 Great. Hello, I'm JJ Rokin, and welcome to Philosophy and What Matters, where we discuss things that matter from a philosophical point of view. Today, we will travel to the philosophy of possible worlds. Now, according to the standard possible worlds analysis of modal sentences, the sentence, possibly, donkey's talk, is true if and only if there is at least one possible world where donkeys do talk. On the other hand, the sentence necessarily 2 plus 2 equals 4 is true if and only if 2 plus 2 is 4 holds in all possible worlds. Now, this possible world semantics relies on the existence of the domain of possible worlds. But what are possible worlds? Are they the same as our actual world? And how do we know them? Now, to guide us through the philosophy of possible worlds and why it matters, we have Daniel Dolan, Blackman, Hank, Professor of Philosophy at the University of Notre Dame, and the author of topics in the philosophy of possible worlds. Hello, Professor Dolan. Welcome to Philosophy and What Matters. Hi, JJ. Great to be here. OK. Before we start getting into our topic, let's first discuss your philosophical background. By any chance, are you related to the more famous Nolan? Christopher Nolan? Christopher Nolan. Alas, no. My Nolan ancestors came out to Australia in the mid-19th century, and I think his recent Nolan ancestors are British. So I mean, probably back there somewhere, but it's a fairly common Irish surname. OK, so how did you get started into philosophy? Yeah, so I was doing undergraduate work at University of Queensland, and there were an amazing bunch of teachers there that I learned a lot from, and they had a real passion for the subject. And after a while, I realized it didn't have to be a sort of side interest. It could be my job. So I did a PhD at Australian National University and went on from there. OK, so having read your book on David Lewis, I know that you're deeply influenced by him. What was he like as a philosopher? Yeah, I'm really impressive. So just able to say sensible things about all sorts of topics almost straightaway, able to be really clear. Sometimes he would kind of make advances on the literature, and you wouldn't even kind of notice, because it looked like he was just kind of setting things up. And it's only when you go back and look, you'll wonder, why does everybody think about it that way? And you realize, oh yeah, it's because Lewis presented it that way once. Yeah, so he was also really impressively able to get on board with other views to explore them. So he would often, somebody would often present a view. And even if he disagreed with it, he could often kind of say it back to them better than they could say it and kind of, and was sort of, I think, very interested in making sure other people's point of view and systems were as good as possible. Even when he thought something was wrong or misguided or he was committed to something that was conflicted with it, he would really try to make it as good as possible, even when he disagreed with it. And I think that's a great thing to emulate. So aside from Lewis, who else influenced you to pursue a career in professional philosophy? Oh, that's a good question. So as I mentioned, there were a group of philosophers at University of Queensland. And so particularly, Andre Galois was a big influence and grand priest. And in Hengphis and Roger Lamb were very big influences as well. And then when I got to, I knew there were very established figures like Frank Jackson and Philip Pettit and Jack Smart. But also, and Peter Menzies, who was my supervisor when I first got there, but also a group of really interesting younger philosophers were younger then. So people like John Hawthorne and Grave Oppie and Greg Rastall. And there was just a kind of, yeah, I think I learned a lot from, I talk to all of them a lot. I'm really grateful for all the time they poured into a student way back then. And I think I keep noticing in myself ways of doing philosophy or approaches to issues that I think, oh, I probably got them, got that from one of those people. So they were probably the biggest influences that got me into professional philosophy. Right. So the ANU really made a mark in your way of thinking? I think so. Yeah, I kind of, it didn't always feel like it at the time. But once I left ANU and realized how much like the other ANU philosophers I was, I think it must have been a really heavy influence. Right. Because, I mean, just style, but method as well as specific views. I mean, I disagreed with a lot of people about a lot of specific things, but kind of, which is so important and how do you argue for them and what sorts of things do you take seriously? And I think I must have been very influenced. Right. Yeah, and I remember Frank Jackson telling me that you came up with the term Canberra plan. Oh, that's very kind of him, but it wasn't me. I think it was originally Hugh Price and John Hawthorne who came up with the Canberra plan. And they meant it kind of not entirely, in a not entirely flattering way. So I think Canberra's a great city, but it's got a bit of a reputation in Australia for being a very boring city. And it was deliberately laid out as the capital. So I think they meant to kind of suggest that the Canberra plan way of doing things was a little bit artificial. But people associated with Canberra took it on. And yeah, so I think I'm happy to say I'm a Canberra planner. So try to find the things that Canberra planners agree with. We don't think we'll agree on any particular thing, but there are kind of clusters of views. Yeah, because in your website, you mentioned about the Canberra plan manifesto. So what is that all about? Yeah, so there's a kind of narrower and wider uses of it. So the narrower use is a sort of idea that a lot of us, I think got approximately from David Lewis, though people like Frank Francy and Rude of Carnap are also important this idea. Basically that one slogan is, if you want to know what something is, look at what it does. If you want to understand what something is, kind of look at particularly what's definitional about something then look at the theory that talks about that thing and then something's one of those things if it does enough of those things. So in the philosophy of mind, a belief or a desire is something that sits between inputs like perception and outputs like action and interacts with the other things, mental things in the right kinds of ways that produce that. So no matter whether you're running on neurons or silicon chips, it's pain if it produces avoidance behavior and there's a response to damage and all of that kind of thing. But then you can kind of generalize it. Well, what's causation? Well, it's the relationship that typically holds between these kinds of things and these kinds of things. And it can support explanations of a certain sort and it seems to be governed by laws of a certain sort. And you can kind of get a job description for something you're philosophically interested in and then try to work out what, if anything, fits that job description. So that it doesn't necessarily settle very many questions. I think you could even be a Canberra planner and think that a functional story that the mind won't work or something like that. But as a way of approaching, what are we trying to find and what does it have to do to count as the right answer to the question when we found it? I think that's kind of the core. But then in my sort of Canberra planner's credo that I wrote when I was a graduate student there trying to work out what the core was that everybody believed. There were all sorts of other doctrines that most people around there believed, like say, eternalism about time, but the past and the future are real. Doesn't have much to do with the Canberra plan method, but sort of something that a lot of people around there believed. Now I like the way you phrased it. So this one is a kind of method. Canberra planning is a kind of method in doing philosophy. So you have a concept and it has a job description and whatever fits that job description is that thing. Is that the main idea here? Yeah, and I guess the other thing that was popular around Canberra was a lot of the really important concepts in philosophy, the job descriptions are sort of pre-theoretic. So causation or pain or possibility. Once what's that? Well, you look at what our implicit understanding of those things are. So a word like electron, you might go to the physicists and take their theory as kind of definitional of what we're looking for, but for something like grief or being the same table 10 years later, that doesn't, you don't go to a specialist theory, you kind of consult our implicit folk theory of what's going on. Okay, so I like that way you're doing philosophy as well. I think I was influenced by that kind of Canberra planning. Oh, good. By a way of doing philosophy. And now let's proceed to our topic, which is possible worlds. The possible world semantics was primarily a logistic device to analyze the notions of possibility and necessity. But what's the history behind the semantics? Yeah, good. So I think it sort of has two streams. So there's a kind of narrow kind of kind of model building in logic where you, if say for ordinary, say predicate calculus for XFX or something like that, you have a sort of domain and some objects. If you're using a possible world semantics, you'll have domains associated with different worlds and objects falling under different things in those worlds. And one of the most, one of the first things people were using possible world semantics for was, yeah, for a logic of necessity and possibility. And for something to be possible, it didn't have to sort of really be true, but it had to be true at some possible world. And for it to be necessary, it had to be no matter which possible world you went to or no matter which world you went to, that was possible relative to the actual world. That's what happened. But pretty quickly, people started finding other applications and sort of very early on other applications. So Arthur Pryor, for example, used this kind of semantics for his temporal logic. So it's eternally true that P, well, what does that mean? It means at all times P, but the times in the semantics are like worlds. They're kind of snapshots of how everything is. And to say something's eternally true is to generalize over those snapshots. Or Yaku Hintaka very early on was doing, working with logics of obligation and for something to be obligatory, it meant it kind of had to happen or maybe had to happen in the best world or the world where things go as they're supposed to. If something's permissible, that meant it happened in some of the worlds where things go as they're supposed to. And again, even though whether or not everything that's obligatory is genuinely possible is kind of substantive question, but the same kind of techniques for modeling these logics were used. But another strand of it is the idea that we can understand constructions in our language. So uses are possible or necessary, for example, it's being generalizations about these things. And that idea that we could sort of give the meaning of a language that way goes back at least to Khanap who had sort of state descriptions and to say something was necessary was to say that it obtained in all of the state descriptions. And people like Khanap were thinking about this for artificial languages to kind of improve on how we ordinarily talk, but was supposed to have the power of ordinary languages. But then people like Richard Montague and David Lewis and Arapati and many others thought that you could use these kinds of devices for natural languages too. So you could understand the meanings of languages like English or French by modeling those meanings with possible worlds and constructions from possible worlds and things like that. So that also was tied in with the logic early on and a lot of people kind of worked on both. But those two ideas I think went together to build these sorts of theories where you explain something or model it or get some understanding of it by associating different parts of your language with different possible worlds or sets of possible worlds or other things like that. Okay, so I think there's something interesting here. So on the one hand, you have people like Arthur Pryor and those doing deontic logic using the technology of possible worlds to understand some interesting philosophical concepts like obligations, time and so on. And you have another set of people like Khanap who are modeling languages. For Khanap, it is the ideal language of logic for Montague and Lewis. It is the natural language. Okay, but how does this 19-year-old kid, Saul Kripke, come in the picture? Yeah, so this is a kind of really interesting episode in all of this. So people had been thinking about possible worlds and how they might reflect on, might help us with understanding logic and understanding sort of at that stage sort of idealized languages or formal languages. But one thing that was really holding people back was it wasn't clear or at least it wasn't sort of common knowledge what the sort of talking about the possible worlds, how it matched up specifically with the formal systems people were using. And in particular, people were looking for a completeness proof for modal logic. So C.I. Lewis, I think back in the 30s, he developed a series of modal logics and but he didn't have a kind of model theory for them. He didn't have a semantics in the logician sense for them. And so, and it was really hard to tell when different sets of axioms kind of picked up the same class of theorems or when they were sort of varying in various kinds of ways to the point where C.I. Lewis' favorite, so he had these S systems, S1, S2, you probably heard of S4 and S5, which are very popular modal logics still. So logics are possible learning necessity. His favorite logic was S2, modal logic, but he didn't even know whether it was different from S3. And it took many years and a really ugly, big proof to show that S2 and S3 didn't turn out to be the same logic. I mean, you wrote down different axioms for them, S3 had an extra axiom, but to show that that axiom wasn't already entailed by S2 was just, I mean, it took them years and years to work out how to do that. And then in the late 50s, this schoolboy in Omaha, Nebraska sent a paper into the Journal of Symbolic Logic with a completeness proof for S5. And he kind of points out how to extend this completeness proof, sort of a brief note about ways of extending this to other things. And he had a follow-up paper where he extended to a whole different class of logics. And he did that by using the sort of, as you said, the technology of possible worlds, basically, he showed that there were a class of possible world models that if there was a counter example to an S5 inference, you could find a model pretty quickly using this class. And he developed completeness proofs for a range of other systems and kind of gave people this system where you can kind of go from the logic to what kind of conditions on possible worlds have put on and you could notice different strengths of conditions and you could go back to, if you thought possible worlds should obey some extra condition, it was much easier to go back and work out what you actually and you'd have to add or subtract to get that condition. And so his work in the 59 paper, which I think was published when he was 19, but written when he was in high school and 63 paper after that really revolutionized the field. I mean, he wasn't working in a vacuum and Jack Copeland has a really interesting paper, what's it called? It's called, sorry, I wrote this down because I knew I would put the name of it. It's called the Genesis of possible world semantics. And he kind of traces how there were dozens and dozens of people around the world using possible worlds or using things that in hindsight kind of look like possible worlds to explore more logic and to explore other topics. But it was Kripke who nice and crisply published this completeness proof and showed people how to go back and forth between modal logic and possible world semantics in a way that wasn't generally appreciated before Kripke. So, and it was kind of, it was a bolt from the blue, right? A lot of the people who had been working on all this all this life, of course, have never heard of S. Kripke and wondered where the paper had come from. And yeah, so it was a really big splash in the field then. And of course, Kripke went on to do lots of really important philosophy with possible worlds as well. His native necessity book is maybe one of the most influential books in the second half of the 20th century and in analytic philosophy anyway. And a lot of that book is kind of using thinking about possible worlds to argue for conclusions about language but also conclusions about the mind and all sorts of things. Yeah, that's interesting. And I read somewhere that Kripke was invited in Oxford and Arthur Pryor was his host during that time because of his semantics. Okay, let's go to the big philosophical question which is the topic of your book, topics in the philosophy of possible worlds. So what are these possible worlds? Perhaps you could divide the question into a metaphysical question about the nature and existence of possible worlds in the epistemic question about the knowledge of them. Let's start with a metaphysical question. Yeah, that sounds great. And make sure we come back to the epistemic one if I forget. Yeah, so as far as the formal systems go, you don't, I mean, Kripke called them possible worlds and things like that, but you could just take them to be indexes in a model that assign numbers to sentences. It kind of, the formal thing doesn't, but the intuitive understanding behind it and the way I think the reason people got so excited was it felt like this was what motor logic was about. It made it really plausible that saying something was possible that went along with thinking there was at least one possible world where it happened and saying something was necessary was sort of generalization no matter which possible world you go to. It's all gonna be like that, or at least all the ones that are possible from the actual world. Maybe there are, whether something's possibly possible, does that automatically possible? Different logics disagreeing about that. So, but then, and certainly when you're talking about natural language, for example, if you take what we ordinarily say about, say options not taken, let's say I'm making a decision and I, I'm saying I'm choosing between going to a Chinese place and Indian place for lunch and I go to Chinese place and have the lunch, then I, this option I didn't take, we made me a lunch that they didn't end up making because I wasn't there. When we talk that way, we talk as if there are these other possibilities and it's all over the place and the ordinary way we talk, but also science, so probability distributions are often usually distributions over the outcome that did happen plus some other outcomes that could have happened but didn't and had some chance of happening but didn't. And so all the time, once you start looking for it, people are not just talking about situations, the actual situation, but alternative situations or other possible situations. And that's really puzzling because it feels like, no, don't talk about the lunch they made for me at the Indian place. There's no such lunch, right? There's no such thing, right? So there are these two things in tension there. Of course there's this possibility where I went and of course they gave me lunch or they didn't run out. Of course there's no such thing as the lunch they gave me at the Indian race restaurant. So what's going on? And one of the things David Lewis is best known for in philosophy is taking that talk about other possibilities at face value in an unusual way which he thinks alternative possibilities to this one are alternative universes that are much like this one or at least the nearby possibilities are much like this one. So there's an alternative universe where somebody like me goes to an Indian restaurant much like that and has a lunch. And I'm kind of talking about that when I'm talking about the possibility of going having lunch at the Indian restaurant instead. And a lot, I mean, it makes it, if you accept that theory, it makes it easy to understand what this other possibility has to do with Indian restaurants and lunches and things like that. There really is a lunch over there. How come it's possible in some sense that there are mountains of gold? Well, there are golden mountains but they're just not now space and time. And if I say that there's 20 different shapes that the golden mountains can have, that seems true and that means that there must be at least 20 different ones with different shapes out there. And so it makes it easy to see how our language is talking about that stuff because there really are, things really, well, really are out there the way we say things possibly are. But on the other hand, it makes it really puzzling. So how could we be talking as if there were all these other universes how could what's going on in those other universes make any difference to the claims we make whether they're true or false? I mean, maybe they should when we're speculating about cosmology or something like that. But when I'm just talking about what I decided to do for lunch, it doesn't look like I drag in alternate universes full of people like me going to different places. So sort of concrete modal realism is sometimes called where the possible worlds, a possible world where a donkey sings is one that has a singing donkey in it just full stop. That's one approach. Another approach that's probably more common among people who work on this is to think that of other possibilities is somehow abstract objects. So maybe it's a set of propositions that describe me going to the Indian restaurant or maybe it's a property that a whole universe has when it's a universe that has me going to an Indian restaurant in it. And that's a property that the world doesn't have but it could have had and that property's out there and that's kind of the possibility. It could have been instantiated by the world and if it had been things would have been this way or maybe possibilities of their own kind of thing. Maybe about all I can usefully tell you is well, if that one was actualized I would have gone to an Indian restaurant and it's an alternative to the one that was actualized. So maybe I could just tell you stuff about possibility and necessity in it. So that's sort of often called sort of abstractionism about possible worlds. So that's another way of talking which doesn't mean we were talking all along about alternate universes full of that's just like this one except I had something different for lunch. But we're talking about propositions representing that or maybe set abstract sentences representing that or a possibility of that happening where that's just sort of basic. So that's probably maybe the other main view but there are a few other minor views as well. So for example, Minongians who think that there are such things as possible worlds but they don't exist and that we just have to get used to the fact that a lot of what we say requires there to be things that don't exist that have various features and you could be probably most natural if you're a Minongian to agree with Lewis about what's going on there that there really is a talking donkey there and stuff. It's just that it doesn't exist but you could also be a Minongian abstractionist in principle. One approach that I thought about a bit in that book and early on, because I thought it was one of the main rivals to these realist views was a sort of modal fictionalism where we take as a convenient fiction that there are these alternatives and maybe if it's just fiction then they may as well be concrete ones like Lewis thought it's just that they're not real it's not true but we talk that way and it's useful to talk that way like frictional surfaces or ideal gases or something. Getting the details of that to work without collapsing into some kind of abstractionism turns out to be a bit tricky but it's a tempting idea. Yeah, and of course there are some philosophers who think taking possible worlds seriously is a mistake. We have these figures of speech but either we have to be able to find something more kind of less committal to replace them with or to treat people as replacing them with or everyone's just making some kind of big mistake. They act like there are these alternatives and they just start and it would fit our speech really naturally if they were and we talk and do science as if they were but there aren't and that was a mistake. So that's almost always with any interesting philosophical subject matter there's somebody who comes along and says there's no such thing that happens even with the mind and the tables and chairs but it definitely happens with possible worlds as well there are quite a few people who think, no, no, no. We have to do some reconstruction here. Okay, so let's try to outline the procession. So you have the concrete legalists like David Lewis who thinks that these possible worlds are actual worlds concrete worlds independent of each other. Yeah, I mean he wouldn't use the word actual for them because he wants to preserve that for this one but that's just terminology. But yeah, there was just like actual things. Yeah, there are human beings as real as may in them. Yeah, then you have the abstractionists who think that these things are just abstractions that we have perhaps they're just sets of oppositions or sets of sentences but they don't really exist in the concrete sense. And you have those minor positions like model fictionalism which takes possible worlds as just sections and but you're right. It seems to collapse to a kind of abstractionism. Right, and you have- Yeah, I mean, once you think there are fictions out there that are that rich. Yeah, you're right. That looks like the sort of propositions that the abstractionists was trying to talk about. So then you also have the mind-onions who think that these possible worlds don't exist but they subsist perhaps. And you have of course those nihilists that don't think the possible worlds are anything. Okay, so we have the positions. What's your position? What's your preferred view? Yeah, so I knew you were gonna ask me this so I've been thinking about it. I think I'm some kind of abstractionist. So I think I'm definitely willing to sign up to abstractionism but a natural question there is, well, what kind of abstract object? Let's talk about these abstract objects. And there I am not sure I'm fully opinionated. So I guess I think they are abstract objects and we managed to talk about them as these things that could have been actualized. It could have been this one rather than the one that is associated in the special way with the concrete going on. Maybe they're maximal propositions that kind of settle every issue or maybe they're super generous things that we can't say something else about or maybe there's something like these properties. I guess I think one reason that I'm not completely opinionated is I think it's a really tough question to tell when we're talking about abstract objects with different terminology and different vocabularies when we're talking about the same group abstract objects. I mean, this comes up in philosophy of mathematics as the other is the number one on the real number line the same as the number one on the counting number line. Natural to say yes, but they're governed by different axioms, so maybe no, but outside mathematics, it's even harder. So are these two set theories talk? Well, I guess that's mathematics again. Are they talking about the same sets and disagreeing about them or are they just talking about their own sets? But if this person's talking about maximal properties and this person's talking about maximal propositions, there's a way of modeling what one of them's saying with the other ontology. But are they talking about different things or are they talking about the same group of abstract objects in a different way? Or maybe there are multiple abstract objects out there that can kind of do the job sort of camera plan style and maybe we haven't said enough to make one of them the unique ones that do the job. So it could even be that it's to some extent indeterminate exactly what the abstract objects that we're talking about what we're talking about possible worlds which ones they are. So that was a long way of saying, and if you ask me a further question about which abstract objects there are, I kind of know. Okay, so I was intrigued because he did not follow David Lewis in his modal realism. What's your objection against modal realism? Yeah. So David Lewis once kind of said that the objection he didn't know had a few was the incredulous stare when people would just stare at him when he said that the possible worlds for these things and this is what they were like. I don't think it's just the incredulous stare though. Like a lot of people ever since I've heard this view I kind of think that can't be right. That's not what we were talking about. That's not what's required. So it's possible that I knocked something off the table. Does that require an entire cosmos with an alternate me and an alternate table? It just doesn't seem like... Yeah, so, but that's not an argument. So it's just an argument. Maybe it's an argument, but it's just an argument. I guess I think it's really extravagant. Basically you have to postulate every single possible kind of thing to have postulated enough things to make his theory work. And I started off not believing in flying mountains and singing castles. And I still don't believe in them and I'd need a good reason to start believing in them. But I'd have to be flying castles and singing mountains and dragons and every possible kind of thing if Lewis was right. And so I guess I think it's unpassimmonious is one problem. And it's about as unpassimmonious as any possible theory could be. So I think that's one of the reasons why it starts off. Yeah, so I'm not saying I couldn't be dragged to it, but it feels like it's got a much bigger burden to clear than the other ones. Sorry, go on. Yeah, so I'm hearing a kind of opum breacher opum breacher argument here. Presummonious argument. I think it was an answer to that, right? Yeah, yeah. So he's got, he doesn't say a lot about it. And I, but the short thing he does say, I kind of interpret as talking about two things. So when he says, well, look, we shouldn't care about numbers of objects, some theories have a lot of objects. We should only care about kinds of things. And I'm just giving you more of the same kind of thing. You already believe in concrete objects in space and time like you find in this cosmos. I'm just giving you more of that. Boy, you already believe in this cosmos. I'm just giving more of that kind of thing. And I guess I think that's which kinds you count matter, right? So if it's actually enough, then nobody has to worry about extra kinds of things because they can say, oh no, I'm just postulating more of what I already had, existing objects or objects or things that fall under the quantifier. So I think it's got to be, it's got to be more restrictive than just more kinds of things where the kind is object defined in this universe. And if you go more restrictive, then it's not the, it's not the extra space times that so much of a worry on the kind's front as the, well, no, I mean, biologists told us there weren't any flying dragons and you tell me there are. So that's a new kind of thing that it didn't look like biology told me about, but I have to believe in it if I'm gonna believe you. But there's also a second thing, which I think is more controversial, but I've taken a stand on it, so I'll tell you about it. He also, when he says the numbers don't matter, I guess I don't think that's true. I think maybe total number of things you believe in might not matter because anyone who believes in set theories already up for all sorts of levels of infinite sizes and things, but within each kind you postulate, I think numbers matter. So the detective who thinks that somebody in the organization has been defrauding it, that's a lot more sane a hypothesis or a lot more sensible hypothesis than the detective who thinks every single person in the company is defrauding it, right? So if the first story works, you wouldn't immediately believe the second one, right? You'd wanna see some extra evidence that they were all in it together or something like that. And so I think if you don't just look at what kinds of things you postulate, but the numbers of things of each kind, I mean, Lewis said you shouldn't and that doesn't matter, but I think it does matter and then he's got huge infinite numbers of, not just, yeah, infinitely many people rather than however many billion there are. Now, however many trillion there'll be over the history of the universe. I like the analogy about the detective looking for a suspect. Now, you don't think about every person in the organization as the whole, right? So it's just at least one, that's a good number to start. Yeah, and if the detective says, well, all of us think somebody was defrauding it. So you can't complain if I think everybody was. No, no, I wanna say more before, yeah. Yeah, so on to the next question about epistemology, the epistemic question. So how do we know these possible worlds? That's a really good question and I think it's a tough question to answer. I guess I think it splits into two questions. One is how do we know what's possible and then how do we know that whenever we think something's possible we should think there's a possible world associated with that possibility if you like or associated with that being possible. And so the first question, philosophers are often interested in possibility in everyday senses. I think we have a whole range of different techniques for telling that. Like how do I know I can get the lid off this bottle? Well, if it's the same kind of as many other bottles that I have taken the lid off and some of it might even be sort of pre-theoretic. So psychologists say a lot about affordances and we pick up whether or not something's in reach that is whether or not I can reach out and grab it. I often don't need to even come up with a hypothesis or reason about it or something. I can just look and I know because I'm used to being in this body and I'm used to having my arms. Obviously if somebody broke my arms or I shrunk or grew or something I might not be able to do that anymore. But the kind of possibility that philosophers are really interested in is maybe metaphysical possibility or logical possibility or possibility in the most generous sense if there is a most generous sense or at least very, very generous senses. So I mean, philosophers are also interested in the kind of necessity that laws of nature have maybe nothing can accelerate through the speed of light that kind of necessity. But a lot of philosophers even think that's kind of contingent relative to logical necessity or something like that. It's nothing incoherent about an alternate set of laws of nature or something like that. And finding out about that I think we probably find out about it in a similar way to the way we find out logical truths or mathematical truths. Big question how we do that. For example, one way we do I think is through our conceptual competence. So why am I so sure that I'm never gonna come across a round square? Well, I know what squares are, I know what round is. That already puts me in a position to know that I won't come across any and also that there can't be any sort of unpack it that's a contradiction in terms or something like that. How do I know contradictions are wrong? Well, I don't know some kind of which inferences seem good and then generalizing to that and checking around to see whether we come up with any kind of examples. I mean, epistemology of logic is really hard too. But I think when we do know logical principle mathematical principle like I as well. It can't be that roses are red and violets are blue but it's not the case that roses are red but that combination can't happen. But I think to know that maybe I need to know a little bit about flowers but I mostly just need to know about ant and have a good concept of ant to know that. Okay, so that, I mean, that's easy bit. That's a really hard bit. But then suppose I find out that it's necessary that whenever a conjunction is true, one of its conjunctions is true. How do I go from that to generalizing about all sorts of possibilities? It's like every possibility is like that. Or suppose I find out that it's possible that one person can split into two people, philosophers argue about whether that can happen or at least whether that can happen and both of them sort of have been the person who started as opposed to there being new people who sort of emerged from the first person disappearing. How do I go from that to there being some kind of object or possibility that has according to it people splitting or something like that? I think that needs something like inference of the best explanation or something like that. So I guess I quieted, Will Venom, Quine and Hilary Putnam talk about sort of indispensability arguments or sort of if you've got a theory that works really well and you keep testing it and it keeps working and you see how all the bits are needed to get what the theory is telling you about the world, the better it does at the bits you can test, the more credit the whole system gets. And if you can pull it apart or find another system that does it a different way that might discredit the system you started with, I think that's probably what's going on in a lot of our very theoretical enterprises and I think something like that's probably what's going on with possible worlds as well. We find this going back and forth between what's possible, what's necessary and talk about possibilities or counting possibilities or comparing possibilities. That's all kind of mixed up together and that system, I don't think philosophers understand the whole system yet but when we do, we'll be able to see how that system is helping us get it right and get us to understand the modal things that we can kind of test or I don't mean test by sort of taking a modal cane and directly doing science on it but see how it fits in with our other things and see whether we go wrong by relying on it and that sort of thing. And then fingers crossed, that'll all work out and that'll tell us, oh well, we do need to think there are these possibilities as well as just, so it's not just possible that there'd be a mountain of gold but there's a possibility according to which there's a mountain of gold and if Lewis is right, that'll be a cosmos with a mountain of gold in it. If I'm right, it'll be some kind of abstract object that has some kind of other connection to being a mountain to mountains of gold or being a mountain of gold or something like that. So that's a very, well, I was literally waving my hands, it was a very good answer but. Right, yeah but I like the way you cash it out in terms of logic. So you have a conjunction rule. So you have a conjunction, you get the conjuncts, right? So that's a necessity but yeah, the question is right. How do we know that? How do we know that kind of principle? So asking that question relates us to questions about possibilities as well, right? So how do we know those possibilities? It's not possible that you have a conjunction that's true and one conjunct is false, right? So I think, but do you have a position here in this kind of question, this epistemic question? Yes, so I mean, I don't have a detailed theory worked out and I really hope one day I have a more detailed theory worked out. I guess I think the, so I had fairly recent paper called naturalized modal epistemology. I think the way we answer these questions isn't radically different from the way we answer other very theoretical questions. So some philosophers have thought we need a special way of knowing modal things. Maybe we need a special exercise of our rational faculties or something or maybe a special- We don't have epistemic access, special privilege access for these things. Yeah, yeah. Or we have a sort of connection to the friggin realm of concepts and their relationships or something like that. And I guess I think not. So I think the main, maybe I shouldn't say the main, the most tempting argument for me to postulate some kind of special way we do it is the sort of argument by exhaustion. Look at all the ways we find out other things. None of that seems to help. So some people sort of start with this idea, well look science or our senses or something can tell us what's in fact going on, but it can't tell us what has to go on or what could go on. So we have to find out some other way. And I guess I think it can, in various ways, it can tell us various things. I mean, sometimes fairly boring things like very possible things like the feasible things like my being able to reach out and grab the cup. That's, I find that out using a, I don't think I need a special rational faculty to tell me that. I think I could, the sort of psychology of affordances will give a pretty good story about how we reliably and systematically do that. And then various theories seem to tell us not just what's in fact happening, but sort of a range of theoretically open options where that doesn't look just epistemic, but it looks like when I, when I learn about the properties of a metal, I don't just learn how that metal is in fact arranged, but I learned various ways it could be arranged. So could you bend it in this unusual way that nobody's bent a piece of metal before? Well, if you're familiar with the metal, a lot of metallurgists will say, yes. We can obviously build you something like that if you want. Nobody's ever wanted that shape before, but we can go do it. And I think they know that modal thing about the metal before they've ever made it into that particular shape. And that we can sort of do more of, here's the hand wave, we can do more of that sort of thing and eventually get ourselves to a position where we have a systematic understanding of modality. So, and I mean, I mostly argue for that by trying to point it all the ways that we seem to have access to facts about what's possible, what's necessary. And that's sort of a proof that we can do it all and we don't need some special faculty, but it starts to suggest that we don't really need a special faculty to be able to do this stuff. Yeah, you mentioned about- But it's very much controversial still whether we do need a special faculty to do this. Yeah, you mentioned about coin and potnums and dispensability argument. I'm seeing the influence here. But the one intriguing part of your book, idea in your book is about impossible world. So what are these impossible worlds and how do they fit in the story so far? Good, yeah. So when, one reason there was a lot of attention paid to possible worlds was all the philosophical jobs that looked like they could do. For example, one that people were thinking about from very early on was as a picture of, for example, mental content. What's going on when I believe that cheese on toast tastes good? Well, that belief is associated with a set of possibilities where cheese on toast tastes good. And if it's a set of possible worlds, it's ones that might disagree in all kinds of other ways, but all agree that cheese on toast tastes good. And so modeling beliefs and desires for sets of possible worlds was a tempting application and helps explain how beliefs could be about things that aren't actual because they could be about things in other possible worlds or however you understand that. But for example, if you're doing that, then you also want beliefs and desires that are impossible. You can make a mistake and approve and think that something's a theorem when it's not or something is a counter example when it's not. And that's a kind of matter of logic and you can be wrong about that. Mathematicians or, and also there are necessary truths, you can believe some necessary truths without believing them all, right? Nearly all mathematicians believe that two plus two equals four, but that doesn't mean they automatically know every mathematical truth. It would be wonderful if math was that easy, but it's not. And so modeling people with not just possible worlds, but with impossible worlds as well, we can find distinct beliefs to have, right? Like the person who thinks something is impossible, let's say they think there are six regular platonic solids in Euclidean space. I think there are only five. I might be an example of somebody who's wrong if I got that number wrong, but if they think it's six, yeah, there are worlds where there are six Euclidean regular Pythagorean solids in their set of beliefs, but other impossibilities aren't in their set of beliefs so that their parents turn into numbers when they weren't looking, that's also impossible, but that's not, they don't believe that, whereas somebody else with a certain kind of delusion might believe that, but might not have a belief about various mathematical things. So one application I was initially interested in is one thing possible worlds seem to do is give us a way of understanding counterfactual conditionals. So if I'd dropped the cup, it would have shattered on the floor. That doesn't look like it's about, just about what in fact happened with the cup and the floor because let's say I didn't drop it. It doesn't look like it's saying that it's necessary that the cup gets dropped and shatters on the floor. It looks like it's logically coherent. So I have a model where the cup falls down and lands on the floor and doesn't break. It looks like it's saying something like, in possibilities, much like this one, except that the cup drops, it breaks. And David Lewis and Robert Stirlnack had developed kind of theories of counterfactuals, kind of using this idea of nearby possibilities and for a counterfactual to be true, it had to be the nearby possibilities where the condition held, the outcome held, but as the lingo has it, where the antecedent is true, the consequence is true as well. But as well as, so, and we use counterfactuals all over the place, including in science and history and so on, but as well as counterfactuals with possible antecedents, we look like we talk about them within possible antecedents. So a historian might be interested in how things would have been different if Napoleon had conquered Europe or what would have happened if female emancipation had taken even longer in the United States. But we might be interested for, say, purposes of entertainment. What would happen if characters of their own free will could climb out of stories and climb into other ones? And you can tell stories or should have movies where that happens. And if you're assessing how plausible the movie is, you might think, well, what would they do? And did they, did the author do that or did they do something else? Or rival logical systems? So for example, I don't think any contradictions are true, but I'm really interested in the question of what would the solutions to various philosophical puzzles be if there were true contradictions? Because maybe they could help with vagueness or maybe they could help with the lie paradox or maybe. And so, and likewise people who do think there really are true contradictions, they probably wonder whether if things were logically as I think they are, what would happen? And they might be interested, they might think that if I were right about there being no true contradictions, then there wouldn't be a good solution to such and such a problem or something like that. And they might even argue for that. So it's not just sort of entertainment, it's sort of philosophy logic in that case. It's not saying that's not entertaining too. Yeah, so, and you can extend the possible world story, right? All the nearby possibilities or the possibilities much like this one, a certain thing happens, you might extend it to impossibilities as well, right? So the impossibilities where characters climb out of stories and climb into other ones but things are otherwise just as they are. That's a impossibility where for example, I'm really surprised and people might be more scared of keeping fiction books in their kids' bedrooms if that can happen, right? So those, but other impossible things, they wouldn't happen if fictional characters could do that, right? It wouldn't turn us all into numbers or something, right? Then that's also impossible. And I think there's even an impossibility where both things happen, where fictional characters are climbing out of things and we're all turned into numbers. But it's not a relevant impossibility to the question of what would happen just with the fictional characters or impossibilities where we all start flying through the air, those aren't the nearest impossibilities where true contradictions provide a satisfactory solution to the lie paradox, right? Then sort of the flying through the air stuff is sort of arbitrary, sort of cracking the lie paradox won't give us an eight flight or anything like that, right? So yeah, so just sort of whatever we were doing with possible worlds, we can often do that for more cases if we've got impossible worlds as well. Yeah, it seems like impossibilities for you are fine grained. So there are different kinds of impossibilities and sometimes they're related, sometimes they're not related. So we need a kind of machinery to think about these things. Is that your whole point here? Yeah, yeah, that's right. And I think the, I think I wanna sort of piggyback on the success of possible worlds. The things people have managed to do with possible worlds. I think you can often do those things a bit better with impossible worlds. So modeling fictions with possible worlds, well, you probably wanna model it with some impossible worlds as well or model belief with possible worlds, you probably want some impossible worlds as well. So in your semantics and whatever semantics this will be, it will include two sets of worlds. You have possible worlds and impossible worlds. So how do they interact in this kind of picture? Yeah, so I'd like a view where, for example, metaphysically, they're the same kind of thing. So maybe, I think that would be a bit harder if you were Lewis because- You're right, right. Well, some people have gone that way and said, yeah, just like there are golden mountains out there, there are round squares out there too. But that's, I find that even harder to believe. But for example, something that were it to be the special one, there would be golden mountains. I think that's a possibility where there are golden mountains according to which there are golden mountains. There are also impossibilities. So if that was the special one, we would all be numbers. I mean, it can't be the special one, right? It's not the special one and it can't be, but it's the same kind of thing as the possible one. At least I'd like that story. Some people think it's worth having split stories, say maybe concrete alternative possible worlds and abstract impossible worlds and some other kind of combination. But I kind of think, I don't know how much of this was because I was an abstractionist to start from, for example. If you're tempted by the idea of worlds as big propositions, yeah, as well as possible propositions, they're impossible propositions, you know. It works, right. Yeah, so and how they interact, I guess I think we've got, so for example, suppose you're modeling them with sets of propositions. I don't think that's a perfect model, but it's a pretty good model. Then you could say, well, look, the consistent complete sets of propositions, those are possible worlds or those are associated with the possible worlds or they model the possible worlds. The other sets, they're the impossible worlds. So in particular, the complete ones that answer every question but are inconsistent or incoherent, a bunch of them are the impossible ones. And I think also there are impossible worlds, things that play the kind of role of possible worlds that are incomplete as well, but just if they were real, it just wouldn't be answers to some questions that there are yes or no answers to and a kind of life excluded middle. So I think propositions either they're true or they're false, but were this impossible world, the way things really were, could have middle would fail, right? There just wouldn't be answers to those questions. So I think even the incomplete ones, you could associate those with the impossible worlds too. But I'm tempted to think of it as there's these general things, worlds, and the special ones are kind of well-behaved enough to be the possible worlds. And then there are all the ill-behaved ones as well. Okay, so I'm thinking that the underlying logic in this kind of semantics will be classical since you're retaining excluded middle. Yeah, I don't think it needs to be for this, right? I happen to like classical logic as far as it goes. And so I happen to think the possible worlds are classical, at least on most days, sometimes I'm a bit non-classical. But I think even if you're a non-classical logician, you should probably think there are the well-behaved ones that obey whatever the correct logical principles are, and then there are the other ones. And I think we should be pretty generous about what kinds of impossibilities we admit. So we should think pretty much no matter which logical system you like, there'll be impossibilities where it doesn't work that way. No, right. That's okay. They don't have to compromise their logic by admitting that it's impossible that their logic fail. Okay, so on a more personal note, you've been in the field in philosophy for so long now. Oh, a long time now. Yeah, so what's your advice for those who want to get into academic philosophy? Yeah, so assuming that everything goes more or less back to normal after the pandemic, because as you know, a lot of the job markets kind of stopped right now and it's really hard to know what the future's gonna be like. But I'm optimistic. I think once the pandemic ends and the global economy will take off again and universities will lead people to teach philosophy again and that sort of thing, or new people to teach philosophy. So assuming everything goes back to normal. Yeah, so I think getting into a well-regarded philosophy program and of course, depending on which system you're in and what country you're in, that can vary quite a bit. One thing that's worth looking at, I think no matter what system you're in is, look at the placement record of PhD granting programs and see whether the people who get PhDs from there tend to go on to academic philosophy jobs or not. I mean, there are lots of great things you can do with a PhD other than academic philosophy. But if you're interested in academic philosophy career, which is kind of what your question was about, then yeah, there are some places where something like two thirds of the people who finish PhDs from there end up with continuing academic positions. But there are some quite good programs where it's only like one third or even lower. And again, the people who started five years ago or seven years ago, people's experience starting now will be different because faculty will be different, things like that. But it's a guide. So get into a program and work with somebody who is well-regarded in the area that you're interested in. But for people in PhD programs, what can they do if they want to maximize their chance at a career and a good career? I think there's a few different things they can. I think one important thing is to see the philosophical world outside your own graduate program. So go to some conferences, maybe visit somewhere else for a semester or something like that because the way philosophy is done at different places, even within the same tradition, can be really different. People can take different things for granted. And I think if your first exposure to that is when you get your PhD and try to get a job, that's not ideal. You don't wanna realize that something you think is obvious, everybody else thinks is crazy in a job interview. You wanna be able to come in and be like, I know where you're coming from, I see why you say that, here's why I find that thing or here's the thing that I think needs to be resolved to make progress on that issue. I think also maintain broad philosophical interests. I mean, there's an inevitable narrowing as you get more specialized, but I think again, to be able to talk to philosophers outside your very specific subfield, I think is really valuable, both prudentially for being able to get on with people and convince them you should have a job and stuff, but also just so much of progress in any little topic often comes from things people have been thinking about on other topics and getting a sense of how it's connected and seeing, drawing connections that other people haven't drawn yet and things like that. I mean, we're really lucky to be living in a time where there are thousands or even tens of thousands of people doing roughly the same kind of things we are, but not sort of stuck in a medieval monastery a month's travel for an expert interested in these things. So we really should take advantage of that. And I think it stops people going stale too, if they're aware of a whole bunch of different things going on and find different things interesting. So that kind of be part of the philosophical world, I think is an important thing for people, especially once they're in PhD programs. I guess also for people in PhD programs, start writing things that could be publications relatively early. So I think, I mean, I publishing, I think it's getting more and more important, especially say in the US, but also a writing sample that shows you can do that already, almost every academic job or at least every academic job at universities, there's some publication requirement at some point and people want to see that you were already thinking in those terms. So I think when I was coming out, I think a lot of people still used just an unchanged chapter from the middle of their dissertation as their writing sample. And even then it was probably not a good idea, but these days I think people will, if you do that, they'll be like, okay, this person can write a dissertation chapter, but can they write something that they can publish? And yeah, you don't need to leave that question marking people's minds when you're trying to start an academic career. So I think that sort of thing, I mean, again, that's something advice almost everybody gets, but start thinking about how to learn the skills of how to be able to write something to be published at grad school and start thinking about that fairly early. No, is a career in philosophy worth it? Would you say that your career is worth it? I'm very glad that I did it. I think it, but I'm very lucky. I've got a continuing position and I've done very well out of it. I think it's, I mean, yeah. I think if you can get a job like mine, it's very, if you've got preferences like mine, it's very worth it. You get to spend a lot of your time thinking about really interesting things and teaching students about stuff that I still find really interesting and talking about them with interesting, smart people and yeah, spending my time learning about stuff that I'm interested in learning. Of course, different people want different things, but I think it's great if you can do this. I guess the one question a lot of people starting out need to think about is the probability piece, right? So I think something like a lot of philosophy PhD programs, something like three quarters of the people start, finish, but then depending on where you are, it could be two thirds of those people end up with continuing jobs eventually. So initially you might have to take a postdoc or a temporary teaching or something like that. But at other places, it's only one third or something like that. So three quarters times one third is getting to be a relatively small number, what, one quarter. And so I think everybody starting philosophy programs these days would be well advised to have alternatives besides academic philosophy in mind. I mean, people fall on their feet too if they don't, if they have an alternative and it doesn't work or they haven't thought about it and they need to. But it's, a lot of people kind of near the end of their PhD program say I wish I'd thought about this earlier or I wish people had told me to think about it earlier. And of course there's a lot to think about at the start of philosophy PhD program too. So there's not time to kind of work on every contingency. But I think it's, I mean, yeah, do a philosophy PhD if that's something you'd like to do intrinsically and depending on where you are, you might have a pretty good job chance at an eventual permanent position. And then, yeah, it's definitely worth it for me. Different people want different things. So people who are really reluctant to move where they're living or where they grew up, where they're in graduate school, that's a very difficult wish to satisfy in the US or Australia or the UK. I mean, the UK, it's a little bit easier because you get a job somewhere else and say you're in England and you're somewhere else in England, you're not too far away from where you started. But yeah, so things like being aware of the kinds of things that you might have to give up like choosing where you live and things like that. But yeah, and the probabilities are tough and they'll probably keep getting tougher. So people starting, even once the pandemic starts lifting, a lot of people will start PhDs and either won't finish or will finish and not get an academic, continuing academic position. And yeah, whether it's worth taking that gamble, I think depends on how much that intrinsically enjoy doing the PhD and how much that value doing it, even if a permanent academic position didn't work out. And also how happy they'd be in a variety of permanent academic positions. So a lot of people end up teaching it, primarily teaching colleges in the US. And a lot of people I know who teach in those colleges love that, they love teaching the students, they love thinking about it, they love not having the kind of pump articles about pressure that people at more research institutions have, but other people would hate it. Other people would feel like they were in exile. And so people have to kind of, some of it you have to sort of do it to discover it. But I think having a, I think when you think about all of those things, I think it's, it will still make a lot of sense for people to do philosophy PhDs and try to get permanent academic jobs. And if you can get a good one or depending on what's good for you, it's likely to be very rewarding and very satisfying. So thanks again, Professor Nolan for sharing your time with us. And for you guys, join me again for another episode of philosophy and what matters where we discuss things that matter from a philosophical point of view. See ya.