Everybody talks about the truth, but nobody does anything about it. And to be honest, how we talk about truth — what it is, and how to get there — can be a little sloppy at times. Philosophy to the rescue! I had a very ambitious conversation with Liam Kofi Bright, starting with what we mean by “truth” (correspondence, coherence, pragmatist, and deflationary approaches), and then getting into the nitty-gritty of how we actually discover it. There’s a lot to think about once we take a hard look at how science gets done, how discoveries are communicated, and what different kinds of participants can bring to the table.
Support Mindscape on Patreon.
Liam Kofi Bright received his Ph.D. in Logic, Computation and Methodology from Carnegie Mellon University. He is currently on the faculty of the London School of Economics in the Department of Philosophy, Logic, and the Scientific Method. He has worked on questions concerning peer review and fraud in scientific communities, intersectionality, logical empiricism, and Africana philosophy. He is well-known on Twitter as the Last Positivist.
- Web site
- PhilPeople profile
- The Sooty Empiric Blog
- Paper on “Is Peer Review a Good Idea?”
- Talk on Why Do Scientists Lie?
[accordion clicktoclose=”true”][accordion-item tag=”p” state=closed title=”Click to Show Episode Transcript”]Click above to close.
0:00:00 Sean Carroll: Hello, everyone, welcome to the Mindscape podcast, I’m your host, Sean Carroll. And as you know, we are not afraid here at Mindscape about getting deep into things and today, we’re going to get deep into some of the most profound ideas in philosophy. We’ve talked about philosophy before. We’ve talked to philosophers about pretty darn deep ideas, about realism, about consciousness, about physics. Today, we’re going to talk about truth, and truth is one of those things you think, “Well, maybe it’s a pretty straightforward idea. Some things are true. Some things, not true.” Anyone marginally acquainted with modern philosophy should know better than that; of course, things are going to get really, really complicated.
0:00:40 SC: The analogy that I have in my brain is Gödel’s theorem in mathematics. Kurt Gödel proved that if you have a sufficiently strong formal system, a system where you can prove theorems and so forth, using logic and math, and deduction, then there will always be one of two choices: Either there are true statements in that system that you can never prove, or the system itself is somehow incomplete, somehow internally incoherent. It’s not consistent. And this is a surprising result, right? You might think, people did think you could prove everything that was true in a formal system, not true. And in fact, that lesson generalizes to other ideas that we might have about truth. It’s very, very hard to be formal and rigorous and careful about what you mean by the word truth.
0:01:28 SC: So today, we’re going to talk to an expert. Liam Kofi Bright is a Professor of Philosophy in the Department of Philosophy, Logic and Scientific Method at London School of Economics. If you want to know why the department’s called that, Karl Popper was an influential person back at the London School of Economics back in the day. So, things Karl Popper was interested in became an entire department. Philosophy, logic and the scientific method, we’re going to talk about all of those things today. Liam is an expert not only on the formal aspects of what truth is. There’s a lot of symbol manipulation and proving things, and being very, very careful and rigorous that goes into this kind of game, but also the down and dirty. How do we figure out what is true? So not only what is true, but how do we figure it out, the game of epistemology. So we can go from this crystal clear set of theorems and axioms and so forth, to really difficult, real world questions. When you say, “How do you figure out what is true?” you might say, “Well, I used a scientific method.” Okay, well, what’s that, right?
0:02:31 SC: What not only do we say the scientific method is, but how does it actually play out in the real world? How do we decide who gets published in journals? What is the role of refereeing? How do different kinds of scientists bring different perspectives to the questions of what is true and what is false? Even if we all think that the universe is real, and it’s out there, and what is true about it is going to be objective for everyone, the ways that we get to that truth might depend a lot on who we are and what we bring to the table. So there’s a lot going on here. It’s a wonderful fun conversation. This is the last hold-over from what I got to do when I was visiting the UK a little while ago. Liam was nice enough to let me delay this while other podcasts got into the queue, so I think we’re going to have fun here. It’s a little brain teaser, let’s go.
[music]
0:03:34 SC: Liam Kofi Bright, welcome to the Mindscape Podcast.
0:03:36 Liam Kofi Bright: Hi, Sean.
0:03:36 SC: I do want to mention for the listeners that we’re in swinging London at a university. So there is construction going on outside.
0:03:45 LB: Yeah, sorry about that, listeners. [laughter]
0:03:47 SC: That’s okay. The listeners are a flexible bunch by now. They’ve learned to accept things. I thought that we would talk about the nature of truth and the nature of knowledge and the nature of science. And then with all the extra time we had left over, we can think about things to say.
0:04:04 LB: Well, but once you settle knowledge, truth and science, we’ll still have about hour left, right?
0:04:09 SC: Yeah, I know. We’ll have to fill in the time, tell jokes.
0:04:11 LB: Okay.
0:04:12 SC: So truth. Let me… Maybe a good way to get this is to sort of say, look, there’s a way that scientists think about what they’re doing. They come with a model. Maybe there’s words like particles and forces and fields involved, and if the model works, if it fits the data, they will say, “Look, I have found some truth. The truth is reflected in the relationship of my model to reality.” So, what do sophisticated grown-up philosophers think about that picture of truth?
0:04:40 LB: Sophisticated grown-up… You’re asking me what the null class thinks? [laughter] So there’s a kind of interesting feature of the way philosophers think about truth, which is as far as I can tell, it diverges from how everyone else in the world thinks about truth, and yet philosophers tend to think of themselves as only explaining the common-sense view. For what it’s worth, my experience of talking with scientists has actually been very different. The scientists tend to be very averse to describing the results of their inquiry as truth, and partly because I think they’re working in the background with a more loaded vision of what truth is or would have to be than philosophers tend to be.
0:05:17 LB: Let me explain what I mean. So philosophical thinking about truth as it’s done nowadays, there’s a sort of classic dialectic, which, for instance, we teach undergrad students. I teach a first-year course, and this would be how I would introduce them to it, where there was in the early 20th century, late 19th century a debate between these kind of stylized positions which we nowadays remember as the correspondence theory of truth, the pragmatist theory of truth, and the coherentist theory of truth. According to the correspondence theory of truth, truth consists in a relationship of correspondence between judgments of some sort, be they utterances or written sentences or something like that, and the facts, where the facts are these metaphysical composite entities out there. So, for instance, if I, say, one of the standard examples, the cat is on the mat, then that is true, just in case there is this thing, a fact out there, which is this structured complex of a cat, a mat, the relationship on that, and the cat actually being the thing which is bearing that relationship to the mat. And that’s the correspondence theory of truth.
0:06:25 SC: Yeah, this seems like the not crazy theory so far.
0:06:29 LB: I think that would probably… That’s the view which especially philosophers tend to think of as the correspondence theory, as the common-sense theory of truth, rather. The pragmatist theory of truth… And what I’m going to do is, I’m going to go through them. Then I’ll go back into what are the problems. The pragmatist theory of truth has many different iterations. There’s not one canonical formulation. The sort of famous slogan is, “Truth is what’s good by way of belief,” and there the idea is something like true beliefs are those which one expects to work for you in the long run, or something like that, where belief has a certain purpose, which is helping us orient ourselves and act successfully in the world. True beliefs are those which consistently and systematically help us achieve those purposes. They allow belief to fulfill its function successfully.
0:07:17 SC: The cash value of ideas was the slogan that I was talking about.
0:07:20 LB: Yeah, there you go, right? It’s the same thing. It’s the things which, in the end, how does this belief play out in my life, in my actions?
0:07:27 SC: What good is it for me?
0:07:27 LB: What good is it for me? In a scientific context, that will often be seen as sort of linked to empiricism, where the idea is that true beliefs are those which sort of bear out experimentally, which when I try and test them, they don’t end up being wrong. And so that would be the pragmatist theory of truth. And the coherentist theory of truth is a bit different from both of those, where the coherentist theory more or less has it that a true belief is a belief which sort of fits into the maximally coherent theory of the world. So imagine you had a sort of fully worked out theory of everything which was able to accommodate all of your observations, sort of everything you come across. A belief is true just in case it would fit into that sort of very, very complex, no doubt will always be hypothetical, but hypothetical totality of beliefs. So I said this is sort of classic dialectical, where these are stylized positions, because more or less all of those have now been abandoned. You wouldn’t find… You very rarely…
0:08:24 SC: Oh, no. I was just going to say that I’ve gone through believing in all those different theories, and I’ve landed at the coherence theory more than anything else. So now you’re going to tell me that I’ve spent my life being wrong in different ways.
0:08:36 LB: Yeah, and that’s just everything. You’re wrong.
[laughter]
0:08:39 SC: Yeah, get used to it.
0:08:40 LB: You Sean, and you listeners are wrong about everything. And me, I guess. Except for the claim that you were wrong about everything. I was right about that. Okay, so the issues… Of course, you’ll find sort of people who have more sophisticated versions of the initial formulations. And so nothing ever dies. It just gets reborn and rebuilt. But there are sort of big problems with each of these which make people a bit wary of assuming any of them. I’ll go in reverse order. So, for the coherentist theory, typically the worry is just there’s no way of specifying a unique best theory under that.
0:09:14 LB: The problem is that multiple very consistent, very coherent, from what we can tell, theories can be formulated which would allow contradictory beliefs in, and now I have no fact of the matter as to which of those… Well, I either have to endorse true contradictions or I have to find some way of deciding between the maximally coherent world views I can formulate. And that doesn’t seem quite right, because the natural way of deciding between a very, very consistent fairytale and a very, very accurate description of the world, is that one of them’s true, and one of them isn’t, but that’s now being denied to me.
0:09:49 SC: The reason why it took me so long to even entertain the coherence theory is that it seems obvious to me that there’s different coherent theories that are completely different. And what sort of broke the tie for me was thinking that, well, one of the things that your theory needs to be coherent with are your observations of the world, right? And so it’s not just they’re internally coherent, there’s some external coherence as well.
0:10:13 LB: Yeah, I should say I probably don’t endorse the coherence theory, but I’m more sympathetic than most. This is just kind of the standard objection to it. That said, what I think people would say to things like what you just said, would be that, there are always lots and lots of ways of explaining my immediate experience, right, so sort of classic examples of…
0:10:32 SC: Brain in a vat.
0:10:33 LB: Brain in a vat, demon deceiving me, the matrix, whatever.
0:10:36 SC: Ultimate brain.
0:10:37 LB: And so the worry is that whatever things I try and bring in to shore up my coherent theory and introduce constraints on it, a sufficiently clever story weaver can find a way to weave those in, and yet still come up with a fundamentally wacky view of things.
0:10:54 SC: Okay.
0:10:56 LB: So that’s kind of the classic problem. Of course, there are things to be said in the favor of the theory, and some people still defend it, but most people I think don’t accept it, and they don’t accept it on [0:11:04] ____ of those grounds. For the pragmatist theory… So the worry for the pragmatist theory is roughly that the thing it just said has almost nothing to do with truth.
[laughter]
0:11:18 SC: It all sounds good, but we’re looking for truth.
0:11:20 LB: Yeah, and so you can formulate that in a couple of ways. Firstly, depending on how sympathetic you are to ideas like this, if you did insist on the kind of empiricism interpretation, then it seems to many people that there are candidates for truth or falsity which don’t look like they’re going to have any experimental predictions at all. So classically, “Is there a God? Tell me some properties of that God.” Whether you’re an atheist or a theist, it seems like, or even an agnostic, in some sense, you’re committed to there being some truths, whether or not we know them or what they might be. And if you interpret the pragmatist maxim in a way which it sort of, experimental validation, is what it means to work out, for me. I just did air quotes. You can’t see me doing air quotes.
[chuckle]
0:12:02 SC: The air quotes are very helpful.
0:12:03 LB: I’m doing air quotes. Yeah, no. Trust me, if you watched me, it would be very good. So the experimental validation looks like it’s just going to miss out on a bunch of things which do seem like candidates for truth, and the other worry is it just looks like you can have kind of useful lies, so if I…
0:12:21 SC: So usefulness is relative to a goal, and you could have different goals. It’s sort of the same thing as coherence, right? It’s like there’s this indeterminacy.
0:12:22 LB: Right. Exactly that, right. So precisely because you can imagine sort of now I have another parameter to play with and determine if something’s true. What I believe, how the world is and what I’m trying to do, and now it looks like I can worryingly make things true or false by varying that third parameter, and that just seems irrelevant. It wasn’t like what we were initially trying to do with truth.
0:12:22 SC: We want truth to be univalent in some way.
0:12:22 LB: Right, in some kind of sense. And so that puts a lot of people off the pragmatist theory. Also, it was unpopular for a while because, in a way which a lot of people don’t think is fair, after the war, Bertrand Russell sort of alleged it was complicit with fascism in some kind of way, because he…
0:12:22 SC: Haven’t heard that one, yeah.
0:13:09 LB: The line being like, “Well, if anyone gets to pick the goal… “
0:13:13 SC: Oh, okay. Oh, I get it. Okay.
0:13:14 LB: And so you end up with some kind of like what’s useful is what’s useful to the Volk, and then we’re in a very bad place.
0:13:23 SC: [0:13:23] ____ Halfway back.
0:13:24 LB: Yeah, then we’re in this old facts world. And so that was another thing which I think contributed to its unpopularity. And then the correspondence theory, which is probably still of those three the most popular of them.
0:13:39 SC: And the most commonsensical…
0:13:40 LB: And the most commonsensical, yeah. It’s more or less viewed by a lot of viewed by a lot of philosophers to… It has a kind of dilemma it faces where either it seems a bit trivial, it doesn’t really explain anything. It’s like, okay, I want to know what truth is. You tell me that truth is consistent correspondence to reality. Well, you’ve just used a lot of other fancy words.
0:14:03 SC: Right.
0:14:04 LB: Which sounds kind of like describing what…
0:14:05 SC: What’s reality, what’s correspondence.
0:14:07 LB: Yeah, and so if you just leave it at that or something not much more sophisticated than that, then people worry that, “Well, it’s not really informative. You haven’t told me anything about truth at all.” On the flip side, and this is where you feel sorry for correspondence theorists, they get that complaint, but then whenever they fill in the details, philosophers start to worry that it’s kind of crazy metaphysics, and…
0:14:27 SC: So it’s both trivial and wrong.
0:14:29 LB: Yeah. It somehow managed to like to tell us nothing and also tell us things which are certainly false, which is something which I thought only my students could achieve.
0:14:39 SC: No, no, I’ve been accused of that many times.
0:14:42 LB: And so there the worry is, if you really start taking seriously the idea, that okay, it’s not just there are cats and mats, and sometimes they sit on each other, but also in the world there is this ordered complex which isn’t quite any of those things individually but which is the unity of them out there in the world, and there is this relationship you can form between your utterance and that thing apparently instantaneously reaching out to this [0:15:08] ____ complex and obtaining and failing to obtain. And people are just like, “What on earth are you talking about?” Like it becomes people’s worry that there’s no way of spelling that thing out which won’t end up invoking mysterious entities which…
0:15:21 SC: Something mystical, unknown.
0:15:23 LB: And so for many people that was… While some people do still endorse that, philosophers aren’t that averse to wacky metaphysics, but that is off-putting to some people and so has sufficed to put many people off the correspondence theory of truth.
0:15:38 SC: Okay. What are we left with?
0:15:41 LB: Well, so for one thing, I said you can always argue about those forever.
0:15:47 LB: Sure. I’m not necessarily giving in. I’m not necessarily accepting, but clearly if people are rejecting those three…
0:15:52 LB: So what I’d say is… This is kind of what I initially was getting at when I said it was distinct about philosophers, because this is, I’d say the following I’m about to outline is the view most distinctive to contemporary analytic philosophy. I never hear it anywhere else, and I think it’s a large part of why scientists and philosophers end up differing so much on truth. So the last view nowadays, which I think is more popular, I don’t think it’s the most popular, but it’s more popular, is what’s called deflationism. According to deflationism…
0:16:17 SC: It’s not a great sales pitch, honestly. [chuckle]
0:16:19 LB: It’s really, really not, and as I explained it to you, it’s going to sound really boring but you know what? It’s your podcast. It’s going to be your problem.
0:16:28 SC: Yeah, exactly.
0:16:29 LB: So according to deflationism, to say something is true is almost just to repeat it. It’s like truth is just a kind of a linguistic device which helps us streamline certain conversations but does nothing more than that. So there’s the classic schema of which is, : X is true if and only if X.” So snow is white is true just in case snow is white, and if snow is white, then, “Snow is white,” is true.
0:17:01 SC: So it’s sort of leaning into the objection to the correspondence theory that it’s just trivialsizing… [chuckle]
0:17:06 LB: Yes. Indeed, right? It’s trying to make a virtue of that. And it’s saying that having that schema, the truth schema, available to us, it allows us to do some things which we might not otherwise be able to do. You have to clean it up a bit. Technically, if you just relied on that schema, you end up with a contradiction. Consider the sentence, “This sentence is false.” If I’m allowed to say, “This sentence is false is true,” then I’m going to be in trouble, because, “This sentence is false,” is true if and only if this sentence is false and then, oh… So you have to clean it up a bit. But if you clean it up to avoid contradictions, then the thought is a theory like that, technically it allows for some mathematical results. You can speed up proofs. You can show that proofs can be made shorter if you’re allowed to involve that kind of thing.
0:17:46 SC: Okay.
0:17:46 LB: And it linguistically allows to do what people call blind descriptions where I can do things like say, “Whatever Sean said last night was true.” Well, I don’t need to know what you said. I can kind of affirm it in a convenient way. So it’s not the only linguistic way of doing that task, but it’s one linguistic way of doing that task. There’s a bunch of kind of tasks truth does for us in terms of expressing agreement or doing some mathematical work, but there’s nothing more to the truth than that. There’s no deep metaphysics here. There’s no kind of interesting theory of what this amounts to. It’s kind of… It’s as if you were having a sort of… I was just about to say an interesting theory of pronouns, but actually that’s super debated nowadays, so it’s not a good example.
0:18:24 SC: Yeah, no, it’s hard to come up with things about which there are no interesting theories.
0:18:29 LB: Right, so truth is uniquely uninteresting. And so that’s a view which is very popular in analytic philosophy, which is [0:18:39] ____ has no option, uptake anywhere else. The relevance of this, I think, is very often when people outside of analytic philosophy, but I think especially in the sciences here talk of something being true, then truth has, I think, in ordinary language, often has a kind of… Has connotations which are sometimes moral, like being a truthful person is a good thing to be; and sometimes a bit mystical, the, “Christ says he is the truth.” And so that seems to sort of imbue it with a degree of religious significance.
0:19:10 LB: And also, unrelated to this. In fact, this would be something which almost all of the theories I just named would reject, but also is often taken to be related to certainty. So if I say something is true, I’m kind of really backing it up, I’m affirming it in a strong way, I’m…
0:19:25 SC: Yeah.
0:19:27 LB: And all of those things are totally separate for philosophers. Philosophers, especially not just, but especially if you have that deflationary view, it doesn’t have any moral significance. It’s just a means of summarizing things. It doesn’t have any connotations of certainty. It’s unrelated to that, and it doesn’t have any spiritual or religious significance.
0:19:50 SC: But if I say something like, “The universe is expanding,” how do I categorize that within a deflationary theory? Do I say… I mean, is it possible to make sense of the statement, “That’s true”?
0:20:00 LB: Well, so on a deflationary theory, you have whatever you wanna say about the universe expanding, presumably as someone who’s very educated in physical theory, like yourself when you say that thing, in the background, there’s a theory of what it means for universe to be expanding. There’s a sense of what kind of things we should expect to observe, and there’s a whole bunch of statements connected with that. When you say those things, that licenses me to now make predictions on those grounds, to assume that you’re gonna affirm the consequences of that claim and stuff like that. But it being true doesn’t add anything to saying… Like saying, “The universe is expanding,” or saying, “It’s true, the universe is expanding,” doesn’t add anything new to all of those entailments which are already there.
0:20:47 SC: Okay. So it makes it very… I don’t wanna say subjective, but it’s from the point of view of people talking rather than from the God’s eye view. Like, “Scientists would like to say that when they discovered the universe is expanding rather than contracting, that would have been just as true if there were no people around talking about it.” Right? Can I still say that in the deflationary…
0:21:07 LB: Well, that’s a bit complicated. In some sense, yes, because you can just say… For instance, you can say the exact sentence you just said, “Even if no one had discussed it, the universe would still have been expanding.” And that has certain… Actually that’s rather complicated but…
[chuckle]
0:21:24 LB: Perhaps that makes certain predictions… Maybe that’s, for instance, affirming that there aren’t any causal dependencies between our words and the rate of expansion or anything like that. And those are all things we can test and we can vet them out, we can agree with them or disagree with them. But saying it’s true doesn’t do the work of affirming that thing. To get that out, you need to make additional claims and make your argument for those claims, for instance, that there aren’t these causal dependencies. So I think what a deflationist would like to say is, anything you were inclined to say by means of using the predicate truth, you can still say under their theory, it just won’t be the case that the property of being true is ever doing any work for you.
0:22:06 SC: I think I’m finally getting it, yeah.
0:22:07 LB: And so that’s why it’s deflationary. They’re not, so to speak, deflating what claims you can make about the world. What they’re deflating is what the role truth plays in those claims.
0:22:17 SC: Right. So it’s not a theory of objective reality or it’s not denying realism, it’s just that this truth relationship is nothing so special?
0:22:27 LB: Yes, that’s exactly right. And so, this view it’s sort of… The common objections to this view are first the… I made that sort of glib remark. You can clean up to get rid of the contradictions. You can guess that…
0:22:42 SC: That’s some work. [chuckle]
0:22:42 LB: Yeah, that will… There are then multiple things you are left with, and you have to decide which are the ones you prefer, and that ends up being controversial. And the other thing people object, and I think this goes back to Michael Dummett, but now there’s a lot of people that make this objection, is that it just misses out a key part of what truth does, which is that we want true beliefs. The point of inquiry is to get true beliefs rather than false beliefs. And the deflationists, while they, of course, have things to say in response to the objection, they seem to be sort of missing out a key role of truth in our conceptual mental lives, if they don’t have that.
0:23:16 LB: So Dummett rather describes it as if you explain to someone the rules of chess, the nature of the pieces and whatnot, and then decline to mention that the point of chess is to check the opponent. You really haven’t explained what chess is, unless you also give the people that detail. And so, the worry about deflationism is its account of the role of truth in our lives, these communicative roles, these technical roles is missing out a key thing and explaining that key thing, you’ll have to invoke something a bit more substantial about what truth is to explain why we care about getting it in the first place.
0:23:47 SC: Is part of the motivation for considering theories like this… There’s this bit that Tarski established about… He was thinking… I guess he followed Gödel. Gödel showed that there were certain statements that I would say are true in a formal system, but you can’t prove them, and Tarski basically proved that there were statements that sounded like they should be true, but you couldn’t prove them. I’m not sure exactly how to say it. It’s like, “The statement P is true if and only if P is true is not a theorem,” something like that.
0:24:25 LB: So, listeners, you can’t see it, but I’m also nodding. I do air quotes and nods when I am on podcasts.
0:24:30 SC: So do I. They’re used to it. I’m gonna get more complaints that I don’t have video on my podcast now, but okay, yeah.
0:24:37 LB: I just love creating problems for you. That’s really why I came on this. So broadly, yes, there is a relationship between Tarski’s work and Gödel’s work, and deflationism was kind of… One of the big props or supports for it was that people were able to take Tarski’s work and say, “Once you have Tarski’s work, that’s really all you need from a theory of truth.” So what Tarski did, more or less was find a way, the first rigorous and consistent way of retaining a lot of that, “Phi is true if and only if Phi,” formula without running into those contradictions. The way he did it was by saying, “What you need is a kind of hierarchy of languages where in the base level, there’s no truth predicate at all. Then, a level above that, you can speak about formulas in the level below that… “
0:25:29 SC: Below it as being true.
0:25:31 LB: “As being true. And then you can have another level of truth predicate where you could also affirm those ones,” and so on, and so on.
0:25:36 SC: And you just live with the infinity of levels?
0:25:37 LB: You live with an infinite hierarchy, yes.
0:25:40 SC: Okay.
[chuckle]
0:25:41 LB: So that’s one way of trying to get a consistent theory out of this. There are other ways of doing this. There are also people who are trying to avoid the hierarchy. So there’s now a lot… There’s a thriving technical field of mathematical logic which is looking at how to do this.
0:25:56 SC: Yeah, it’s pretty fiercely technical if you look at these papers, right?
0:25:58 LB: Yes. It quickly gets…
0:26:00 SC: It’s gibberish, really.
[laughter]
0:26:01 LB: Gibberish. How very dare you, sir? How dare you. And there’s also a thriving theory of trying to give axiomatic theories, where you have axioms governing the truth predicate and see how they behave. So there’s a thriving technical field which is related but not as closely related as you might think to those philosophical theories I just mentioned. Like Tarski, for instance, in his work, at various times suggested he was doing something kind of like a deflationary theory and also something like a correspondence theory, and it’s not really clear that he knew…
0:26:38 SC: What the correspondence is. [chuckle]
0:26:41 LB: Yeah, he certainly didn’t seem to know or care that much which exactly these was, and I have just guaranteed you a bunch of angry emails from scholars.
0:26:50 SC: Yeah, oh good. I will forward them to you, no worries.
[laughter]
0:26:52 LB: No, okay, well. Darn.
0:26:57 SC: But, so, good. Now you convinced me that I have little idea of what truth is, but also given the history, 100 years from now, no one’s gonna accept any of these theories, and we’ll have a different…
[laughter]
0:27:09 SC: We haven’t settled yet anyway, right?
0:27:12 LB: Yes, it does look like…
0:27:13 SC: That’s right, we’ve not equilibrated.
0:27:15 LB: Yes. Well as far as I can tell, philosophy only ever knocks things out of equilibrium.
[laughter]
0:27:21 LB: But yes, right now I’d say, if I was to make a prediction about where this is gonna go, is there will eventually have to be some kind of unification of the advanced technical work with the more traditional theories of what truth amounts to. There are some logicians who are now in a sort of building on Tarski’s initial suggestions to be serious, he was a very good scholar, obviously.
0:27:47 SC: Sure.
0:27:47 LB: And who are trying to claim that in some sense you can see this technical work as a vindication of deflationism. But all I need to do is lay down sparse or fairly minimal rules.
0:27:58 SC: Some rules. Yeah.
0:28:00 LB: So you’re starting to get that. There are other people going in other directions too, but I predict that as time goes by, we’ll get more of a sense of how the technical, logical, mathematical, more scientific work relates to the traditional, high level philosophical work, and that will give us a better understanding of what truth is and what kind of role it plays for us.
0:28:19 SC: Okay, good. I wanna go back to moving this to connect with science, once again, but I think there’s still some more philosophy that we need to get out of the way.
[laughter]
0:28:30 SC: When we talk about true statements or when we talk about the statements, different kinds of truth, there’s things that you can imagine proving logically, things that you have to go out and experimentally verify. And I know that… Let me try to say this in my words, and then you can fix it. [chuckle]
0:28:45 LB: Happy to.
0:28:49 SC: The notions of the reasons why things are true, some things are true in virtue of themselves or the definitions of their words, and we call those analytic. There are things that are virtue of true because that’s how the world is, but it might have been otherwise, and we call the synthetic. There’s also how we know that things are true. There are things we know just from reason alone, a priori. And things that we know by going and doing experiments, a posteriori. So these are two very similar sounding things but a little bit different. Did I get it mostly right?
0:29:19 LB: That’s, yeah, a reasonable summary of the traditional distinctions you’re drawing, yeah.
0:29:24 SC: Yeah. And so to me, and I guess to David Hume, it sounds like there’s this very close connection between a priori truths, ones that are true that you can show just by logic, and analytic truths, ones that are true just by virtue of their definitions. Whereas there’s another relationship between a posteriori, synthetic things that you go out and… How did I define synthetic? I forget. But synthetic sounds like a posteriori. Analytic sounds like a priori.
0:29:55 LB: Yeah.
0:29:56 SC: Do people basically buy that? I know that Immanuel Kant tried to argue there was wiggle room there.
0:30:02 LB: Yes. So Kant famously argued that there’s a category of things which are synthetic making substantial claims, not just given in their definitions, but really making claims on the world but which are nonetheless a priori, which are things you know by means of reason, reason alone. And as far as I can tell, and this is another one where the scholars… Eventually I’m just gonna have to say, “Scholars, just deal with it,” like, “You can’t get me fired. I’m safe now.”
0:30:29 SC: That’s right.
0:30:31 LB: And so, subject to the correction from scholars, Kant seems to have been kind of thinking about then relatively, relatively recent Newtonian science and thinking… He was very confident like, “Finally we’ve achieved knowledge. We know something about nature,” which is that it works on broader Newtonian lines. And sort of the dominant epistemology coming from many sources, primarily Loch and the European continent, was that this is kind of experiential knowledge. We know this because we’ve done the right kind of experiments, gained the right kind of experiences, and it validates our claims, and we finally worked out how to do that.
0:31:12 LB: And Kant saw that that couldn’t quite work, and the reason it couldn’t quite work he thought, and I think many people agree with him, is that some of the claims that seem to be necessary for Newtonian physics didn’t seem to admit of those kind of tests. For instance, it seems to people at the time, although we would now disagree with this, but it seemed to people at the time that that space has something like Euclidean structure was something we presupposed when did our experiments. We didn’t test that, rather when we were making… Testing claims about the distance between things or the speed at which things moved, we were then presupposing it.
0:31:51 LB: And so Kant worried that this seems therefore like there’s this worrying fact that a bunch of really foundational claims to this science which he really wanted to validate as achieving knowledge of the world, were just not supported by the means that we had to support things. And so Kant built sort of a philosophical apparatus to explain how there could be some claims actually which were substantial, for instance, that space has Euclidean structure, while at the same time not requiring empirical validation; we could know that just by reflection. And so that was, for Kant, that was equal to synthetic a priori.
0:32:25 SC: Okay. And that sounds wrong especially because it’s not even exactly true that space has a Euclidean structure, right? So that might have been a little premature on Immanuel’s part.
0:32:36 LB: Yes. Right, so it’s a little bit complicated, what the relationship is between what Kant said and what eventually became refuted. It did seem…
[laughter]
0:32:51 SC: He’s not a famously clear writer.
0:32:53 LB: Not quite, no. And in particular, the issue is it’s not like what happened was we stopped having apparently a priori roles for some kinds of geometric structure in space. We may not use the particular one which Kant thought we had to use, but still there’s something playing a similar… Some people might argue there’s something playing a somewhat similar role. I, to be frank, have no real opinion about this and…
0:33:23 SC: Yeah. Those people are wrong, just so we know, yeah.
0:33:25 LB: Yeah. So, but some people claim that at a sufficiently high level of abstraction, you could maintain Kant’s claim, he was just too specific in saying this is particular structure space has to have, but he wasn’t wrong that space has to have… That we need to a priori attribute to space some kind of structure in order to do our reasoning and…
0:33:45 SC: But there… You see, this is where we’re getting back into truth, right? I’m open to the idea that there are preconditions for us making sense of the world, but that’s different than saying those preconditions are true. You can then say, “And I have made some sense of the world, therefore, empirically, I have verified that this particular presupposition works.”
0:34:07 LB: Well, I guess, this does actually… This might be a point where which of those theories of truth we endorse makes a difference to how you wanna understand what’s going on. So, more or less, for the deflationist to say that it’s true that space has such-and-such a structure is more or less just to repeat that space has such-and-such a structure. So, once you’ve told me your theory, you’re not adding anything to it by telling me it’s true. And to that extent, kind of there’s not really… He would just disagree… I say, he as in I am the deflationist. I would just disagree with the claim that like… To say that space has a structure is different from saying it’s true that space has a structure.
0:34:49 LB: On the other side, to a correspondence theorist, maybe there might be more involved ’cause maybe when you say that space has this structure, there’s something extra involved in saying, “And that corresponds to some external thing,” maybe some claims are different, and they don’t need to have that correspondence theory. This would be a version of what people call truth pluralism, where they have that… There’s kind of more to the correspondence relation because different claims correspond to the world in different kinds of ways. And so I need to tell you not just that space has a structure, but what kind of correspondence relationship you need to be looking for in order to bear this kind of claim out. And that would make it such that there is like a little bit of extra work to be done, at least in explaining exactly what you mean when you say space has this structure. And so this might be a point where like which of those theories of truth seems tempting to you, at least makes a difference to what kind of things you think you need to do to fully explain what you mean when you make claims about the structure of space.
0:35:46 SC: Okay, that’s fair, but it does sound like, at least provisionally, I’m okay in making my naive statements about, “There are mathematical and logical truths, and there are scientific truths, and they’re different”?
0:36:00 LB: So, no, I just didn’t get to that yet.
[laughter]
0:36:02 SC: We have all the time. Don’t worry.
0:36:05 LB: Alas, in the 20th century, this is kind of associated with Quine, eventually Kant’s… This division between analytic and synthetic truths, truths which are true in virtue of their meaning versus truths which are true in virtue of something about how the world is, independent of language, came under quite stern criticism within philosophy. And more or less the criticism was that like, so to speak, it’s a little bit arbitrary how you categorize things on one side of this versus the other.
0:36:41 SC: Okay. [chuckle]
0:36:44 LB: So the thought being that… I’m not quite sure what the thought is ’cause I’ve always thought the attacks were just silly.
[laughter]
0:36:51 SC: We can categorize whole regions of philosophical debate as silly. That’s okay.
0:36:55 LB: Oh, I’m on Twitter, I do that all the time.
[chuckle]
0:37:00 LB: But I guess it’s just the thought being that this notion of an analytic truth, something true in the virtue of meaning, when you try and cache it out, it’s kind of similar to the complaints about the correspondence theory. When you try and cache it out, what you will find is it depends on notions which are themselves quite mysterious, and we don’t have a theory of it. That you have to sort of invoke notions of synonymy or necessary equivalence which themselves aren’t well theorized. And once you try and really think about this or really make sense of it, eventually what you’re left with is something more like, “Well, there are some claims which I’ll do more to defend than others.” In terms of, “You really have to do a lot of work to make me give up on the claim that all bachelors are unmarried men.”
0:37:48 LB: Like that’s something which really, really, really requires a pretty drastic change in my worldview. But you could do it, like you could… It’s just, so to speak, I haven’t yet found the circumstance where it would be convenient for me to give up that claim, but you could do it. Other claims are closer to being revisable in light of my experiences. And so what a lot of philosophers came to think was this notion of analytic versus synthetic was just a bad way of making theoretical sense of the difference between how entrenched beliefs are for us, and really all there is is this entrenchment, and that’s a sort of much more that comes in degrees…
0:38:25 SC: So Quine is trying to give us a spectrum, rather than just the two options?
0:38:28 LB: Yeah, more or less, things we more or less central to our belief structure and…
0:38:34 SC: But are you advocating this or mentioning that there are people who advocate it? ‘Cause it does not sound very convincing to me. I think that all bachelors are unmarried men. [chuckle]
0:38:42 LB: I am mentioning it rather than advocating it.
0:38:44 SC: Yeah, okay.
0:38:45 LB: I tend to think that… It’s one of those things where it relies on, a crucial point in this argument is that there are these somewhat mysterious notions we just can’t make sense of. Quine, the person who most influentially launched this attack, was making it in the ’50s just before there were a lot of developments where we developed tools where I think we can just make sense of those notions. And so I think it was sensible in light of the state of theoretical development at the time, but I don’t think it really bears out. And so it sort of entered philosophical lore that this was refuted, but I think it was sort of, that decision warrants re-examination in light of later developments.
0:38:57 SC: Yeah, I have this vague feeling from a completely unqualified perspective, that if we fix some things up in our use of language and logic and things like that, there would be a clearer distinction. And maybe it has to do with possible worlds and if something is true in every possible world or something like that, that’s different than being true in some possible worlds but not others?
0:39:43 LB: Well, that’s… Some of the technical machinery which philosophers use to analyze a bunch of concepts will be in terms of, “Oh, it’s good modal logic,” which is the logic of possibility and necessity. Well, how philosophers tend to think about that is they will have possible worlds, which depending on how you describe them, are either concrete real existent worlds, just not ones we can physically reach…
0:40:05 SC: Yeah, that’s the one scientists don’t care about, but yeah, there are philosophers who buy that. Yeah. [chuckle]
0:40:08 LB: Or more sensibly, just complete descriptions of the world which may or may not be actual, which you may not endorse, which may not be true, but you can give various descriptions of the world. And…
0:40:20 SC: I should… Sorry, I really need to… Before I forget, mention that Max Tegmark, a previous podcast guest, really does believe in all the other possible worlds also. He’s a mathy version of David Lewis, yeah.
0:40:32 LB: One of the genuine, and I mean this, genuine joys of being in academia is whenever I think something’s just too wacky, and no one could believe it, there are always people…
[chuckle]
0:40:41 SC: There’s always very respectable people…
0:40:42 LB: Ten times smarter than me who have passionate defenses of that, and I’m just left in awe of what the human mind can do. But, yes, right. So you could view possibilities as just a stats thing or as descriptions of how the world might be, and then the apparatus of modal logic is a way of reasoning about like, “Well, given… If we think of ourselves in this space of possibilities, what follows? What’s necessary from here? What’s only possible from here?” Etcetera. And some of that machinery can be brought in to try and make sense of the notions required to make sense of analyticity. I don’t wanna give anyone the impression that this is settled, but I happen to think you can do it but there are reasonably intelligent people who don’t think you can do it, but those are exactly the developments I have in mind. And that just kind of it’s a nice illustration really of if you follow the story, there’s…
0:41:38 LB: We begin with… Already in the early modern Europe, appearance of a interaction between philosophy and scientific reflection, where you have philosophers and scientists trying to think about like, “How can we be confident in what seems like a very well grounded theory?” From that we get new technical mathematical machinery designed to try and help us make sense of those notions, which philosophers then used to problematise some of the initial notions of theory that it was built on. But then that itself spurs new technical developments, and we are getting better at being able to validate this. And it’s a really nice virtuous circle of critique and response to critique, in a way, which is developing us new tools and allowing us to make sense of things which were hard to make sense of before.
0:42:22 SC: Good. And I think this leads us right into the scientific practice that I was hoping to get to. The system you just described is what science would ideally like to be like with a little bit of data thrown in, right? We come up with theories. We compare them with data. We still unfortunately teach kids, at least in the US, that there’s something called the scientific method that Francis Bacon laid down, and all scientists follow. And a little baby hypothesis grows up to be a theory and then finally graduates to being a law. None of that is actually true in scientific practice, but there is some sort of the give and take between hypothesizing and collecting data is certainly there. What is your read on a better way that we might teach high school students how science is done than the Baconian scientific method?
0:43:14 LB: Yeah, right, I think this is a really fascinating topic, and it’s actually something I think about, professionally. Where exactly as you say, it’s the same in the UK, people are taught this kind of stylized vision of scientific method, and it really just doesn’t correspond to how I think science in practice works and even how it should work. It wouldn’t be good if we tried to make science work like that.
0:43:37 SC: And the irony is very thick, right?
[laughter]
0:43:40 SC: A theory that has no relationship to experiments. [chuckle] Okay.
0:43:46 LB: So I think… Right now, I think the best bet I could have for how to reform scientific practice… That’s not right. Scientific education would be to base it actually on some ideas from a different philosopher, this would be Dewey from the American pragmatist tradition.
0:44:03 SC: Pragmatist. Yeah.
0:44:03 LB: Where Dewey was this famous educational theorist is actually one of the places where he had most influence, and he had sort of learn by doing was the philosophy. And he really actually kind of modeled his whole theory of education on what he took to be a more realistic model of how scientific inquiries is done. And the idea would be you sort of give people problems, materials with which to solve those problems, and space to reflect on what they’re doing as they’re doing it. And so you give people some materials and tell them, “How can I make this chemical do that? To turn this colour?” Whatever it might be.
0:44:43 SC: Yeah.
0:44:43 LB: They then get some space to play around with the materials they got, and then afterwards they’ve gotta tell you what did you do and why.
0:44:50 SC: Right.
0:44:50 LB: In the process of doing this, you’re gonna learn a bunch of facts about scientific inquiry because in some sense, science is that kind of activity but scaled up and in much more communication with other people. And so, for instance, a classic example. Here’s something which doesn’t tend to feature much in stylized descriptions of the scientific method but it’s actually super important and is the subject of much debate now around the replication crisis. At some point, when you’re testing a claim, you have to stop testing it, stop seeking new evidence, and decide, “I have enough evidence to write this up.” How do you do that? What are the rules which decide how you do that? That doesn’t feature at all in the normal picture of the stylized picture of scientific method. Sometimes people talk as if at one point nature rings a bell and says…
0:45:34 SC: It’s binary. You got it or you don’t, right? Yeah.
0:45:37 LB: Right, exactly. If you set children that task, they would have to discover that problem because you have to at some point work out like, “Now we’ve done enough to be confident this is the way it’s done.” And in so doing, they could reflect on, “What kind of features of the scenario made me think that now was the time to do that? What would have made me think we need to do more? What made me think we need to do less?” And so by engaging in scientific research, they’ll learn key features of how the practice is done, how these decisions can be made in a way which don’t fit the stylistic method but aren’t arbitrary either, are responding to features of your investigatory situation. And we’re doing something which is kinda fun. [chuckle]
[chuckle]
0:46:14 SC: It’s way more fun than learning all the elements or whatever.
0:46:17 LB: Yeah. And so this is the kind of thing which I’d be really interested in there being much more experimentation around.
0:46:21 SC: Yeah. But I think the part that is valuable or important to me that maybe you didn’t mention there is this idea that we are comparing different possible worlds. We have hypotheses, just going back to the previous discussion, like a big part of the scientific method is saying, “I don’t know which world we live in. There’s different theories that might describe the world. They’re all theories that describe some world. Which one do we live in? So, we’re gonna collect more data, and maybe it’s not binary, but at least as good Bayesians, we will increase our credence in one of these hypotheses and decrease in the other.”
0:46:56 LB: Well, actually, I would hope that a good process of doing education would help people do that. Because one of the things one wishes to learn is what kind of experiments are worth doing? And eventually you need to learn that some experiments don’t differentiate between hypotheses, because in either of these possible worlds, you would expect the same given this test. And so, in learning about like, “Well, how do I do an experiment which decides or at least gives me evidence relevant to deciding between these hypotheses?” You’re having to think about it and by the process of doing that, you have to hold in your mind, “Well, what would be true if this hypothesis and what would be true if that hypothesis?” Listeners should be aware, I’m now doing very helpful hand gestures for each hypothesis. So what would be true in each of those hypotheses, do the working out what the consequences would be and how they’d apply for the materials you have available. And that is as a baby version of something like hypothetical or deductive reasoning which is if not the be all and end of all science is, at least part of it. And that is also a type of reasoning which necessitates you thinking about different possibilities.
0:48:00 LB: And so again, allowing people… Throwing people into scenarios, as long as they have the time to communicate and reflect with each other, really I think actually teaches in practice a lot of these things as long as you guide them to point out features of like, “Well, here’s what you’ve just done.” And so, yes, I certainly agree that a core feature of any scientific reasoning is, in some sense it’s sense it’s counterintuitive, but it’s actually important that we reason about what’s not actual. Science requires that you reason about possibilities which aren’t true, which aren’t accurate descriptions of the world, and that’s kind of interesting in many ways. I also… To come back to something that you first said, a core reason why when you say, “I have my model. Isn’t that true?” Almost all models, in fact, involve false claims.
0:48:45 SC: Right.
0:48:46 LB: Right, so that can’t quite be right. And so, there are al these times in science where, and I would think people would learn really by doing this is in some sense science doesn’t just consist in saying true things. There’s a bunch of non-true stuff you have to deal with, but that doesn’t invalidate it either.
0:49:00 SC: No, not at all.
0:49:01 LB: And so one of the advantages…
0:49:02 SC: It validates it, right? [chuckle]
0:49:03 LB: Exactly. That’s how in fact we gain reliable knowledge.
0:49:05 SC: Yeah.
0:49:05 LB: And so one of the advantages which philosophers like Heather Douglas or Philip Kitcher often claim for citizens or young people being able to do this kind of more engaged, Dewey-an education is that a lot of attempts to undermine public trust in science involve pointing to these features which people aren’t aware are a typical feature of scientific practice, but which is just part of how we do things and pointing out like, “Well, did you know that these models involve a bunch of false assumptions? Then how can you really trust what the climate scientists are saying?” If you knew how science works like, “Yeah, the models involve false assumptions. It’s not… It’s how we do modeling.” Right?
0:49:42 SC: Yeah.
0:49:43 LB: And so by giving people practice and seeing for themselves that like, “Oh, I made some simplified assumptions. That wasn’t a problem. That’s what allowed me to get the answer.” It would immunize them in some sense to certain kinds of spurious skeptical attacks.
0:49:57 SC: Right, right, and… But the idea, I really just… Maybe I’m just repeating myself, but the idea that we could be wrong, that we are not simply, purely reasoning our way to how the world works is very crucial to me in science. I remember hearing a talk by an unnamed evolutionary psychologist who was founding an institute or something like that. And they made it very clear, “The purpose of our institute is to show that blah, blah, blah, is true.”
0:50:26 LB: Right.
0:50:26 SC: And I wanted to raise my hand as like, “That’s not really very science-y there. How about to find out whether or not it is true?” There’s a slightly, there’s a slight but important distinction there.
0:50:35 LB: Well, I’m an employee of the London School of Economics. This department was founded by Karl Popper. So you know…
[chuckle]
0:50:43 LB: I’m legally obliged to agree with what you just said, but there are some philosophers, heretics, who might think it’s at least a bit more complicated. Now, I mean, jokes aside, I honestly am actually pretty sympathetic to a broadly falsificationist view of what we’re up to in science. It doesn’t always sit so easily with the Bayesianism, set that aside for now.
0:51:04 SC: Sorry, I’m much more Bayesian… I’m not very falsifiable. I think I got in trouble by writing a paper saying… The title was, “Beyond Falsifiability.” And to point out that, “Look, this was Popper put his finger on some things that were really, really important, but it’s not the final solution of the demarcation problem,” and I think that Bayesianism helps get us closer to a more nuanced view.
0:51:29 LB: Right. Okay, in that case I think we’re probably sympathetic. I think there’s something to be said for the falsificationist picture, but I agree it’s not the be all and end all.
0:51:37 SC: Either way, you have this idea that you could be wrong.
0:51:40 LB: Yes. But what I was gonna mention is there’s a vision of scientific practice which is in some sense dependent from Kuhn, although I’m quite sure Kuhn would probably not want to be associated with some of the people you perhaps have in mind, wherein the way science works is by setting up paradigms. Those paradigms involve not unquestionable, but at least relatively unquestioned assumptions. And it’s legitimate to go forward like, “The way I do science is in light of those assumptions, I’m now gonna draw out the implications and test for other things you can make sense of in light of those but not necessarily test those.” We’re in the Lakatos building right now, and so, Lakatos is famous for saying, “Scientific theories have a protected or a hard core in the center and that… Eventually that can be penetrated, but it’s of reasonable procedure to begin with assumptions which you are less willing to give up.”
0:52:38 LB: In some sense, that actually relates to the Quinian point we were saying earlier, where many who do think a sort of reasonable vision of how rational inquiry goes, is you have some beliefs which are much more central, which you’re much less willing to give up. And it’s okay to do things by first trying to swap out the less core beliefs and seeing how you can do things. And of course you can imagine the sort of quasi-Bayesian justification for that kind of procedure.
0:53:00 SC: Well, this is why one of the issues with falsifiability as a solution to the demarcation problem is there’s plenty of examples where I have a theory. It’s very well established. An experimental result comes along that is in contradiction to that theory and everyone in their right mind says, “Oh, don’t worry. That result will go away.”
[laughter]
0:53:18 LB: I still think light draws faster…
[overlapping conversation]
0:53:25 SC: Faster than light, yeah. That would be an example, yes. That’s right. Yeah. Scientists, they have their pre-conditions. They have their, what we call cherished principles, this is what we like to call them. And you don’t wanna give them up, but eventually, you end up doing it if you get enough evidence against it. And to me, that kind of fits in with the correspondence theory a little bit. Sorry, with the coherence theory, I wanted to say a little bit.
0:53:46 LB: Can you say more about how?
0:53:48 SC: To be coherent is to be coherent amongst a bunch of beliefs, and some of them will be more switch out-able than others, right?
0:53:56 LB: Right. So…
0:53:58 SC: Some of them will be more foundational, some of your beliefs.
0:54:02 LB: So the thing about both the pragmatist and the coherence theories of truth is they clearly pick up on aspects of our mental lives which are important. We do in fact have goals, and beliefs do help us fulfill those goals, and there’s some reason to prefer a belief if it helps you fulfill your goal. Our beliefs can be more or less coherent, at least in the sense of being self-consistent or not being self-consistent, and you may prefer to be self-consistent. I think there are true contradictions. I don’t know if I… [chuckle] One may…
0:54:28 SC: That’s another podcast, yeah.
0:54:30 LB: One may prefer to be self-consistent. And in that case, and so for that reason, the coherence theory must be picking up on something because it’s more or less just saying that particular virtue of your belief system being coherent, like that’s the one which makes the difference with truth. So, what non-coherence theorists are gonna say is like, “We have a way of recovering, within our theory, the importance of coherence.”
0:54:53 LB: So, for instance, the correspondence theorist will tend to have a metaphysical theory about what the facts are and just say like, “It turns out that the facts can’t be such that they’re inconsistent of each other. And so if your theory is incoherent in the sense of being inconsistent, we know it’s false because we know that the facts aren’t incoherent.”
0:55:11 SC: That’s fair. Everyone wants to be coherent, just whether or not that’s the [0:55:15] ____, yeah.
0:55:18 LB: And so, right. And so what you’ll find is that theories of truth… And this is, I think, one of the reasons why theory of truth tends to… Something which attracts maybe people to deflationism, is it tends to be the case that you can recover whatever nice things you wanna say about one theory of truth within the other theory of truth. And without just being a justification for deflationism, it can prompt the thought that truth… It doesn’t really matter what you end up saying about truth. You always find a way of saying the nice things you want to be able to say anyway. And so truth isn’t really doing much there. You just have some additional principles you’re able to bring in, which get you the goods you want.
0:56:00 SC: Is it sort of a dirty little secret of philosophy that so much of it is driven by things you want to be the case, [chuckle] and then finding ways to make the things you wanna be the case, be the case? Or maybe that’s a virtue. I don’t even know. But certainly, you wanna think of moral philosophy that often ends up being true…
0:56:14 LB: Well, Quine said…
0:56:16 SC: But maybe even epistemology true.
0:56:17 LB: I’m not sure it’s either dirty, little, or a secret.
[laughter]
0:56:20 SC: Okay.
0:56:20 LB: So for that reason.
0:56:20 SC: But other than that…
[laughter]
0:56:25 SC: I think it’s dirty.
0:56:26 LB: It’s certainly the case that… So is it a secret? People say mean things to me on the internet all the time which suggest that they think philosophy is so… So plenty of people probably know about philosophy. Is it little or dirty is a bit more complicated. Because, yes, it’s true that philosophy as a field precisely because it’s relatively unconstrained by direct empirical data or relatively unconstrained by rigors of proof, mean that if you come in with like, “I’m really committed to keeping this principle,” you can find a way of constructing a worldview which retains that principle, but which has other desirable features. And some philosophers more or less just think… And that’s the game.
0:57:18 SC: That’s what we’re doing, right.
0:57:19 LB: The game is you come in with the things you’re really committed to…
0:57:21 SC: Here are my commitments.
0:57:22 LB: And you have to then… The world will, no doubt, still present you with puzzles. So I just mentioned I believe there are true contradictions, suppose you don’t. What do you wanna say about sentences like this very sentence is false?
0:57:32 SC: So you do believe there are true contradictions?
0:57:34 LB: Yes.
0:57:34 SC: That was [0:57:34] ____.
0:57:35 LB: I do. So I more or less think what we want to say about, “This very sentence is false,” is that it’s both true and false. And so if you don’t… And I’m not going to defend it right now, but if you don’t think that, you have to tell me something because it turns out there are sort of pretty minimal and apparently plausible things you might say about truth which will get you, “That sentence is both true and false.” And so it’s not…
0:57:58 SC: Roughly, the alternative must be that it’s meaningless or unsayable. Out of bounds, somehow.
0:58:04 LB: You can out of bounds somehow. You can have it not well-formed because you’re applying a truth predicate to itself in a wrong kind of way, or you can have it such that it gets no value, like it just fails to be evaluated. But in any case, building a worldview which gets you either out of bounds or meaningless or not well-formed or whatever, that requires work. And so in other words, it’s not a trivial game, even if you do come in with your strong commitments like to build a theory which does that. And in the process of building those theories, you often are forced to do things which are themselves intellectually quite valuable. And so, some people will say that this activity of elaborating upon and defending one’s core commitments is actually in the end intellectually very valuable, because if you take that game seriously, you’ll find it’s really difficult, and what you have to do to play it well is just a real driver of innovation.
0:59:03 SC: Okay. Some of my best friends are philosophers. I don’t want to…
0:59:09 LB: No, I’m perfectly happy with being self-loathing.
0:59:12 SC: There we go. That works.
0:59:13 LB: Yeah.
0:59:16 SC: Speaking of self-loathing, let’s not let scientists off the hook. You already mentioned the replication crisis. I want to get a little bit down and dirty into the difference between the pristine way in which scientists naively think of their own pastime and the ugly truth of how it gets done and… Because the ugly truth always becomes a little bit prettier when you really face up to it, right? So…
0:59:39 LB: I thought we were being self-loathing. You’ve gone two minutes…
0:59:40 SC: I’m not self-loathing at all. Someone’s gotta love me. [laughter] “Why not me?” is my attitude. It’s certainly not the YouTube commenters, I know that.
0:59:50 LB: I love you, Sean.
0:59:53 SC: Alright. So the replication crisis. I think most people have heard of it, but why don’t we say what it is?
0:59:58 LB: Sure. So in social psychology, especially but not just in social psychology, it’s, in relatively recent years, been found that a bunch of results, which people took to be secure, which were featuring in textbooks and which public interventions were being based on, in fact don’t replicate. Which is to say, when we try and retest the experiments which initially validated those phenomena were real, we find either that we can’t get the effect at all, or it’s much reduced in its strand of full degree of influence than we initially thought. So the sense is… That’s just replication failure.
1:00:33 LB: What made it a crisis were, in a sort of ill-defined sense, this happened so many times to so many results that people felt that there was a systematic problem here. I raise that explicitly, because I’m often a little bit underwhelmed by the arguments of that second part. Like, what is the rate at which we should expect true ideas in science? I think people are blase about assuming it’s a crisis. But nonetheless in the course of… The fact that this prompted a kind of… Panic reaction is too strong, but this prompted quite a strong reaction from the scientific community and the social sciences and statistics especially, and nowadays as well actually in philosophy of science, led to a bunch of problems being uncovered with the way science was done or at least things which people were pretty clear could be improved upon. And so now there’s a quite broad reform movement sweeping through the social sciences, which is trying to address the problems thus identified.
1:01:32 SC: And presumably part of it, part of the worry that it was really serious was that, like you say, it’s not just random but reflects systematic biases in the field. When you get an exciting result…
1:01:41 LB: Yeah.
1:01:41 SC: That’s where you publish it, right. When you just get a null result, no one is going to care.
1:01:45 LB: Yes. One of the fascinating things about the replication crisis is that a lot of the things we are now responding to it in light of, were… All of these problems, people were aware of them in some sense for a long time. And so there’s the kind of… The classic one is the file drawer problem, right, where if I get a positive result, suggesting there is an effect, that’s exciting, especially with a new thing, I can go and publish that in a high prestige journal, which lots of people will read and become aware of. If I fail to find anything, and even if I just replicate that, that’s not exciting. That’s much harder to get published, and often that information will only be available to the particular scientist who did the thought experiment. And so there’s a massive bias in what kind of information is transmitted to the community at large. Now, that has certainly been a thing which people have discussed a lot in light of the replication crisis, but there’s no way that people only became aware of this possibility inside of publishing…
1:02:39 SC: Right.
1:02:40 LB: In 2015. That’s a thing people have worried about for a long time. And so one of the things the replication crisis has really done, it’s like the replication failures were dramatic and drew people’s attention to problems, but what we’re actually now doing in light of that is really cleaning up house in ways we probably should have done a long time ago. And so, right now, of course, there’s a lot of anger in some ways as people are sort of arguing with each other about what should be done. Some people are feeling defensive, because they feel like their life’s work is under attack. But in the long run, I’m actually very optimistic, that what this is, like, only doing what we should have been doing for a long time and, hopefully, it will lead to a much stronger, much improved scientific process in the future.
1:03:22 SC: But once you do admit that even scientists have biases, then maybe they have other biases as well, right? This becomes a problem of much wider scope, and we can start questioning… People take it seriously, right? It sounds like you’re questioning their integrity. Although you’re really not. You’re just doing epistemology in some sense.
1:03:44 LB: Yeah. I mean, so there are all sorts of other biases besides just the bias towards publishable or fancy and more exciting results and other ones which, of course, get much attention are sort like demographic, like gender biases, for instance, and who can get published and where.
1:04:04 SC: We don’t have those in physics, but I’ve heard…
1:04:05 LB: Yeah.
1:04:05 SC: That you have those in other fields.
1:04:07 LB: Yeah. Like the gender utopia, which is physics. Certainly that’s like one source of bias, which has really negatively affected a lot of people’s lives if we’re being serious. And so, other kind of biases you might see are, for instance, towards not just things which the scientific community will find exciting in terms of publishing in fancy journals, but also, and this is very worrying in light of what we do as science, there might be biases towards the kind of thing which the media will pick up on, on which policy people will pick up on. And so that you get some cases of what looks like outright fraud and that outright fraud being driven basically by the sense that there’s a way to clout with the media and with policymakers.
1:04:50 LB: And so, yeah, all of these things do influence how scientists behave. They can have the social biases, which are very common. They need to get published in order to advance their career. In fact, most of my research is about the effects those career incentives have on working scientists. And, also, like anyone else, they want to change the world in ways they view to be desirable. And so that desire to influence policymakers or people who are going to make a difference to the world, that can really affect their behavior. And, yes, some of these seem like they admit of relatively simple fixes. It’s not exactly clear how to check the file drawer problem, but there are both technical statistical, actually Bayesian methods…
1:05:29 SC: We can do the statistics better. That’s what it’s called.
1:05:30 LB: We can do the statistics better. And you can also reform how journals publish or what kind of things that give an uptake in such a way that we can mitigate at least the effects of the file drawer problem. But, do you have a solution for gender bias? And do you have a solution for the subtle influences of wanting to be influential in the world? These sort of things… There are a bunch of things here which are going to be much harder to really fully correct for, but which all seem… Which is part of the phenomena we know as the replication crisis. I think that what we have to do, so to speak, is do our best to fix these things. But also in some sense it’s just quite right for people to decrease their trust in this institution insofar as their trust was previously based on a image of scientific researchers or science as a social process which made it immune from those things. It’s not immune to those things. And so a well calibrated agent trusts it to the degree that they have taken into account the real presence of these biases.
1:06:27 SC: Sadly, there are no well calibrated agents, right? People either think it’s infallible or completely untrustworthy, right? That’s the problem.
1:06:34 LB: I’m not so pessimistic about that. But the reason I’m not so pessimistic is just because I think I encounter people quite frequently who are of course not perfectly well calibrated, but much better calibrated than just like unthinking deference versus total skepticism, do climate scientists rule the internet too style reasoning. And those are people who are familiar with it as a practice. I find that both philosophers of science and also working scientists don’t tend to have this kind of attitude except about their own work, right?
[chuckle]
1:07:11 LB: And then that’s just are you self-loathing or wildly overconfident, and those are the only two options. And so then relating then to that previous book on education, this is kind of what I meant by immunizing people, I do think that it seems that familiarity with the practice, being engaged with science tends to actually give you a sense of there really is something unusually reliable about this as a way of reasoning. It’s not just any other way of reasoning about the world. There really is something special about it. But at the same time, it’s not magic either.
1:07:41 SC: No.
1:07:41 LB: And it’s only as good as lots of smart people repeatedly failing, but maybe failing a bit better could be. And that’s about right. And so I hope that actually by sort of in some sense democratizing science, by bringing more people in, by giving more people an experience of the practice of scientific research much more calibrated to how it’s actually done rather than the sort of quasi-authoritarian [1:08:05] ____ retrospect, “There are some laws we’ve discovered, and you are to believe these laws, and the test consists of you repeating these laws to me.” Rather than that, we have definitely… Give more proof as a sense of like what this enterprise actually involves, that would actually really help. That would create those calibrated systems. So I think there are some people who have it now, and I’m optimistic that there’s a reasonably exportable method of creating more people who are calibrated in that kind of way.
1:08:30 SC: I would like to be optimistic also. But before I become optimistic, I do want to emphasize there’s even more ways to be pessimistic. I think that when it comes to biases like gender biases and racial biases that exist in science just as much as in society, sure, but then there’s these softer biases which probably also exist in society where people who have idiosyncratic ideas will find that they just don’t get paid attention to, or they don’t get jobs, or they don’t get grants, and it’s not like anyone is banning them, but they’re just choosing not to pay attention to them, and their ideas can wither on the vine or not get paid attention for that reason. And of course, they will say it’s ’cause the system is biased against them. Maybe they’re right, or maybe their idea just isn’t very good. Right? And do we have any objective way of telling these things?
1:09:20 LB: No. I think that one of the things the well calibrated agent with regards to their trust in science just has to learn is that… And this is a thing which I do think which Popper, even if he wouldn’t have put it exactly as I would put it, was he really had the fundamentals of this idea in his [1:09:38] ____. Science is, so to speak, a quasi-evolutionary process. There are lots of people throwing up lots of ideas, lots of the time. Most of those ideas don’t survive. A lot of them for good reason. Occasionally, we’ll miss one which we should have paid more attention to. But so to speak, we haven’t got anything better than just having lots of people try their best and throwing themselves at a problem come up with the ideas they come up with.
1:10:04 LB: And as long as there are incentives to pay some attention to new ideas, which there usually are in science, we can talk about that in a second, we can reasonably hope that eventually we will catch good, valuable new ideas, and we will actually have something. We will eventually be able to build on them. And I think those incentives to take new ideas seriously do exist in science because if a new idea really does work out, then I can use it to bolster my ideas.
1:10:33 SC: Yeah.
1:10:33 LB: The thought being that if I have access to some new experimental technique which actually works, some new mathematical theory which allows me to make inferences I couldn’t otherwise have made or whatever, then suddenly, I can do a thing to bolster support for my ideas. At least I have some incentive to be trying to search in the field and looking out for things which other people have missed which might give me an advantage in the race to get my ideas taken seriously. And so the incentive structure of science does provide at least some push towards taking those ideas seriously. I’m not pollyannish. Actually, a lot of my work is about how I think we can try and do this better. But these are features of the present system, and they just need to be built upon and strengthened, rather than we’re not starting from nothing here.
1:11:12 SC: You did propose the radical-sounding idea of getting rid of pre-publication peer review, is that right?
1:11:21 LB: Yes, I did. It’s funny, I am someone who spends too much of my life on Twitter. And I’m always getting into political arguments about the hot button issues of the day on Twitter.com. And yet nothing that I have said has drawn me as much controversy…
1:11:36 SC: I was going to say, yeah.
1:11:36 LB: As the claim we should get rid of pre-publication peer review. This was in co-authored work with my friend and colleague, Remco Heesen, and the pair of us have this paper which was kind of, in some ways, a summary of the last few years of both of our research. We’d sort of… We’d realized that we were sort of both coming to a similar conclusion, so we put our heads together and co-authored this piece. And what we argue in there is that most of the things which pre-publication peer review is meant to do for us, there’s either no evidence that it does it or some evidence that it fails to do it. And so…
1:12:12 SC: So, just to be clear, the words “pre-publication” are important here. So tell the audience what you… How you’re defining the thing that you want to get rid of.
1:12:19 LB: So, right now, how do many fields, not in all fields, actually. It’s already changing in some fields, but in many fields the way new ideas get taken up is, not privately, but in relatively small groups, a scientist or a group of scientists produce a manuscript. They maybe share it with some of their friends, get feedback, and then eventually they will submit it to a journal. The journal will have an editorial board and some editors, that is to say, and reviewers. And the editor will send it to reviewers, the reviewers will decide whether or not they think it’s a good paper. In light of the reviewers’ advice, the editor will publish or not publish.
1:13:00 LB: So up to that point, the people who decided whether or not this idea is worthy of more attention from the scientific community are whatever the scientist’s immediate peer group, the editor, and whatever the reviewers think. This then decides what claims are put out there to try and garner more attention or not. Our point about this process is it’s very labor-intensive, it takes a lot of time for lot of scientists, it’s actually also, and this is not really relevant to our point, but it’s a fairly extortionate business. The scientific publication houses are like the most profit-making entities…
1:13:35 SC: I know, it’s a bit crazy.
1:13:36 LB: Yeah, because they just get a bunch of free labor from PhDs. Yeah, so a small group of people have put in a lot of effort to decide whether or not an idea is worth sharing. It then gets out there and gets whatever uptake it gets once it’s in a journal. More or less, our question is all of that labor in deciding whether or not something is worth putting out to the broader community and having them assess it, what does that achieve? Is there anything… Is there any firm evidence that that’s doing something good for us, rather than just directly having… Putting it out in some kind of archive, which is actually increasingly the case in physics and maths, that is how are things done. Actually, a lot of economics is kind of moving this way. Some fields are going this way, but just putting it up on some kind of publicly accessible space and going directly to the stage where the community are able to assess it as a whole.
1:14:24 SC: Philosophers also often have this idea of a working paper where there’s a draft that, for years, will be on their website, and people comment on it before they submit it to a journal.
1:14:34 LB: Yeah, in some sense it kind of… We want to make all of science work, operate more like working…
1:14:40 SC: More like philosophy, yeah.
1:14:41 LB: Yeah. I often think that the world would be better if it was more like how I think it should be, how I’m familiar and comfortable with it. So, yes. What people claim in defense of peer review are a number of things, and sort of the paper really consists in us going through the various rationales that have been offered for it, and trying to assess the empirical evidence for it. Now, the paper’s called Is Pre-publication Peer Review a Good Idea? And that’s partly to reflect the fact that right now, we think the evidence isn’t always as strong as we’d like it to be before making a clear assessment. So we are clear that to some extent, what we want to do here is just prompt people to like… Tradition has given us this expensive, very large bureaucracy we all spend a lot of our time on.
1:15:28 SC: Yeah.
1:15:29 LB: Don’t just inherit it and pass it on. Do we want to have it and what’s it doing for us?
1:15:33 SC: And you’re not even necessarily questioning peer review, but the idea of not letting anyone read the paper until it has been peer reviewed.
1:15:39 LB: Exactly, we’re questioning pre-publication peer review. In some sense, we say this in the paper, there’s a sense in which we want more peer review. We want more people to be able to access and evaluate the paper and for the opinions of the scientific community about the merits of a paper to reflect the judgments of more scientists. So, that’s in some way what we want, so we kinda link it to the open science movement. There’s a sense in which we want open peer review. In fact, open peer review, though, is a word for a different kind of thing, but related to what we’re proposing. Whereas, the claims that, for instance, peer review helps us sort the good ideas from the bad ideas or sort false claims from true claims, there’s really not much evidence that peer review is achieving that. And in some sense, it’s because science is hard and two people will…
1:16:29 SC: I don’t think it’s that hard.
1:16:29 LB: Yeah.
1:16:30 SC: No, okay. Actually, it’s very hard. Yeah. Just kidding.
1:16:32 LB: Two or three people deciding…
1:16:34 SC: Just kidding, audience.
1:16:35 LB: No, Sean Carroll is actually winking at me now. He tells me he’s not kidding, he means it. He just thinks you’re incompetent. But…
1:16:43 SC: Irony. Irony is everywhere here. You never know whether I’m serious or not.
1:16:49 LB: In some sense, the judgements of a couple of people who just happened to have got it first, they’re not necessarily gonna be great at deciding whether or not this really is an idea which was worth…
1:17:03 SC: We all have examples of bad papers that have gotten through peer review and good papers that have been stopped.
1:17:09 LB: Yeah, and there’s a kind of… There’s a… You don’t want to lean too much on this kind of flashy but one-time experiment, but illustrative, there’s this kind of famous experiment where people anonymized and lightly changed non-essential identifying details of classic papers, incredibly highly cited, resubmitted them for review, and I think it was like half were rejected on grounds of serious technical flaws.
1:17:16 SC: Right.
1:17:16 LB: And…
1:17:16 SC: Oh, in fact, there’s another study, I think that most highly-cited papers were rejected in one part, in the real world because the super highly-cited papers are gonna be a bit more daring, a bit more… It’s way easier, as a practicing physicist, I can promise. And I should say, parenthetically, I love refereeing in peer review. I appreciate the effort that goes into making papers better both from referees and from editors.
1:17:16 LB: I do not, and I want future editors assessing my work to know I don’t value you. I don’t value what you do…
1:17:16 SC: There you go. Having said that, if you write a paper that is boring but true, it’s way easier to get it published, than if you write a paper that is interesting and daring.
1:18:16 LB: Right, no, and that is one of the things we address in the paper. There’s a real reason to think that there’s a kind of small c conservative bias in peer review, precisely because, and this is, so to speak, inevitable, whenever we assess a new idea, we are assessing in light of what we already know, what we already believe. That’s inevitable, but the problem is is you’re preventing other people from being able to engage in that practice, unless a relatively small number of people decide it meets their standards, and I think that on net this is just a disadvantageous procedure.
1:18:49 LB: As I said, we are clear and I wish to be clear in this podcast that when I’m being not being cheeky, my real opinion is, there are plenty of times in a paper, where we say what needs to happen here is rather than assessing things according to people’s intuition about what this is doing, there needs to be proper sociological work on what’s actually happening, and very often people don’t do that, they just have a sense that, “Well, surely this is the sort of thing which peer review must do. It must be selecting for high quality work, right. Otherwise why would we do it?” And no, look at the evidence, it’s not clear it is.
1:19:20 SC: Yeah, look at the evidence. It’s just like the teaching of the scientific method. I mean, there is this idea that we can just think our way into the truth. But sometimes you have to look at whether it worked.
1:19:31 LB: Right. No, exactly, and peer review is something which I think was intuitive to many people that it surely can’t hurt and yet, and yet…
1:19:40 SC: It does filter out a lot of truly bad stuff, right?
1:19:42 LB: Yeah.
1:19:43 SC: That is a true fact. It’s the edge cases that we could improve a little bit.
1:19:48 LB: Yeah, even that is kind of complicated, because what’s the importance of filtering out truly bad stuff, because people put truly bad stuff on the internet anyway. So I hear…
1:19:56 SC: Yeah, I’ve heard.
[laughter]
1:20:00 SC: The internet.
1:20:00 LB: If you go to the wrong bit of the internet, there are people with false opinions, just putting them out there.
1:20:04 SC: They allow that?
1:20:08 LB: Yeah. Who knows? The government ought to do something. In the meantime… Please don’t… In the meantime, the reason it’s kind of important, scare quotes, to filter things out is the sense that being put in a journal is in some sense a stamp of approval. It’s a validation of that opinion. In a world without journals, what does it matter that one more person puts up a wacky enough very well validated opinion on the internet. What we say, the one thing we’ll be, this is one of the points where we acknowledge most problems for our work is, is we don’t… We think if you really think about [1:20:44] ____ journals, this is that thinking through the hypothesis, the possibilia kind of thing. If you really think about that world, precisely because there’s no apparent authority figure saying, “This is worth taking seriously,” a lot of the things which peer review is meant to do, we don’t lose anything by not having it.
1:21:01 LB: That said, there is something valuable about having a means by which non-scientists who might be interested in science can know where to look for the decent ideas. Now again, that peer review should…
1:21:16 SC: Or even scientists in a different field?
1:21:17 LB: Yes, yes. Oh, yeah, exactly. Or people who just are not expert in [1:21:19] ____ matter. Now, of course, that peer review is a good way of doing that kind of depends on you thinking it’s selecting for good ideas, which we previously argued against. But even if peer review isn’t doing that, it is probably a good idea for something to be doing that. And so to speak, right now, peer review is de facto how we’re doing it. Maybe a good thing to invest time and mental effort into doing would be thinking about how to do that in a better way.
1:21:42 SC: Okay. So that’s one example of the bomb throwing. Maybe to wrap things up, we should talk about standpoint epistemology.
1:21:50 LB: I was just thinking, one example of the bomb-throwing, what’s the other? Oh…
[laughter]
1:21:52 SC: I think that, look, I think the case that has been made indubitably over the course of this podcast is gaining true knowledge of the world, whatever that means, is hard, is sticky and is not quite as algorithmic as our high school educations might have led us to believe. And once you admit that, it opens up a whole bunch of things, like who decides and how do we decide where the truth is once we’ve found it? Is this kind of consideration… I know you’ll tell us what standpoint epistemology is, but how serious is the connection between this sort of social sciencey or even philosophical sounding idea and the real practice of science?
1:22:41 LB: Right, so this is a big one. So, standpoint epistemology is a sort of idea which comes out of… Actually comes out of Marxist theory, but it sort of gets more famously taken up in feminist theory in the late 20th century, and which nowadays in the early 21st century, for future listeners, is…
1:23:02 SC: You’re an optimist at heart.
1:23:05 LB: I think climate change…
1:23:05 SC: Like 23rd century listeners are gonna be, like, “Oh, thank you.”
1:23:09 LB: I just think climate change isn’t gonna kill us quick enough for that to be irrelevant to people. So, which nowadays has kind of become a heated topic of dispute outside of just the academic circles where it began. The broad idea, stated very generally and very loosely of standpoint epistemology is something like who you are, what social position you occupy makes a difference to what you can know and how reliable you’re gonna be at certain kinds of reasoning tasks.
1:23:42 SC: So is it really what you can know or how you know it or both?
1:23:46 LB: Depending on formulation, can be either or both.
1:23:49 SC: Okay, because what you can know, that’s a really radical claim.
1:23:56 LB: Well, funnily enough, I was going to say that what you can know is the easier one to defend.
1:23:58 SC: Oh, no, okay.
1:24:00 LB: I have, insofar as I’ve weighed into this debate at all, it’s from an old blog post, where I argued that if you look at kind of some of the precise formulation or the more detailed formulations of standpoint epistemology which exist in internet philosophy encyclopedias or places where people are trying to define it clearly for public audience, then it looks more or less just like the… It looks to be, so long as you believe that what kind of experiences you’ve had in your life make a difference to what kind of things you’ll be a reliable informant about, then some kind of standpoint epistemology is gonna be true, right? So, if you assume that in a very segregated town, but where the, say, and this is sort of 1950s America, the black people are going to be the janitorial and the maid staff for white people, then the black people are gonna know both about the black quarter and the white quarter, but the white people aren’t gonna go into the black bit of town.
1:24:53 LB: And so if you’re asking people about this town, its layout, where things are, and you’ve got limited time and resources, concentrate on the black people. They’re more likely to know, especially sort of the black working class people, they’re more likely to know about the time and layout of this, the structure and layout of this town, just ’cause they experience more of that.
1:25:10 SC: Yeah.
1:25:10 LB: That’s a very basic example. I’m not saying that that would be the most efficient way of finding out about this town, but there is some reason to think that if you are picking an informant in that kind of way, a thing, a reasonable correlate for you to look out for is the demographics of who they are and what kind of social position they occupy.
1:25:27 SC: I mean, it’s kind of inarguable that there are correlations between one’s identity markers and one’s knowledge.
1:25:32 LB: Right, exactly, and so, in some forms, like standpoint epistemology can more or less just be trying to think through the implications of that fact, and so, okay. In that case, that might make a difference if you’re doing certain kinds of ethnography or certain kinds of anthropology or sociology of this town. And so already while at a high level, it’s sort of obvious point. Like in practice, it might make a difference to how you carry… Your research design.
1:25:55 SC: Yeah.
1:25:55 LB: Who you wanna ask things. And so even that apparently trivial point can have implications. And [1:26:00] ____ of the implications, which are sometimes forgot, is there are classic examples which I not gonna remember quite right, but I’m gonna get you the broad structure of for early 20th century anthropology, where you’ll find that anthropologists was literally only interviewing the men in town, because they had some sense, it wasn’t just their own biases. They did have some sense that the men were the important ones who knows what’s up.
1:26:24 SC: Partly gotten rid of that.
1:26:26 LB: Yeah, yeah. So partly based on their biases but also partly based on what informants were telling them about the town. Except they were… You know, the male informants.
1:26:34 SC: The coherence theory of truths.
1:26:37 LB: But there were times in some of the rituals they were interested in, when the men and women would go into different places. And then, of course, if you only interview the men you literally just can’t find out about some of what’s going on here. And so it sounds trivial, but there have been times when people, even that simple correlational point has been forgot, and it has been forgot in ways which track people tending to trust more of those at the top of social hierarchies, and that just doesn’t… It does always work like that.
1:27:02 LB: And so I think even in its very simple form, standpoint epistemology can be contentful and important. I actually think another more controversial, but I would defend it as the same as this, is actually the Me Too movement. A lot of the… Like the slogan “believe women,” I think there’s a very reasonable explanation of it… Well, there are lots of reasonable explanations of it. But one reasonable thing it’s picking up on is if you are going to a workplace, and you wanna know who’s gonna reliably say who’s a sexual predator, who’s dangerous here, then you have to think of who has the authority to get away with things and what’s the distribution of hetero versus other kinds of sexuality, and it’s gonna be just mainly heterosexual men have power in office places.
1:27:39 LB: And so it’s mainly going to be women who are going to be the ones with relevant knowledge, and so there’s some strong reason to believe them, to trust them. They are the ones who know. And so again, that seems to me to fall out. That’s a really contentful point but it falls out of the quite basic thing. Now, let me say, that’s the point where I think standpoint epistemology is non-trivial, has implications of scientific and social life and which we could afford to take more seriously. There are other versions of standpoint epistemology which apparently do want to be saying a stronger thing, and so often a slogan would be, “A standpoint is an achievement.” And this, again, goes back to the original Marxist formulation, but it has some of the… Other more modern formulations also have this feature. Wherein, the idea that your social position gives you access to certain kinds of knowledge isn’t just meant to be an empirical fact that, well, these kinds of people who tend to have these kind of experiences have also a high rate, chance or something like that.
1:28:34 LB: But rather, there’s something which is a more substantial thing, which is like the proletariat standpoint, which is seeing things from the perspective of a worldview properly attuned to those people’s interests, or something like that. And seeing things from that worldview will tend to be producing more accurate or a more informed worldview. I think with that one, spelling out the content of that standpoint, in that sense, the achievement, the property proletariat or women’s or whatever standpoint, I think that becomes philosophically much more difficult, because you can just imagine, right, what is the true woman’s standpoint?
1:29:12 SC: It becomes a little normative also in a different way than simply epistemological.
1:29:16 LB: In some sense epistemology just tends to be normative. We’re trying to say, when we say what knowledge is, usually knowledge is taken to be a good thing. And so, if I tell you what knowledge is, I am in some sense saying that you’d have to…
1:29:27 SC: Fair enough.
1:29:29 LB: Succeed by way of belief. And this is only, it’s kinda trying to do that as well. People who have this stronger version will say like, the proletariat standpoint or the standpoint of oppressed women or something like that, that standpoint is more successful at knowledge gathering in order to produce…
1:29:47 SC: Okay, so there’s an empirical claim that could be tested.
1:29:51 LB: Well, I mean… I’m not a good spokesperson for this view ’cause I don’t believe it.
1:29:57 SC: Yeah, I know, but okay.
1:29:58 LB: But they might not wanna agree with that, because it depends by what you mean by empirically tested, because if these standpoints can come apart in what kind of things are gonna count as methods of validating knowledge, then you might worry that any test you do is gonna…
1:30:11 SC: It would have to be relative to that…
1:30:13 LB: Yes, be relative to the thing, and it will either be circular or biased against it.
1:30:16 SC: Okay.
1:30:16 LB: And so…
1:30:17 SC: Well, then it becomes less interesting to me. I mean, I…
1:30:19 LB: Right, yes…
1:30:21 SC: If it’s just true by construction or irrefutable, I guess. But just to make it super duper clear, none of this is going to claim that, whether you’re black or white, or male or female or a cis or trans affects whether the universe is expanding.
[laughter]
1:30:25 LB: Oh, no.
1:30:25 SC: Or whether you know the universe is expanding, right?
[laughter]
1:30:25 LB: No, no.
1:30:25 SC: It’s not like who your identity is affects the truth about the world. That’s just not the point.
1:30:25 LB: No, right? So I guess some brave soul might wish to argue as much, but typically the kind of claims people are concerned with are the sort of things I spoke about. So it’s gonna be about power relations in the workplace, or…
1:31:07 SC: Human experience.
1:31:08 LB: Human experience, like where it’s safe to go in this town if it’s a sundown town, that kind of stuff. And so usually if you look at… This is the thing, for all of the kind of cultural wall controversy around it, often, not always, but often, if you look at the kind of claims standpoint epistemologists are interested in, it’s usually sort of independently plausible that the sort of people they’re saying have an advantage of some sort in knowledge gathering, would have an advantage about knowledge gathering in that case. I in no means wish to say that’s in all cases uncontroversial. In many cases there are gonna be controversies, but there’s a kind of core set of examples which people will tend to find plausible, and then it’s just how far can you expand that, rather than whether or not there are any. People don’t usually expand it all the way to like the fundamental forces of the physical, the cosmos or something like that.
1:31:55 SC: Well, I think that I bring that up intentionally ’cause I guess my first exposure to ideas in this universe was as an undergraduate taking philosophy courses for the first time. And there’s sort of two extremes that are equally silly. One extreme, silly in my view anyway, one extreme is that super radical view where who you are literally changes the physical truth of the world, and the other is that must be what you mean to be claiming if there’s any relationship between who you are and what you know about the world, right? And I think that the lesson of all that we’re saying here is that there is a difficult but good place to be in the middle of these claims.
1:32:34 LB: Yeah, and I know that that way of phrasing things actually… And I think you’re quite right that that is often how the debate is carried out. And that way of phrasing things actually comes back to something we discussed right at the beginning, which is a thing which I think philosophical reflection tends to want to be more careful about is separating out claims about how you know, and claims about what’s true. And so very often some of the most heated debate about, for instance, standpoint epistemology, people attribute to standpoint epistemologists who are saying who you are makes a difference to what you know, the claim that who you are makes a difference to what’s true.
1:33:07 LB: And those are very different claims, and they should be. And one of them is far more plausible than the other. I mean, and so to speak, standpoint epistemologists are paying a special attention to social positions, especially social positions in the hierarchies of relationships of oppression, but in some sense everyone is gonna be a kind of standpoint epistemologist, something like presumably having received a PhD in Physics puts you in a better position to understand and reason.
1:33:34 SC: I probably know something.
1:33:35 LB: Yeah, exactly right.
[chuckle]
1:33:36 LB: So everyone thinks that some life experiences are gonna be pertinent here. The standpoint epistemologists are just drawing attention to a class of experiences of being oppressed in certain kind of ways, which they say make significant differences in these areas. So sort of the basic idea behind standpoint epistemology, it should be sort of no more foreign to you than saying that… If someone says that someone with a Physics PhD knows more than something without a Physics PhD about whether or not the universe is expanding, they’re not claiming that whether or not the universe is expanding depends on whether you have a Physics PhD.
1:34:05 SC: That’s right, yeah.
1:34:06 LB: It’s the same thing.
1:34:07 SC: Well, this is good, because it makes it sound like I had organization in mind when we started talking, and that always makes me look good even if it’s just kind of random.
1:34:14 LB: I believe it.
1:34:16 SC: Alright, Liam Kofi Bright, thanks so much for being on the podcast.
1:34:21 LB: Thanks for having me, Sean.
[music][/accordion-item][/accordion]
So happy that you had Liam on the pod, I saw you two interacting on Twitter a few times and was really thrilled about that possibility. You are both among the most humble people I know on the internet.
I really enjoyed this episode, thanks. I’d love even more getting into the weeds discussing ideas about truth — more on the deflationary viewpoint, for instance, because that was new to me. My big takeaway was thinking about teaching the scientific method differently — I’m a pro-science educator and want to incorporate some of this into my teaching.
It seems like the options arriving at “truth” using one of the four possible approaches doesn’t place enough value on inductive reasoning. Usually we describe our beliefs about the truth or falsehood of a statement in terms of a prediction of the likelihood that new information we have yet to discover will change our beliefs about the current state of the world. None of these models correspondence, coherence, pragmatist, or deflationary approaches really does any work to describe the probability that new information we have yet to discover will change our current beliefs.
There has been a major breakthrough in creating a commutable algorithm for logical induction. To me this seems much better at defining truth in terms of probabilities. It does a much better job of making predictions about the likelihood that new information will change our beliefs about the state of the world. I think that talking to some of the researchers who contributed to the paper on logical induction would be really valuable because their theory seems way more practical than what I heard discussed during this podcast.
https://intelligence.org/files/LogicalInduction.pdf
Scott Garrabrant, Tsvi Benson-Tilsen, Andrew Critch, Nate Soares, Jessica Taylor. With contributions made by Abram Demski.
Pre publication peer-review? As a layman, I have been subject to popular physics books the have a sharp US Republican idea, starting on Chapter 3 in a general consumption physics book, (a book on ’emergent theory), a kind of sneak attack. Labor intensive? great. It weeds out the trouble makers that have a specific political or ideological view. The science propaganda machine is alive and well, and idiots like me rely on Sean Carrol, on Liam Kofi Bright. Industry of all stripes inject pro-industry ideas into science, attempting to sway governments, and the broader public. Big money on the line.
Go ahead and put it on your blog, put it on line somewhere. Put it into a discussion with. a peer enthusiastic with your paper’s view. Destroy more of institutions? Don’t they run public university research, using public grad students and professors, for private gain?
On quantum mechanics .. What if a camera was pointed at electron.. The result is on a screen on the camera.. Another camera is pointed at a honeycomb shaped sphere.that will show the screen in some mirrors and not in others . So it can record observations and non observations ..so it is both looking and not looking ..just a thought
Sean (or anyone):
Multiverse: Different far-out areas with different properties affecting the way local physics can work. Does anyone discuss how there can possibly be borders/boundaries between such pocket universes? Would the strings/branes containing the extra dimensions simply be of different geometries?
How can two universes be “next to” each other, and would there be strange interactions at the border (or no interaction)?
As I understand it, the difference in the properties of a given universe is dependent on the geometries that are holding the tiny extra dimensions. If this is right, are the proposed neighboring universes “next to” each other in any real sense, or can there be no interaction — one of them is inside a black or white hole so there would be an information firewall or something.
Really enjoyed this episode! As for The Truth, one of the best laughs of my life was when Lionel Hutz explained to Marge Simpson that there’s “the truth…” and then there’s “the truth!” Yup. By the way, “This sentence is yellow.” 😉
Many-Worlds in praxis: So, today I used the Quantum Random Number Generator app by CrimsonLabs to call up one of my personally curated “Top 20 Songs Of The Year” playlists of which I have one for every year from 1966-2019. The API was successful and in this world, I got 1991. It was somehow very satisfying in a cerebral way to understand that Many-World variations of me were playing the Top 20 of the Year for all the other 52 years at the very same time, trusting my parallel selves fulfilled the deal.
Pingback: Mini-Heap - Daily Nous