General Discussion — Bitbol, Grinbaum, Pooley Part 2

0:00 The metaphysical realist, I know, Michael Atiyah, rightly, he coined it after, with a good impression, he plans to be a coherent realist, first ontology, after epistemology, all that Kant tried to do is before epistemology and after ontology. And Kant says, if you like an approach, a correct approach, according to the problem of knowledge, you must avoid to approach the problem, having in mind the ontological presupposition of the distinction between subject and logic. I remember always the victim of a theory that he applies to Kant as an ethical machine. The philosophy of Kant's theory of knowledge does not know from the beginning the distinction between the sound, the mind, the element, and the peace. See the point. No surprise, let me take a couple more weeks beating my own drum here. So, first again, the moral. With an example, hydrodynamic description of water, it's an idealization, it's false, but it gives us real understanding. It gives us real knowledge of the world. Now, I want to apply that moral to the line of thinking that's gone on about realism. Let's start with everyday common sense realism. I hold up a red piece of paper. You know, there it is. It's real. It's red. If this isn't knowing what's real in the world, nothing is. This ought to be a paradigm case. what I see what I feel is problematic is that we interpret this case by saying oh they're objects and they have properties so that's what it is to be a realist it's to identify objects to identify properties and say which ones go with which but wait a minute when I hold up a red piece of paper I'm already indulging in an idealization an idealization about the object

2:30 now make a different point about color perception than I did in a talk. Color perception, semantically, it's treated as a non-relational property, but the actual facts about color are deeply relational. So it turns out that our paradigm case of knowledge already involves deep idealization, and so this, I'll call it, hard-boiled realism of out-there objects and properties and which ones go with which doesn't work even in this primary case. So then I'm back to what's exemplified by the hydrodynamic model. We have different ways of getting at the world. These constitute real knowledge, but it's not in the way of what I just tried to describe as hard-boiled realism. You don't think there's an element of hard-boiled realism in the hydrodynamic approach to water? Not with what I mean by hard-boiled realism. Again, there are objects out there, they're completely determinate, I have names for them, I have names for completely determinate properties. I just meant an analogy to the paper being read. No, no, no, so you understand now what I mean by hard-boiled realism. Yeah. I mean, I'd like to say there's a lot about the hypnotic description of water that's true in whatever useful sense of the word true. Small t. I guess. So I'll take a full challenge again. So I'm joking, I'm persuaded by the red piece of paper case. So I'll go on to this that red is a color gap, so you can set up of shades of red and we pick up a series of pieces of paper, on the one hand is one that's definitely red and on the other side is one that's definitely yellow and then there's shades all the way in between. We don't in our normal use of colour predicates distinguish sharply between very tiny differences in shades. Nonetheless, that doesn't prevent us from saying one's on this side is definitely red and one's on this side is definitely yellow. so you can have predicates which have what cases of which paradigm cases are being read paradigm cases are being gathered yes the application of the predicates can be a barrier but that doesn't present preventative facts

5:00 about what the central applications are so I think that kind of structure already takes into account what could be true in the sort of idealization in the case of relevance and it's not something that raises any problem for the readers at all even if they're not saying something absolutely precise, they're saying something I was appealing to an entirely different way in which appealing to red involves an idealization. I take it that the semantics of the word is going to be the same for a person who's not aware of the fact that the facts about color vision are deeply relational in someone who is. From that I conclude that when the semantics of the word has the form of attributing a non-relational property to the world, that is, it has the form of seeing red as, you know, whether it's a graded property or a specific one, as a completely determinate, not in sharp color, but maybe continuum of shapes, which is out there independently of facts about the observer and facts about the environment. That's what the idealization Can I just ask for a bit more clarification about what you mean by semantic? It seems to me that exactly how we should think of everyday color is controversial amongst, you know, philosophers who worry about color. But, you know, the view I kind of like is that the real issue is not between the semantics and, you know, the relational aspect. It's between how colored things are played to us in experience and the relational aspect. So the writing industry will, when I look at a colored thing, the color presents itself as being, you know, a non-relational property. Now, I think you can question that. When I look at this blue bit of plastic, it appears to have a surface property which reflects light in a certain way. So you could think that what's relational about color is its disposition to reflect light.

7:30 So I think you've got to say a bit more to convince someone who wants to cling on to being able to completely endorse everyday realism that there's some idealization going on. I propose to move on to the next question. I would love to address that, but I don't want to... We're going to get in a circle here, and I don't... We've got lots of topics to talk about. Do we want more questions, or do we want to stop? Oh, no, if we're going to stop, I'll respond to that. Where is that? I wanted to ask a little more of these extra realists. So one thing that's puzzling me is the relationship that the New York binds between common sense and science and realism. Contability of common sense and science and realism. So, as you know, as a young man, I was a realist and I was sitting in a cold box. But I mean, the Middle Republic of the mid-70s wanted to say common sense can be completely screwed up. So when I defended realism about relativity it was, hey, it was really real in space-time. We used to think about space and that's just wrong and now we can. But it's real. It's important to me for some reason to follow the Middle Republic of the day. It's real. With quantum mechanics, I wanted to defend realism It's a good Hillary Putnamian. I thought, oh, well, the logical gentrification is good because it can allow you to be real. It's not something that's a good person. But the world is really strange and the logic is strange. But that's fine because everything is real. We all have the property. Obviously, this is not the game. and as you see, now I'm still into the scientific evolution. That's what I really like about science. It's just the fundamental concept. Now I know it's a real thing, but I want to somehow be a compliment for that. So, I mean, given these, well, one way I can ask is, you said, well, one problem with what sounds to you about how to increase, oh, there it is that the 3DF can work. Yet, how does that exactly fit between common sense and science?

10:00 Because you might think common sense is kind of like a problem. I'm having trouble... Isn't that that's more a meta-level problem than a ground-level problem? So the common sense of realism is based on the thought you can understand that you can have a reasonably precise view and the world is concerned that they're not that true or false. It would be like a table with the bottles on the table. But they're the same. The bedrooms and chairs and electrons and photons are at the same time. That makes me really nervous. But of course, John succeeded in making the lesson on this a moment ago. And the meta-level problem might be that when one comes to one's description panel, it thinks that the common sense talk is functioning, and that needn't itself be a common sense thing. What needn't be a common sense thing? Well, I've slightly lost my train of thought. What was exactly the problem you thought there was, or the tension you thought between a common sense? Well, I just heard a little bit of a tension when I thought I heard Oliver say the science of common sense has to be continuous and this is a fundamental thing which we can use in the next degree but then he went on to say the reason we don't like continent kind of views is that they want to be morphed and we don't want to be morphed because one might think the common sense view is that we start from our point of view on earth and the way things appear to us in everyday life and then we learn more and more in science and our thinking in the anthropomorphizing way that behind that common sense thing is something that's really quite So my point about the meta-level, ground-level distinction is that the concepts one enjoys in normal talk about the everyday world will typically be ones that are more sensitive ones that we divide, and then when we're doing science we try a little bit harder to let the world force on us the appropriate set of concepts for its description, and those are sort of different kinds of things. but where the continuity comes is in the nature of the world relation the sentences about macroscopic object can be as simply transparently true or false and true or false in the same language, the same kinds of facts about meaning language and the world as more research scientific state that we've had to go through a process of concept creation in order to say the notion of truth and knowledge framework

12:30 give Tom I mean it's just just a brief comment on what Chris was saying it's brought to before an issue that I've never really understood which is that our semantic practice in using is predicated on objects in front of us, red objects, blue objects, shaped objects, hard objects, objects that appear and don't appear. That's really where our semantic practice, our linguistic practices begin. And then what science does is somehow boost from that. And the idea that you can apply the same kinds of ordinary language semantics that apply to middle-sized dry goods quarks and, you know, gauge vector bosons and all the rest of it, seems to me to have been an extraordinary extrapolation of the notion of a semantics. And somehow the Tarsky schema is supposed to apply here as well as, you know, to snow as white is true, if and only if snow is white. I've never understood that. Well, what do you mean by apply semantics? Well, it's theorizing about semantics, and then they're saying some things which might be true or false, of things in the world. So I'm not sure where your worry is, because we can still manage to say true things about the world and manage to pick out what we might suppose are these pro-telling agreements to be... Well, there's truth in the world that's always examined in a court of law, and you swear to tell the truth, that's going to help you. And that's about what you've seen or what you've seen, reliable evidence. And that's mostly perceptual and it's intersubjective. and it's not highly theory dependent and this notion of evidence that you can boost into your semantics to apply the whole you know, facts about the world and truths about the world seems to me to be just an amazing extrapolation Many people try to say that semantics are not the logical positivist they say, okay, don't take things at face value like referential statements about things into complicated statements and logical constructions about sense statements or something like that. That proves to be a pretty hard thing to do.

15:00 And so it's a better starting position to say, take the semantic structure of the language at face value and then worry about it and determine whether a lot of things you take yourself to be referring to or the statements you take yourself to be making a true or false. If you do that, then you're probably okay. I just want a small interjection. Antics, as a philosophical subject, actually didn't arise from worrying about ordinary dry goods. Starting with Frege and going through Tarsie, it arose from a certain concern with mathematical language and its logic. And Tarsie's theory of truth is arising in that very context. It wasn't arising as a theory of the meaning of ordinary things. What we call philosophical semantics has a place in a certain mathematical tradition. But there's another tradition in semantics that would be linguistic. Yeah, that's true. Yeah, I'm going to change the topic. So, I want to get... That was enough philosophy, guys. I have a question for people who do quantum gravity things around. To go back to this discussion of systems, we had several different points of view on the need to talk about systems. Do you feel that any of the languages, you know, just to refer to this language discussion, do you feel that any of the existing languages can be fruitful enough to provide a framework, to provide a language that would give a theory without the notion of system? Without a notion of system? Yeah. So the actual theories of physics don't usually contain the notion of system as part of their terminology. So the standard model talks about leptons sports. It doesn't talk about systems. Well, right. Yeah, but no, that's a particle. A particle is a quantum system, in a way. Well, that's what... but that's not something you read about it's a part of the language that's something you read about when you're talking about all sorts of different kinds of quantum theories and trying to find what's in common and say well we're studying a system I'm just saying that I think that

17:30 physicists write down specific physical theories you can say okay they're describing a different system but I don't think because the term of system needs to be in those theories and I don't think we need to really know what a system is very carefully. But it's used a lot in, you know, experimenters parlance and, I mean, in people trying to design quantum computers or actually build devices. So I think in non-relativistic quantum mechanics, you might find it a lot more than you find it in quantum field theory, for instance. just because maybe because it's more like it's more operational it doesn't come in as naturally in quantum field theory if you're assembling black boxes and make them the other things that you want them to do the idea of systems it's very important boxes and systems so one of the two as I see it these two big questions in the foundations of quantum theory, not just mechanics, theory? Which one is a question we didn't really debate at all, so let's just, I'll briefly sort of mention it, the question of why real numbers, why real complex numbers? That's one question. Then there's the other question of what about compound systems and subsystems? Where do you get this idea that When you put two things together, they form something of which, you know, dimension sort of adds up and things like this. So the rule for bringing this from here and this from here and treating them as one. And this is not just it. Well, the usual place where this appears is mechanics or quantum informational that say things and the tensor product problems and things as such. but not only we're using this all the time in field theory too we write amplitudes for reactions where one particle will give to others and we still assume something about how dimensions so this is a common thing but this somehow presumes that we have this postulate about compoundness and being a subsystem that. No, for me, it's the contrary. The problem is not why we bring these two things to make

20:00 a system, but the problem is why we consider this system as two separate things. So this is exactly, and for me, many problems of quantum mechanics come from there, because in fact we have one way... But the problem is still there. In this reformulation, the problem is still there. I'm not sure. Well, I mean, when you consider one thing is two, you write a tensor product. Well, how do you know? If we think that we have only one system, then you have four degrees of freedom in it, and then you make, when you consider one system like that as isolated, you make an approximation, because this wavelength, this wave function has some entanglement with that one that you completely neglect. And so this is a source of... Well, the problem is that this approximation gives you the best available physical theory. Yes, yes. Well, sometimes. Of course, not always. I mean, for the interpretation, too. I agree with you. This is a perfectly, well, effective theory. From the point of view of interpretation, the problem is not why we bring these two things, but why we separate them, and why, when we do that, is a very strong operation, when when we, why when we do that this works. Well I agree, I mean this can be also seen as a problem, let's say in loop quantum gravity in the models where you have a fundamental graph, you'd like to say that something of this graph is an elementary particle so you're going to have one graph and then you're going to you're going to sort of try to say that something there either local or maybe maybe some other invariant or something else is a system is a particle so you're dividing this this is becoming a dialogue so Rob you had one thing to say and then maybe we'll catch our chips maybe defer to anybody no it's not too late no it's not too late I wanted to address Michelle's question but there can be ways the world is that prevent happens to get an exhaustive picture of it. And I was going to start saying with Chris, make a distinction about what you might mean

22:30 by an exhaustive picture. So if you think about prisoners in Plato's cave, then they might be presented with various, the shadows of various three-dimensional objects. And so the sense in which if we're always confined to the cave and seeing the shadows, they'll never get an exhaustive picture of reality, they'll never be able to see all the aspects of any a clever prisoner might make a mental model of the way the world has come up with a theory that actually I'm complying to seeing these two-dimensional representations of what are three-dimensional objects. And so in a different sense, that person could have an exhaustive view of reality because they would have, although they don't have the view from nowhere themselves, they can imagine what such a view would be in that modern model. So I think it probably meant the latter sense of exhaustive picture. So my sense is just that in the first sense, I think there are very good reasons to think that we can never, we are like the prisoners in Plato's Cave, so obviously the stuff I was talking about is reasons for quantum theory why you're in that situation. But I think even in the context of Newtonian physics, if you just gave me Newtonian physics, I think there are already good reasons to think that you can never come to have complete knowledge of the world because of action-reaction. if I'm going to have a memory, if my internal state is going to have to respond to the state of something around me, it has to exert a force on me. But by action-reaction, I have to exert a force back on it, so its state comes to depend on what my initial memory state was. And if I don't have a complete accounting of what my initial state was, I can't take into account how it's changed. And so immediately by finding out information, I've changed the world in a way that I don't count that. And so I think classical or quantum, them, there's good reasons to think we're like the prisoners, but I don't see why there's good reasons to think that we can never have an abstract view from nowhere conception of what all the aspects are that we can't come to grips with. But I'm willing to be convinced that at all. I know that some of us are fatigued. Others would probably like to continue, but I suggest but you do it on the porch or where it's a more relaxed setting. We'll wind up with that.