have to pull a fruit out of... I mean, not to read out everything. Another example, I mean, I should have put a slide. It's that if it's S, S prime, S, S prime, S, S prime. And that, you would say, ha-ha! Because of the regularity of change, I don't need this defensive system. I just have to have a fluctuating pattern of regularities. And that would surely be better for the best system than just... Tuck, tuck, tuck. So we have to imagine a system where there's no... If we want changing laws, there's no more elegant way to express it in the best system. But of course, if it's a regular change, then it will be a new law. It will be a new law. We discovered that this... When you do that... And you don't need this. It won't be changing laws. Of course. We need something like a good fact. But it's always good fact in the Mosaic, but in the best system. Yes, brilliant. Okay, governism. Okay, that's us. It's one of the biggest part of the paper. It's the longest part of the paper, so I will be very short. Because we explore many, many ways. So there's two main ways to get changing laws in Armstrong. Or the necessity relation change among the same universals. Or you have a system subjected to a change of properties, but conserved is identity. And so there's a certain sense to say the law change. I will not discuss the second one. We explore the second one, and I think it doesn't... It's too costly, for what it was. For Armstrong, I will just explore the first one. And first, a remark. We will presume of something epistemologically that I will not justify, but I will argue that and we will just presume it's true. Because there's a difficulty when you distinguish, contrary to the Mosaic of facts, which is fixed, you want to distinguish what is related to necessity, what is not, because there's an explicit difference between contingent and necessary in the Armstrong. And you have to always be able to make the difference practically. Example, I have a gas in a box. I should have bought a box. The gas in a box, you can see that it's governed by certain necessity relation among these universals, its properties. I opened the box, the behavior changed quite radically. We won't say there's a change of laws. We'll say there's a change of initial condition or boundary condition, because we know how to make the difference between the fact that the contingent facts open and close the box, and the laws of nature change it. I will presume that this distinction is always possible, because if we don't do that, the problem becomes untractable. It's so complicated to know when laws change if we cannot know when it's the condition of the universe changing. So I will presume it's possible, but you have to, this is a very difficult problem to make sense of these theory of metaphysics in real physical case, because physicists are not that clear about what is a change that is a change of laws or a change of the initial condition of the universe. And sometimes you would like to say that the initial condition of the universe are almost law-like, but they are not. So I will presume it's always possible, okay? Always possible. I don't know how to make the distinction, but we can do it. Okay, two cases. First, the change between two situations of proof facts. So I have the same properties, the same universal, they are governed in a certain way, they are behavior in a certain way, and after a time on the same system, they do some things. They do some things. If it's, if it happens once, okay, but if it happened once, or if it's not predictable, I'm strong would say, yeah, but there's a problem, because if you have a regularity, if you have something causally captured, it should be captured by my theory, because my theory has been designed to capture regular change. So if it's regularity happening, it's problematic, because the same property seems to, remember the laws are in virtue of the properties of the universal, so the same universal seems to change radically. If it happens once, maybe it's a miracle, but if it's happened more than once, it's a problem. So the second way to see it is to go to, a miracle? Yeah, it could be a miracle, in the Lewis sense, something that happens once, and it's not explained by the Mosaic of that. We're not talking about Lewis here. Yeah, but for I'm sorry, it would be the same. If you have a, You have a lot of miracles? No, he's talking about a book, or what's the universal, he's talking about, he wants to make even the thing that should not be modeled, that you should not be able to model, not to model in the theory, because the theory is supposed to explain all those relations. So if, for example, I have this and it's reproducible, so I have an entomantic system, automatically I'm stronger, go to the higher level. And say, no, I cannot stop there, what's the point? I have to go here. I have to say, okay, this change must be covered by a relation between universal of higher level. So NNA is the law, the universal associated with this law, this is universal associated with this law, and there's an insistation relation I put into here, because, a priori, it's not obvious that it's the same. N, it's the same kind of insistent, because it's not universal describing the first level natural property, these are universal and a higher level. Okay, if you put another in N, Rodrigo Alberto, Rodrigo, Rodrigo had a very typically metaphysical argument, he said, but if it's another N, how do I know it's an insistent? I know it's an insistent because it's itself, an instance of a more general concept. So, if that I have two necessities, I need a third one to be able to see these two necessities as an instance of the third one. But yeah, but if I have a third one, how do I know it's a necessity? And to infinite them, that's a typical metaphysical argument that it's not cool to have a lot of stuff. On the other hand, I think it's easier to see that without the argument at a higher level, that here it's maybe a higher level of universal, but it's still related to physical, because here it's the dynamics of effigy. So here it's a higher level, but it's still physical. If we believe that physics is coherent or that science is coherent and the behavior is coherent, it should be the same kind of necessity related to different dynamics, because we're still talking about the same universal downstairs, the same effigy. Okay, but once we have that, there's a question. It's quite uncomfortable to have the same universal exhibiting different laws, but these laws are itself governed by something higher without there's something incoherent that the fact that you have to go to the second level. How do we see that? It's not obvious, even if it's the same N, why there should be at the first level the same properties governed by different laws. Even if there's a mentalism, because it's governed, so you have to go when there should be some kind of causal relation. So there's two possibilities. Or we can N, or we can universal, and if you look in Armstrong, when Armstrong is faced with that kind of problem of that, not exactly the same, but problem where universal could change different locations, for example. The easiest way is not to touch the necessities to go to quasi universal. So why did we think it was the same properties changing? Because it was not the same properties. The electromagnetic properties here are not exactly the electromagnetic properties here. They are still in the same category, but they were different instance of more general electromagnetic properties in universes. So you go to quasi universes, and you have a notion of changing laws. Of course, for a metaphysician, this is not changing laws. It's just saying that you just get the laws not changing again, but from the scientist position. That would be changing laws. Oh, I have certain electromagnetic behavior in certain conditions. I put the thing in a certain specific condition governed by this, and I get a new kind of electromagnetic phenomenon, but it's still electromagnetic because it's governed by it all. Okay, yes. Why wouldn't it be governed by a metal law? I mean, can't it just like, focus, focus, change? Like, focus, focus, all of a sudden change? Yeah. I assume that there will be something regular behind it. If it's not regular, it changes up in once, and it's not reproducible, and it's not possible to do anything to convince yourself that it could be again the same. Yeah, you would just say good fact. But that would be, that would return to a new case. There's a sudden change, no unexplainable, whatsoever. You could still do a reduction over in the adaptive properties in the environment, okay? Yeah, but if it's not governed by a law here, it's not even here that you will put these two in this asco, the universal, the same universal. Would you say these are still electromagnetic properties if you don't have this regularity of changing of a nomenclical regime? Maybe you would say, okay, now they become something else. If it's something else, it's another case. It's not a case of, it's a different case. Because if I have a system, if I have a system, and suddenly I say, yeah, I have a system. And I say, look at the system, I put the blind. I have a system, I put the blind, okay? And after that you say, I put in, you say, aha, changing, and you would say, no, no, it's not the same thing. It's just a thing that became, that's reverse replacement. It's not change. So you need a certain continuity, and this continuity is through properties of the same kind. It's why dynamical change is important compared to just the position of change or change different at different place. Shut thing different at different place. It's not change. Is there a good systematic theory of quasi-universals? No. I'm sure it's entertaining the fact. Yeah, I remember reading this back, I don't know, there's maybe things, but I don't know if there's a systematic theory of quasi-universal. Nobody really thinks it's quite useful. Except I'm strong. Except I'm strong. But it would be formally difficult because you need to provide a condition of identity and a condition of change. Because you need to say, they don't look the same, but they are the same properties, and this one don't look the same and they are not the same properties. They are not the same quasi-universal. And that's not easy to do. It would need to, maybe to have some kind of dexical instantiation of universals, which is, in my knowledge, does nothing good, but should be done in a certain sense. Alexander, one question. But I have to say, they was a serious, I don't know. I'm strongly serious, he says, well, quasi-universal, maybe the world is like that. There's, it would be much more complicated, but maybe in the world is structure like that. In this case, in which the laws change, all the other conditions are kept the same. I mean, it's... The other laws does not change, yeah. Yeah, but everything, I mean, it's still hard to not see it, like especially in which you're changing. No, it's not spotting, it's changing. Oh, no, but it's not, neither, I mean, you're not changing the domain, for example. For example, I can grab your keys and just throw it to the floor, and I'm gonna see, there's a regularity that, when I just leave the keys, they are going to fall down. Many times, but of course, if I do the same and then in the space, it's not gonna come down, but it's not that the laws change. Yes. But we can always make the difference between laws and initial condition. Okay, but in that case, some of the case, I need to keep this. Because magically, we can do that. Okay, I don't know how. Okay. So I would say, oh, in the case of space, it's conditioned or different, but the laws are the same. Yeah, indeed, in this case, the law is the same. I mean, the same reason why my keys are falling here is the same way. But also, if you accelerate some particles, and you see some, in the case, for example, and you say, okay, this is, I can describe that special through a law, but I could also just accelerate even faster, and then probably I'm gonna see other the case. So, yes. But I mean, in that situation, it's not a change. It's not a law that is changed. You are changing just that. Yeah, now, in this case, we are presuming that you are able to show that it's not a change of initial condition for the same laws. Okay, so to get it right, it will be the situation in which I am throwing your keys here every day, and at some point I'm doing the same, and one day just start to go out. So something like that. No, I mean, do you think that can be a change? Yeah, what? The example of, I'm throwing gifts on quasi universal, maybe it would help. It's John's garden. So John's garden, it's a garden somewhere. I don't remember the details, but there's a wall. Each time you take an apple, and it enter in John's garden, it become a peach. There's no apple in John's garden. Each time you take a peach in John's garden, it become an apple when it goes out. Systematically, all the time. Never exception. Extremely regular. There's no apple in John's garden ever. But each time you take an apple, and you do all kind of condition, you throw that, you do it behind, okay, it all become a peach. Okay, and the reverse is true. So it would say, this is not just an initial condition, it's really property that the rules change in John's garden. Yeah, but I mean, I think that's not very fair. I mean, in the sense that the law there is that, I mean, always that you want to enter an apple into the garden, it's going to become a pear or whatever. But that's the law in this case. It's this regularity. It's not existing. No, no, no. I would say one thing. I would say, no, no, it's a meta regularity because it's not governing the natural properties of what is governing apple. It's very stable in the world. We can do all kind of experiments. The laws governing peaches are, we can do a lot of stuff. But it's a change of the neurological regime. So it's governed by a meta law. Okay, but Anna. It's a case. I think the best case would be a case of diaprenic emergence. If you want my opinion, is that if you have quantum electromagnetic here and quantum fluid here that is governed by different electromagnetic quantum laws and it's regular and we can reproduce it in laboratory as many times, but there's an incompatibility with the laws here and the laws there. That's a better case. Yeah, lady. How is not that just a change of the, again, the dominion? I mean, it's just that you're just changing something in the way with your right of laws and then, okay, you're gonna have different. Yeah, but your question is epistemological. It's not an ethical question. It's how do we recognize these cases? And I agree. It's, I don't know. You know, how, I don't know. How do you recognize that we are in this case? Okay, so our quasi-universal related to a quasi, compared to, oh, maybe we were just wrong about the properties and I played out completely. We were just wrong. Okay, but by the same time. But I don't know. By the same time, we wouldn't know if we are in a world in which Newtonian mechanics is, Newtonian mechanics is the Newtonian second law. It's true, but it becomes something like the Schrodinger equation at the, I'm just circling, with the certain condition. It's not that, quantum mechanics is the true theory. It's just that the Newtonian mechanics becomes. Yeah, quantum. You are going to, okay. That's a good case. If there was some conditions where the world become, the world that is classical, certain condition of energy cycle, become quanta all the time, complicated because of the measurement problem. Yeah, but it's all the time. Yeah. Okay, okay. That may be a case like that. Okay, okay. Okay, so I've been talking a long time. I'll try to accelerate. Okay, so essential dispositionism is super of style to changing laws. So we thought that we could not invent something that's plausible. And I have it here to present a very strange argument of Pauline phrase, but we all signed behind. We all said, okay, yeah, then after working with Paul, the only Paul can think about something like that. And it's very clever, but very weird. So, because it's a theory that is really against changing laws. Essential dispositionism is really against action. So to do the demonstration, we'll work on a toy model. Okay, let's imagine that Huxla, the law of springs, is a fundamental law. So springs are like quantum fundamental stuff in physics today. So it's a fundamental law. It's not a composition of other laws, okay? Just for the, that is fundamental in the world. This law asserts that the extension of a spring is proportional to the applied force. That's just the Huxla law. We will also presume, which is not acceptable for every dispositionalist, that we can identify what to be a spring, the universal to be a spring, and universal special extension, and universal force independently of the Huxla. So I can't say it's a spring, even if I did not test that they are spring in the sense of the Huxla. Not all dispositionalists would accept it. For the example, yes. Though, so let us suppose we can identify springs and extension independently of the Huxla. Okay, so in our world, world W, the Huxla takes this specific form. Let us call this law L. Let us suppose that in another world, very close to our world, one finds a similar spring. A similar extension of a plan, and the Huxla is a little bit different. It's another constant. So, dispositionally, if we apply the same stimulus, we don't get exactly the same manifestation, because it's a little bit different. So according to a very strict dispositionalist, we would say it's not the same law, because it's not the same disposition, because it's not the same manifestation for the same stimulus. Okay, so a very strict dispositionalist would say, okay, let's imagine now we are not a strict. We're not a strict dispositionalist. We want to make a difference between very different disposition. We would say if it's just a little bit difference in the constant, it's a dependent, it's a little bit difference in the condition, the initial condition, but it's still a very similar disposition. Let's now make the thing a little bit weirder. So now let's suppose that one final law of the form KX2. So again, the strict dispositionalist would say, yeah, it's complete. No, it's not the same stimulus, the same thing, but the one that accepted before that, okay, there could be not exactly the same manifestation if I include a little bit more information about the initial condition in my disposition. Would it, or she says it's the same? Depending of what kind of dispositionalist you find, some would say, okay, no, it's still the same. So in that case, because qualitatively it's the same. Or they would say, no, no, when it's X2, it's quantitatively too different, even if qualitatively it's the same. Okay, yeah. Why is it qualitatively the same? Because it's still proportion, it's still a direct dependence between extension and force. Okay, I'm not sure I master what the metaphysician call a qualitative compared to a quantitative, but for them, qualitatively, if I have extension and another stuff, and it's always the same, even if it's a different factor, square, it could be the same qualitatively, but not quantitatively. No, it's not the same. For a strict dispositionalist, it's never the same anyway. Each time we have same stimulus manifestation is not exactly the same, it's not the same disposition. That's the strict. But we could have, at least in principle, if this example is not completely crazy, you still have essence about qualitative relation that differ quantitatively. So if I can identify springs and blah, blah, blah, independently of Oxford, I agree that the essential part is only the qualitative relation and not the quantitative one. I could have the same disposition, qualitative, that change quantitatively between, at least in principle, between two. And that, for us, is the best at dispositions, essentialists can do to have changing laws. If you think it's too strong paid, it's too cocky. Okay, disposition of exclusion changing laws. But it's the minimal change we could find to have something that looks like changing laws, at least quantitatively and not qualitatively. Because of course, if you change the essence of things qualitatively, you're not an essentialist. This position isn't anymore. I didn't get the result. The difference in the first case with the key to key prime? The first case is that, okay, in each case, the strong dispositionist will say no. Okay, because it's same stimulus, different manifestation. Completely different. The first case was to make you more agreeable to at least nuance the fact that you could say, yeah, there could be different manifestation, but it's still mostly the same. And why is that not enough? That was a convincing case, but that is a rare thing. Yeah, but the first is not a changing laws. It's a change of the initial condition. It's a change of the way you build the disposition. The second one is supposed to push you a little bit towards, maybe it's not explainable just by initial condition. Yeah, I think if you're an essentialist, dispositionist, laws cannot change. But if you want changing laws, it's the best you can do. You might have similar cases for the, I don't see the big difference between the government and the government in the first case. There's very strong argument against meta-disposition. That's why you cannot have like the... But the disposition might be actually real disposition, not a meta-disposition. Yeah, but if it's position to change... But if, okay. But that is maybe... If you have a change of disposition that is regular enough, and you want to explain dynamics in the world by the action of disposition, essentialist disposition, you will need a disposition that a certain way represent this change of dispositional regime, which is a meta-disposition. If it's a meta-disposition, there's a good arguments that they cannot be, it's a flat ontology dispositionist. You cannot have meta. So if it's not governed by a meta-disposition, it's governed by something else. It's maybe a by-product of the action of the disposition at the first level. If it's a by-product, it's in the manifestation. It's in the disposition. So it should be already there. So it cannot change. Because the manifestation to be... To appear to change disposition should be a manifestation of the disposition. It's in the disposition itself. So there's no change. I mean, it looks like change, but it's no change. Yeah, it's not convincing, I know, but it's the best we can do. It's absolutely right that if you have a disposition, you have some manifestation, and you could expect that always. But you also have the dispositionist ontology. All these very, very complicated real-life masks, mimickers, and disposition interweaving in each other, things like that. So in different cases, someone could say, okay, yeah, I mean, this is the disposition, but now we have a different look because there's another disposition in the real or a mask that is impeding this disposition. You're right. It's true that if you have a bunch of dispositions together doing something, and at the level of universal, like, biological universal, it could look like changing because it's the same fundamental disposition, but organized differently. We exclude this case. We are only working at the fundamental level. But of course, you're right. It's probably more plausible to explain, for example, evolution change with disposition based on, yeah, the way that this disposition is organized is different enough that at the level of observation, it looks like different regularities. Yeah? Yeah. But that, yeah. No, you're right. It's an interesting case, it's just that we did not work. Okay, okay. At that level. Okay, so conclusion. It will be very fast. If we have such a kind of change in the mosaic, specifically disjunctive best system, difficult, that difficult for inductive reasoning because if we were able to anticipate from this region to the other one, it would not be a disjunctive best system. So it's still an epistemological challenge, but they have hope. If these change are regular enough for us to identify the meta law and to be able to guess that we are facing quasi-universal, we can accommodate a certain change. If they are regular enough, empirical or nothing. We could restore our confidence. Of course, the first time we saw this case, it would be, oh my God, what is happening? What should we do? The laws of nature seems to change. But if we can have a certain empirical control because they are governed, we could be able to restore some confidence in inductive reasoning. Of course, here it's more problematic because, yes, there will be a qualitative continuity, but such a non-regularity at quantitative level that if they are brute facts, if they are not explainable by a meta laws, by a meta disposition or by something very regular, for example, people that accept more than an essential disposition, they accept also categorical properties. There's a lot of problem, a lot of difficulties to overcome, to restore the principle of induction. So conclusion, if you want to have changing laws in a certain sense and still be confident in prediction and things like that, a theory that explicitly allows for meta laws, like the quasi-universal version of Armstrong, is exactly what you need without surprise because there's meta laws. The fact that there's no obvious way to have meta laws in Lewis because it's the same deductive system describing a plain Mosaic affair, and the fact that this position also is a plain ontology makes that if there's such a change, a change of the homological regime, we are not that confident that the principle of induction is compatible. No, I'm still here, but I already talked a lot, so. Good. We discussed the dispensational parts on the back end of the, for the Lewis part, how much does the measure of zero do with your proof? And to translate the, if you take a concrete case, just translate the opposite of the singularities in spacetime stuff, because then you would have, you could say that it's the laws of generality, it applies in the whole spectrum, even if the case of singularities exists because of measure zero, right? So if you have said this kind of proof, you could say that generality applies everywhere on spacetime, and there is no issue with the singularities. Yeah, but singularities are not, okay, singularities are, the horizon is not just a point, so it's not a good measure. The horizon, no, but the singularities itself are not just zero. If you have the, Yeah. If you don't have a cosmic, there are some changes that you could see. Yeah, but still, you have phenomenon, if we had only generality, and no quantum stuff, maybe I would be inclined to agree with you. But because we have other kind of phenomena that are on the horizon of what happened to singularities, that seems to change the game, the rule of the game, I am less inclined to see an analogy with this case that is very formal, it's just a proof of existence. But if we had only, only relativity, and no other quantum stuff, on the frontier, maybe. So, the relativity itself is not a flawed theory, it's a fact that, No, but relativity is a flawed theory because there's other kind of physics. Yeah, that's amazing. If everything was governed by relativity at a certain scale, and we were completely happy with that, yeah. But it's because relativity does not work well at the beginning of the university, and there's also quantum phenomena at the frontier that we can at least have prediction. It's not just, we don't know what happened inside the blah, blah, blah. So. Does that serve a reason why does the vision is always important? Well, it's important to be sure that the best system, you know, it will still be the same system because if their zone is very wide where there's the different physics, obviously the previous best system is not applicable. Okay. Agree with your conclusions? But, to me that for induction, I mean, the word inductive reasoning is kind of fake. And so, when these metals get involved, it seems that you need other inductive principles. So you induce metal laws rather than balls, or both of them, but they need to distinguish them. And if you do hardcore induction, without like opening up to this possibility, and then like a formal possibility, there's a formal difference between the two kinds of laws and levels, maybe more levels than two, I don't know. The induction is going to go wrong on a formal level, it seems. Of course. So it's not the principle and the principle is safe, but it's not the law. It's because, yeah, I present that in a metaphysics conference where we discuss the principle of induction. Of course, if I have a strict Bayesian system, and I don't take an account normally in a strict Bayesian system, I don't go to a higher level, it will fail. Until it sees enough of these regularities to build another base to, blah, blah, absolutely. Yes. But if you build another base, it seems to get rid of the, of the problem. Of the changing laws. Yes. And then basically, yes. Yes, yeah, you're right. And I was also wondering, in the deductive system case, couldn't you, of course this was not what David Lewis was thinking of, but I mean, there's no, this is purely syntactic, right? The deductive system might just have two levels too. Or for you, I was already, it's like. Yeah. I have another. We were discussing that. I was, I have another paper where, where it's still under review, where I discuss, could you have metals in the base system? There's a part of the paper about that. Because the base system says nothing of what is in the action. Is it facts? Is it first level laws? Is it metals? The base system is the base system. It's a list of stuff. Finite with deductive rules. That's it. We have no idea what are in these, except at one point they have to talk about the properties, the natural properties, but how is it that the, the base system is just saying, there's a finite list. And you could ask yourself, what kind of stuff could be there? What kind of stuff? And you could say, okay, maybe there's facts, maybe there's first level regularities, you know, something that's really expanded. Maybe there's metastoff. Nothing is excluded a priori. But there's some constraints when you think about it. One of, for example, if you have a Laplacian, a Laplacian, if it's sufficient for the base system to have a Laplacian deterministic detective system. So you have a certain number of facts and certain number of differential equations with time. And it's enough to predict most of the, what you need to predict. You know that in this system, hiding a metal arm would be redundant because it would be an automorphism of the same model. But if you don't have such a system, just imagine we have, there's no way to describe the mosaic without some probabilities or there's no way to describe the mosaic without some part that you don't know. So some stochastic aspect. Maybe there could be, in the list of actions, something that looks like a metal arm. It will relate to different models of the same basic other part of the system. That's possible. But there could be, no, they're surely, if we are very careful about what could be a base system, we could surely exclude some things, but a priori in the best system is anything that is deductively powerful enough. So there could be very meta principle there if they do the job. But they must not be redundant. That's the point. Yeah, I wanted to ask about, this is really cool. I wanted to ask about, I don't even think it's a tension, I just think it's like a, so you said at the end, and I'm like, this is what made me think of this. You said at the end, like, oh, you know, in some sense, this is a very surprising, right, that to make this work, you need metal walls. But I actually want to push on, because on the one hand, I think it's exactly right. Like, okay, sure. You know, if you want to be able to do induction across law change, you need metal walls to guide that. On the other hand, this is the direct kind of a sort of backwards way to phrase that, that is a little surprising, which is, you know, you guys did all this work in that other paper to try to find these, you know, minimal interventions on these systems of laws that could make something that looked like a good replacement for changing laws. And there's a sense in which the conclusion that you get to hear when you think about induction is like, yeah, nice try. But no, like if you want to change, if you want to be able to do induction across changing laws, you don't need these like fake changing laws. You need like real, metaphysically, really real, exactly the thing that you were saying in the beginning, like you were hoping you could avoid having to say like, oh no, it's just actually a serious change of law. That seems to be what you really need, right? Yeah, you know, it's well said. This paper I'm working on shows how wrong my previous paper was. Because at least I thought that at least in our previous paper, we were changing so minimally the thing to keep the metaphysics sounds, but make the scientists happy. They say, oh, okay, I see how it's compatible with what I see. It just means it has an unfortunate consequence that you did not realize at the time. Yeah, this unfortunate consequence is bad enough that maybe it's not the way to go. Yes, so the answer is yes. Is there any other question? Yeah. I don't know if I'm full of myself if I try to undertake the technical execution, so just to ask, what do you want to say is a pretty good foundation? Why don't you buy the UMEA? Because you don't think you're reconvinced about what you present. So what is the UMEA? There's absolutely right. There's other ways to discuss induction in a very epistemological way or we have a certain bias towards regularities and so we go to, there's all kinds to discuss induction even in the worst case and say it's not so bad. But what is interesting is that this principle of uniformity of nature that seems to have played a role inside the history of science for hundreds of years in a form or another. Or it was the University of God, for action of God in the car. Or this analogy of nature, which is a very strong notion in Newton or in Mills, this principle of uniformity of nature. They always seems to have a connection between why can we do science because the world is such and such. And these metaphysical theory of laws were supposed to do that. They were, of course, the main idea was to discuss counterfactual. There's all kinds of things they wanted to do with this theory, to explain metaphysically, to have a ground, a metaphysical ground, to explain a lot of the practice. But of course it should be compatible with the fact that science is inductive and working pretty well. So it's why, but on the other hand, you know, probably the good answer to inductive wariness is surely not metaphysical. It's maybe more in the practice, in the way we concentrate our research on the part of the world that seems to fit, blah, blah, blah. Maybe we have a bias, you know. We were discussing that this morning, you know, the physicist Vignaya, who won the Nobel Prize for the introduction of group theory in quantum mechanics. So the more meta-law you can imagine, you know, everything was coming from the meta. Explicitly says, yeah, but you know, there's laws and there's non-laws. We are interested in that part. Who knows if the majority of the world is maybe chaotic and non-nomological at all. But this is not what we are interested in. So maybe it's a selection bias. In fact, the principle of induction works just because we select for the phenomenon where it works. That could be a very neo-Cantian answer. I'm putting this paper in a certain context of people that believe that metaphysics is the answer to almost every problem, but you're right. The last proposal certainly seems true in psychology. But this idea that we concentrate on certain subjects and not on other, or we project certain things on the world and it fits more or less, and it justifies our practice, it's not ridiculous when you look at history of science too. So it's, I think these neo-Cantian conventionism, very seriously, I think they, maybe, they are absolutely right. On the other hand, there's a lot of things that seem regular and why I believe that, not because of science, because of technology, the stability of technology, the reliance, the fact that technology survives to dramatic theoretical change. So there's something in the action in the world that seems robust, at least in a way that seems robust, at least in our scale, at our scale of the world, in this region of the universe. Combining those last two answers, right, to give you, I think, a better answer to that question of the relationship between this and the last paper, right? So it's not just that the last paper is wrong, it's that you're mapping the like, every move here has a philosophical cost and you're mapping the cost structure philosophically speaking. Right? So, I mean, because I guess, you know, we can always, we can always take two steps farther back and underline. So the other option that remains here is to double back and reject a different premise, which is now we can go back to the physicists and say, guys, stop talking like they're changing. Like, knock it off. Like, this just doesn't, there's two, the costs for that move are too high. The costs for that move are too high. Get over it, knock it off. Maybe that, you know, and like, so that's always, you know, there's always, there's, you can play with each of these independent variables a little bit, right? I think that's kind of a cool way to think about it. Maybe, so maybe induction is the variable to play with, right? Maybe the nature of the laws is the variable to play with. Maybe the physics is the variable to play with. But like, that's kind of a neat, you know, it's really cool to see how all these things are, how they're interrelated in that way. Thank you for the suggestion, because I always trust scientists more than this philosopher, and so I presume that when they are talking about let's explore the possibility of changing laws, they have good reason for it. But maybe, maybe if one of these theories of metaphysics of law, so we have to believe that they are a good approach of the metaphysics of law, show that it's too costly, yeah. Surely enough to be worth exploring, yeah. Yeah, but now people can see the breaking of this kind of stuff. If you believe that it's too much of a cost to change laws, why we actually believe in this inter-breaking, because inter-breaking is an easy case where you have two different kind of set of laws in the last course of time. Yeah, but if spontaneous inter-breaking, you can always say, yeah, but these are a regime of a meta-lock, of the end theory. That's the point, you can say that you need a meta-lock to do a lot for this. You need a meta-lock, okay, yeah. So that means spontaneous inter-breaking, the instance of a meta-lock. But I didn't want to... This idea, when we discuss about the end theory and all that, in fact there's this theory general and there's all these different regimes that seems really different. It looks like philosophy. So we have no idea what is the end theory. We have no clues of what it is. If you don't spoil a strategy breaking at the start of the year. Yes, but not that it's a sign that there's a more general stuff. But if you believe in metaphysically, but cosmologically, they can seriously... speaking, you have no idea what they're saying, but it's generally two different kind of law set of laws and there is no hierarchy in principle. Because there should be one, otherwise we definitely have an issue. So it should be Armstrongian. Sociologically, speaking of taking all this seriously and how we should react to it, I read all of this stuff. Does the existence of these kinds of cases and things like end theory, are people taking it more seriously? Because I remember the meta-law discussion, there were a couple of articles poking in with Arnett, like that one bit of Armstrong, where he talks about its sum and life. Is that, are people picking this up? In the metaphysics tradition, there's not a lot about that. To my knowledge. A guy, I don't master all metaphysics. To my knowledge. But of course in science, there's always been discussion about level of laws and principle of invariance. Are they laws or principle of conservation? Are they meta-laws or constraint over the laws we can build? And even, is a biological laws a constraint on the physics thermodynamics of the being? Or the thermodynamics is a constraint on biological laws? There's always discussion of different nomological regimes in science. I guess you've got more blank stuff, but that's metaphysically, but in metaphysics, to my knowledge, there's not a lot about that. Because it's tough, you know. There's good arguments, there's strong arguments that it's very difficult to explain the relation between a law, even in Armstrong, and its instance. Why govern? There's something non-temporal here, there's something temporal there. It's goes all here. If you had a level like that, it's just making all the problems of Armstrong much worse. Much worse. To the point that some metaphysician would say, okay, it's done. It's done. I'm sure it's done. Let's do this position. It's one level. It works. So yes. I'm not surprised that they did not explore systematically that kind of hierarchy for a good reason, because it makes the theory weaker. On the other hand, there's a famous exception. Langer. The theory of laws of Langer is built to have a hierarchy. It's built to do it. It's explicit. It's all designed to be able to make sense of how physicists talk about the level of the neurological level. But it's a very specific, unique, interestingly weird theory. Yeah. You also did mention Langer, because I wanted to talk about the tree standards. The tree standards. Of course, there's other interesting. There's primitivism that is very more and more popular. There's where the dynamical principle are not explained by anything. They are just there. There's Langer that is always a candidate around even if it's such a bizarre approach and it's different. You have to buy tough things about uncontextuality of counterfactual facts which is difficult to buy. And there's other, there's of course one of the reviewers always saying divine decree. A law is a divine decree. It has been very popular for thousands of years. And that, you can have changing laws. It's the will of God. If you have a God that can change his mind or her mind the laws can change. If you have a God, a lightning god that cannot change that is submit by, that has to submit to rationality laws cannot change. So of course, there's other possibilities. I just chose these three because they are the more discussed today. But maybe, maybe they are not the good one. You should be confused. The idea that there should be a theory of universal that, that of course universal that is very well made, that is indexical, that has something interesting that should exist. If we had such a formal theory it would be much more interesting because we could have a framework to understand how necessity relation here transform and necessity relation there and we could we could have the conceptual framework to understand the continuity. If we had universals that are not these Platonistic or Aristotelian stuff that are the same for all eternity but I don't know how to build this that kind of formal theories. Sort of process ontology Yeah, but process yeah, could be a pros, I don't know. I don't understand process ontology also. I wrote a paper about that but it's wrong. Like everything I wrote in mind. It's all, yeah, it doesn't work. I have a question. It's about you were talking about how this could accommodate principle of induction and at some point you were presenting these five principles as the structure of the human mosaic and it's something like that those principles guarantee the principle of induction. Among other things. Among other things, yeah. But at some point I had this doubt that some of the principles could be seen as already assuming some kind of principle of induction. For example, physicalism. Suppose that you are the human and you say, okay, I only material interactions or constant closure or something like that. So what you are saying is that we have this regularity that always material things interact with material things. But how do you know that in the future probably we are going to have minds that emerge and they start to interact with the physical stuff. So at some point you are assuming that the interactions are regular enough to suppose that in the future things are still going to interact with things. But who knows? You are absolutely right. This is something in metaphysics that is difficult for a philosopher of science to swallow. Is that in the metaphysics, they built everything they want by hand. So you are like if there is a mosaic, fundamental physical, meteorological reduction, universalism, blah, blah, blah you have everything you want and you say, yeah, but you just put all the consequence by hand. They say, yeah. But disposition. Dynamics has been a mystery for thousands of years. Where change is coming from? What is changing? It's in the world. It's in the disposition. Do you need to explain change? No. It's in the disposition. Okay? And they would argue, okay this is extremely bad because this necessitation relation between universal, it's difficult to understand how a necessitation between universal that relate to certain kind of properties, connection between property generate change which is temporal in the world. So this job is very fishy in Armstrong because it's not put by hand. It's still a gap. You adopt that. So we should be charitable. The goal of metaphysics is not the same goal as philosopher of science. The goal of metaphysics is to build nice pictures of the world. Pictures that help us think and relate problems together. But you're perfectly right to say, why? So the fact that we have to add these five things should be a very good reason to think that the Mosaic of facts is to it's just not rich enough to do anything but Lewis would say, no, no the Mosaic of facts I have very strong reason to believe that facts are like that in the world. Therefore since I want to to get to all the interesting results I have to patch this system to make sense and it's coherent with my causality of time travel or take the mode of stolons and say that the Mosaic of facts is what we have access to. Science is possible therefore these five things are true. Which true? Sometimes he kind of sounds like that. Science works so the fact that science works is a good reason to believe that these five hypothesis are so strong and rich. And that's the bizarre thing about metaphysical project is that they should stop to say they explain anything because to explain is not to have the answer in the premise it's not the goal the goal is to have a picture that will help you to maybe think about other problems and other things. So if you have, for example the Armstrong conception of laws and metal laws maybe it would be interesting to think about interesting case where there seems to be regime change with the same kind of properties. And okay how would connect to for example my theory of our theory of Olivia and me of transformational emergence. Maybe it's the right frame to think about transformational emergence. Maybe if it's true or false and whatever it explains this position will never explain change ever it's in it you just put change in it by hand so the goal is we should not ask this theory to do something it cannot do but it but your comment is very because of this limitation of metaphysical project I understand why some philosopher would say maybe it's not worthwhile to invest in metaphysics. Makes sense of course if you say that to metaphysician they don't understand because they believe metaphysics is the center of philosophy but yeah I find birds interesting. How far can we go if you cannot do everything putting by hand almost everything that is complicated it's interesting we learn something when you read the book of Alexander and at the end there's this bizarre passage arguing oh yeah there's the problem with symmetry maybe my theory is incompatible with modern physics but probably physics is wrong we learn something we learn something it's going to the end of an idea to the extreme consequence of a metaphysical hypothesis and it forces you to say that actual physics is wrong because it's multi-level and it cannot be in reality I didn't go that far to say that we shouldn't give funding to metaphysicians but we have to put metaphysical project at the right place they cannot do more than if you want to justify the principle induction through these five principles you have to assume at some point the principle induction so in that case it's not giving any justification or explanation of the principle induction because you already assumed it but the gain of Lewis' program of laws I think the main gain is that could you get all the counterfactuals claim without positing in the system some kind of modality power so at least that's why it's an interesting metaphysical project especially for me is trying to show that you don't need this position you don't need necessitation connection between universal to get at least if the project works to get to make sense of the way scientists talk about justifying counterfactuals