Episode 68: Caldwell’s “Clarifying Popper”

Unofficial, reviewed transcript. It may still contain mistakes; check the podcast when wording matters.

Speakers are labeled by first name where identified. Unidentified-speaker labels may refer to different people. Bracketed question marks indicate uncertain wording.

Transcript

[00:00:11]  Bruce: Welcome to The Theory of Anything podcast.

[00:00:13]  Unidentified speaker: Hey, Peter.

[00:00:14]  Unidentified speaker: Hey, Bruce.

[00:00:15]  Unidentified speaker: How you doing?

[00:00:16]  Unidentified speaker: Good.

[00:00:17]  Bruce: So let me go ahead and try to summarize Bruce Caldwell's paper, Clarifying Popper, which was in the Journal of Economic Literature in 1991. So this is something that Brian Boutwell sent to me when he had questions about how to apply Popper's epistemology to his field, which would be one of the softer sciences. And this paper is really about maybe the softest of sciences, or maybe not even a science, which is economics. How do you apply Popper's epistemology to economics? Economics in particular creates some fairly severe problems with trying to apply Popper's epistemology because you have so many causes and so many things going on at the same time that when you have a theory in economics, there's basically no way to falsify it. So, Caldwell's paper does, I think, a really good job of clarifying what Popper's epistemology is. Although, Peter, you did find that critical piece of this paper, which we'll also discuss, but I think he does a pretty good job.

[00:01:22]  Bruce: And even though apparently this other person disagrees with me, I think it's a fairly good summary of Popper's epistemology. And I also think that he does a fairly good job of explaining why it's hard to sometimes apply Popper's epistemology to certain fields of study. So let me take you through kind of his argument as I understood it. I'm going to try to summarize. So Caldwell makes a distinction between falsification and critical rationalism, though he admits that Popper never does this. So this is, when we talk about in this paper, falsificationism versus critical rationalism, understand that this is a Caldwell thing, not a Popper thing. But I think that distinction turns out to be valuable in trying to understand his point, okay? And in fact, I think it's a valuable distinction, if somewhat confusing, because usually those two are used as synonyms.

[00:02:11]  Bruce: But I think it's a valuable distinction that's worth bringing out, that there is kind of two critical rationalisms. There's the stronger falsificationist version and there's the weaker critical version that doesn't necessarily rely on experiments and falsification. So falsification is an epistemology under this way of speaking. That's built up around the demarcation criteria of falsification. And it demarcates empirical science versus everything else, which might include both metaphysical theories, which could be [of] value, but also pseudoscience, which is not [of] value. Now, I've mentioned this in our Popper Without Refutation episodes, that the demarcation isn't really between science and non-science. It's really empirical science and non-empirical theories. And I've also mentioned that it doesn't even really differentiate between pseudoscience versus valuable metaphysical theories.

[00:03:01]  Bruce: I mean, a metaphysical theory might in fact be a scientific theory, but we just don't have instruments yet to be able to test it. So it's really just empirical, as of today, science and non-empirical. So Popper's big breakthrough was realizing that verification won't work as a demarcation criteria with empirical science versus non-empirical theories. Because if we adopt verification as the basis for empirical science, then the theory unicorns exist must be considered valid science, because who knows, maybe someday we'll verify the existence of unicorns somewhere in the universe. So many people might see the theory unicorns exist as empirical. Because I'm using the word unicorns, that's going to turn some people off, but maybe I could say platypuses exist prior to the existence of platypuses. Isn't that an empirical theory? I can go out, I can find a platypus, and I can empirically verify my theory.

[00:04:08]  Bruce: Okay, so it seems like maybe that should fit the concept of an empirical theory. And in a certain loose sense, if you want to take the concept of an empirical theory to include something like the platypus, I guess I wouldn't object. But I would object to saying that that's a good scientifically empirical theory, even though platypuses actually do exist. And the reason why is because there is no way to test this theory. For the exact same reason why you can't test unicorns exist, right? Presumably, unicorns don't exist, at least not on planet Earth. But how would you ever know, right? I mean, like, you would have to literally go search the entire world, you know, searching absolutely everywhere. In theory, you would still not have verified [as spoken] your theory because they might exist on a different world, and then you'd have to search the whole universe.

[00:05:00]  Bruce: And basically, any theory that's based around existence of something or that requires verification can never really be used as a scientifically empirical theory because you can't build experiments for it, okay? So this was kind of Popper's big breakthrough according to Caldwell, although that fits exactly what I said in my Popper Without Refutation episodes also.

[00:05:20]  Unidentified speaker: So I agree with him on this.

[00:05:22]  Bruce: So Popper did not believe that there was a scientific method of mechanical knowledge creation, since one of the main ways we create knowledge is through creativity and coming up with bold conjectures. But Popper did create a methodology for refuting theories by laying out conventions or rules of what we're not allowed to do, such as ad hoc saving a theory from refutation, okay? Here's a quote from the paper. “Falsifications are always more interesting than corroborations, for they force scientists to reexamine the theory and test situation to see what went wrong. It is always possible for scientists to protect a favored theory by blaming a refutation on something else. Popper's response is to lay down certain methodological prescriptions, against which he terms conventionalist stratagems or immunizing stratagems, whose sole purpose is to protect theories from being refuted. [Wording unclear.]

[00:06:17]  Bruce: Popper also uses the [phrase] ‘ad hoc theory adjustment’ to refer to any change in a theory designed solely to save it from refutation.” However, this seems a little unrealistic. Going back to the paper, quoting: “Even in situations where clean tests are difficult, scientists should specify in advance what sorts of results would lead them to abandon their theories.” Now, I'm not sure I agree with that statement, but this is what Bruce Caldwell says in the paper. The problem here is that even if I take into consideration everything that I just said about the problems of verification, and so I'm going to concentrate around refutations instead, the simple truth is that it's just not obvious whether an observation refutes a theory or not. For the [Duhem–Quine] problem, because of just the fact that you can argue over whether the test was valid or not, that it turns out to be difficult to ever actually show this is a falsification.

[00:07:14]  Bruce: And you end up actually needing a second theory to explain it before you can actually see that an observation should count as a falsification. So the most important way we avoid ad hoc saving theories is via the rule that you're not allowed to add auxiliary hypotheses to your theory, except when their addition increases the degree of a theory's falsifiability, by which we mean when its empirical content grows. So it's okay to save a theory, according to Popper, by adding to the theory, which is what people do when they ad hoc save a theory. It's just that he makes the rule, if you're going to save this theory with a new theory, that new theory must itself be testable. It has its own unique tests that we can then go out and try to refute it with. If you aren't doing that, then we consider this theory an ad hoc save. Even if it's the truth, we consider it an ad hoc save.

[00:08:14]  Bruce: It's up to you to go figure out how to make it into a testable theory. Caldwell argues that Popper's methodology is prescriptive rather than descriptive. So he really emphasizes this idea. Caldwell really emphasizes the idea that Popper was not saying this is how science is done, but this is how science should be done. I'm just not sure I even agree with Caldwell on that. It is true, I can find quotes from Popper where it is obvious that he's making it prescriptive. He's saying this is the way things should be done. But there is so much in Popper that seems to be far more descriptive, where he's saying, look, I'm trying to describe what it is scientists are actually doing. And I have always seen Popper as less [prescriptive?], although there is a prescriptive element and there is a descriptive element, I've seen it as 80% descriptive and 20% prescriptive, where Caldwell clearly sees that reversed.

[00:09:11]  Peter: It seems like, as we've discussed before, a lot of what Popper is talking about is more of an attitude than a methodology.

[00:09:18]  Unidentified speaker: Yes.

[00:09:19]  Peter: Is that a fair way to put it?

[00:09:21]  Bruce: It is, and that's why Caldwell sees it as prescriptive. Okay, because Popper does equate this to a certain kind of attitude at times.

[00:09:31]  Unidentified speaker: Okay.

[00:09:31]  Bruce: Caldwell is taking the stance, look, if I'm prescribing an attitude, that's prescriptive, not descriptive. So it may be that scientists don't have a critical attitude. Okay, but they should. Okay, now the reason why I disagree with Caldwell on this is not because that statement was wrong. As strange as that may sound, right? It's just that I'm not sure it matters if scientists actually have a critical attitude or not that much. Because I think the institutions of science make it so that even if every scientist had a non-critical attitude that was not following Popper's prescriptions, I think science would work anyhow because they have no choice under the set of institutions but to move to empirical tests, and they end up following these conventions that Popper is giving anyhow. A real-life scientist will ad hoc save a theory, exactly like Popper says not to, and every other scientist that disagrees with them will call them out as making an ad hoc save, and so they will be forced to figure out, well, how can I test this ad hoc save?

[00:10:45]  Bruce: At which point, it's no longer an ad hoc save. Because of that, I see this less as a prescription. Yes, Popper's being prescription [as spoken]. Yes, Popper is saying, you should have a critical attitude towards your theories. But I don't know if it matters if you actually do because your opponents are going to force you anyhow. And that's the beauty of science, is that critical rationalism is descriptive of how a scientist must ultimately behave if they're going to be successful in this institution. Because the institution exists in such a way that you have really no choice but to follow Popper's prescriptions, even if you don't want to, even if you don't intend to, you will need to. And that's why I see it as more descriptive and less prescriptive. But I admit it's got elements of both, and I think he's admitting it's got elements of both also.

[00:11:41]  Unidentified speaker: Okay.

[00:11:42]  Bruce: So I think Popper's epistemology is first and foremost descriptive, but Caldwell is looking at it from the point of view of economics, where falsification really has played little role because it's so hard to falsify economic theories. When I try to put my brain into his space, thinking of it in that way, I think I understand why he's saying it's more prescriptive than descriptive. Because in economics, if you're a fan of Popper, you pretty much entirely see Popper as prescriptive. We should figure out how to make our theories falsifiable, but I don't actually know how to do that today, right? And so because of that, Popperians that are economists would tend to see Popper as almost entirely prescriptive because they don't know how to apply Popper to their field. Thus, Popper's rules are aimed both at action and attitudes of scientists, says Bruce Caldwell in the paper.

[00:12:42]  Bruce: Now, Caldwell notes, for example, that praxeology is at odds with Popper's falsificationism. One of my friends, Logan Chipkin, he is a huge fan of David Deutsch and Popper, but also of praxeology. He sees those as two of three of the most important theories. He calls them the big three. And yet, from a certain point of view, praxeology is—outrightly claims to be non-falsifiable, right? That it is impossible to test. They try to come at it from the standpoint of it's just intuitively obvious that it's true, which is something you're not supposed to be doing under Popper's epistemology because your intuitions can be completely wrong. That's why we want to test our intuitions and use them only as sources of conjecture, not as sources of truth. So, but I think a lot of people would say praxeology is interesting, right? And it kind of makes good sense to them.

[00:13:41]  Peter: Now, praxeology is associated with the Austrian School of Economics. Is that right?

[00:13:47]  Unidentified speaker: Yes, that's correct.

[00:13:48]  Unidentified speaker: Okay.

[00:13:48]  Unidentified speaker: Okay.

[00:13:50]  Bruce: So it's kind of their mathematical proof, so to speak, of how they arrive at their conclusions. They start with what they think are these really obvious intuitive truths, and then they say, okay, well, then I can now derive this, et cetera, et cetera, and they kind of move out from there to come up with all their economic theories. And yet, at some level, praxeology upfront tells you, this is just true, there's no way to test it. But it's just obviously true, which is a very noncritical attitude, a very at odds with Popper's epistemology critical attitude—noncritical attitude. On the other hand, how exactly do you apply Popper's epistemology to economics when there's no school of economics out there other than praxeology that's more testable, right? I mean, given the fact that there is no known way to directly test any set of world views on economics, I would definitely say praxeology at a minimum needs to be treated as a completely valid metaphysical theory, right?

[00:14:59]  Bruce: And if its competitors can't be tested either, then that's kind of okay that it can't be tested. But this still leads to, well, then how do you apply Popper's epistemology to economics? So Hayek, by the way, Caldwell quotes Hayek as saying that he finds it difficult to falsify economic theories. And he quotes Hayek throughout, by the way, talking about how he kind of doesn't believe in falsification because he feels it has no real role in economics. So this has led economics to fall into two camps, according to Bruce Caldwell. Those that see falsification as a yet-to-be-attained goal for economics and those who reject falsification entirely because it isn't useful in economics. So here is Archibald against falsification that Bruce Caldwell is quoting as an example. He says, “My problem is due to the circumstance that much activity in economics is devoted to empirical work with hypotheses that do not satisfy Popper's falsification, falsifiability criteria.

[00:16:09]  Bruce: Should I judge that the hypotheses and the associated empirical work fall short of some ideal scientific standard, or should I judge that the methodology calls for what is impossible and that it should therefore be altered? My own judgment is that many of the irrefutable hypotheses in economics are important, that they are incurably irrefutable for good and fundamental reasons, and that the activity of comparing them with observations is useful—[too] practically useful to be acceptably called metaphysics, so that it is the demarcation rule that should give way.” Okay, that's a really strong statement, and I think it's exactly the type of challenge that we want, right? We're starting with critical rationalism and falsification, and we're showing that economics has something useful even though it can't be falsified, and it even does things useful with evidence, even though it can't be falsified.

[00:17:11]  Bruce: Therefore, this is possibly a counter example to Popper's epistemology, which is, from a Popperian standpoint, exactly what we want. We want to try to criticize Popper's epistemology by finding counter examples to it, okay? While we do eventually... Now, this is Caldwell's point again. While we do eventually falsify after a fashion some economic theories, say Malthusian population theory, it is practically impossible to do it except over a very long period of time, making falsification useless in determining if Malthusian population theory is worth taking seriously upfront or not. Okay, so Malthusian theory is the one that David Deutsch criticizes, where we're going to grow population and we can't grow food, and so eventually we're going to have mass starvation. That did not turn out to be true at all. At this point, most people would consider that a falsified theory of economics.

[00:18:09]  Bruce: And yet, if you're living at the time that he came out with this theory, you wouldn't have any realistic chance of coming up with an experiment to falsify it. And in fact, Deutsch doesn't. Deutsch criticizes the theory not using falsification by saying, well, you're not taking into consideration knowledge growth. So it looks like that, in fact, they're right, even in a case where we could say, yes, you can, Malthusian population theory has been falsified at this point. Therefore, this fits Popper's epistemology. That would have done you no good at the time, right? And ultimately, the best criticism of it had nothing to do with falsification.

[00:18:50]  Unidentified speaker: It was just a criticism.

[00:18:52]  Peter: Apparently, his predictions were very accurate. And in one sense, he was quite a good scientist, but his worldview was just missing a piece of the puzzle there, which was of knowledge growth [hesitation unclear].

[00:19:10]  Unidentified speaker: Right.

[00:19:12]  Bruce: Now, it's tempting to just dismiss falsification as not relevant to economics or possibly social sciences, but this violates another idea from Popper, namely the principle of the unity of science, where Popper claims that we should, quote, view that all theoretical or generalizing sciences make use of the same method, whether they are natural sciences or social sciences. To give an example of why this view is problematic, Caldwell takes a look at Popper's rationality principle, where Popper claims it only makes sense to assume that what someone says or does is rational. I would note that this is equivalent to my admission to always read everyone as charitably as possible. That is to say, you should never read someone as merely saying something incorrect if there is any way at all to read them as saying something that is correct.

[00:20:02]  Peter: Is that the rationality principle very similar to praxeology? Is that kind of just a different way of saying the same thing?

[00:20:11]  Bruce: They are related, yes, because economics is making the assumption of rationality, not just praxeology, but I think all forms of economics start with the assumption that we're rational actors and that the choices we make are rational. And even though we know that's not ultimately true, you almost have to start with that assumption when you're trying to do economic analysis, because to not start with that assumption is a much bigger problem. And this is actually where Caldwell is going with this, so I don't want to give too much away of his argument yet. So Popper's reasoning is simple. It's just too easy to declare anyone confused or irrational anytime you want. It's the quintessential example of the Deutschian bad explanation, because it can always be used, therefore you can't learn anything by starting with the assumption that someone is merely behaving irrationally.

[00:20:59]  Bruce: However good this argument is, it has a rather obvious problem that we know sometimes people are irrational, say inconsistent or contradictory things or are confused. So starting with the assumption that they are saying something rational could be incorrect. Now Popper admits that this is the case. He says, you know, of course what I'm saying is rationality principle, it's false. I mean, like, we know it's false, but I'm suggesting that you apply it anyhow even though you know it to be false. So this leads to an obvious contradiction for Popper. If we are to reject a theory for being false, which is something that Popper says throughout his writings that we should reject theories that are false, then the rationality principle is clearly false, so we should be rejecting it. So the rationality principle is clearly metaphysical in nature, as per Popper, and it goes against his prescription of striving for falsificationism.

[00:21:55]  Bruce: Doesn't this then mean that either we have to give up the unity of science, because social sciences would then have to be said to... If social sciences rely on the rationality principle, or if economics relies on the rationality principle, doesn't that mean that they are not using falsificationism and not using Popper's epistemology, and that therefore we either have to give up the unity of science, or we have to give up the rationality principle? Now, for Popper to say, well, no, we can't give up the rationality principle, it just makes sense to use it, even though it's known to be false. Isn't that an example of an ad hoc adjustment of the kind that Popper says you shouldn't be doing, because it can't be tested, and in fact, he's admitting that it's wrong. So it seems like we've found a contradiction in Popper, where we've now shown that his epistemology is incorrect in some important way.

[00:22:45]  Bruce: So quoting the paper again: “If one takes Popper's falsificationism seriously, a conflict arises between the doctrine and this analysis, and it results in social science explanations that are either metaphysical or ad hoc or inconsistent with explanations of natural sciences.” So Caldwell resolves this apparent contradiction in the same way that Deutschians will be familiar with, because it's the same way David Deutsch does it in his books. Remember, this paper predates Deutsch's books. And I don't think Deutsch had any access to Caldwell. So this is a solution that Caldwell came up separate from Deutsch. And in fact, I don't think Caldwell came up with it either. I think based on that article you gave me, Peter, that this has been around for a while, maybe has been thought of independently. So he resolves this contradiction by pointing out that experiments and falsification is just one type of criticism.

[00:23:41]  Bruce: And that one can think of critical rationalism not as saying that you have to make your theories falsifiable, but you have to make them open to criticism. From this viewpoint, the rationality principle, despite being non-falsifiable and known to be wrong, and thus ad hoc, still makes perfect sense because it makes it easier to criticize theories. You start with the assumption that the person is being rational from their viewpoint, rather than simply declaring your opponent confused or irrational upfront, which then becomes uncriticizable. So, based on this, Caldwell says, “It is vaguer, no doubt, and inferior in many respects, and it's irrefutable—or lack of testability is its great vice. [Wording unclear.]” “But as long as a metaphysical theory can be rationally criticized”—this is actually, sorry, quoting Popper—“I should be inclined to take seriously the implicit claim to be considered tentatively as true.”

[00:24:42]  Bruce: So Popper kind of had started off his career thinking that falsificationism was the right epistemology, and then kind of had an epiphany at some point, and Caldwell quotes him to prove this, that he came to the realization that metaphysical theories were still of value, that they could still be criticized, and that we ought to still use the same epistemology on metaphysical theories, non-empirical theories, as we do with empirical theories, and that you simply replace experiments and falsification with the concept of criticism. And I think this is exactly what David Deutsch is getting at when he talks about criticism is just—sorry, observations from an experiment are just a kind of criticism, and that actually there's a lot of other criticisms that are more important in that we use them first, and he uses the example of grass cures cancer.

[00:25:39]  Bruce: There's no reason to even go to an experiment because you've got good explanatory reasons why grass should not cure cancer. And so you eliminate all these bad theories upfront using criticism before you even get to the experiment that is then going to be used for falsification. And by rethinking this, this generalized version of falsificationism, that's what Caldwell is calling critical rationalism. So that's why he makes a distinction between the original epistemology, which was about the demarcation of science and about falsifying through observations, and this more generalized version, which he calls critical rationalism, which is where you use criticism to try to eliminate theories instead of specifically observations and experiments. And he says, by doing that, we can now remove the conflict. We can show that economics follows critical rationalism, even in a case where it's using something like the rationality principle.

[00:26:39]  Bruce: It just makes perfect sense. It doesn't matter that it can't be falsified. That's okay. It's the reason why we apply the rationality principle to economics is because it makes it easier to criticize economics. Whereas if you didn't apply the rationality principle, there would be no way to criticize economics, and there would be no way to judge between theories. So this is what Caldwell does, and this is how he resolves the tension here. What is the connection between critical rationalism and falsificationism?

[00:27:10]  Unidentified speaker: It's evolutionary epistemology.

[00:27:12]  Bruce: So in a past episode, I talked about Donald Campbell's theory and Popper's glowing endorsement of the theory, where he claims that all inductive... Campbell claims that all inductive achievement comes through falsifying theories, but by all forms of variation and selection. [Wording unclear; the following discussion generalizes beyond falsification.] Campbell uses the term blind variation, but I argued that he really meant variation. This is the ultimate generalization of Popper's theories that Popper believed combined all forms of knowledge creation, whether biological evolution, scientific theories, non-empirical theories, history, animal knowledge, et cetera. He believed all of them came through evolutionary epistemology. So here's now going back to the article. “Evolutionary epistemology provides the epistemological foundations for critical rationalism.

[00:28:02]  Bruce: This doctrine emphasizes the similarities between the growth of animal, including human knowledge, and the evolution of species. Bold conjectures are analogous to blind variations”—to take the term from Campbell rather explicitly here—“which would be mutations in nature. The process of criticism is analogous to the process of natural selection. Evolutionary epistemology provided an empirical basis for epistemology in processes found in nature, as well as an argument for realism, the survival of both ideas and organisms depends on their fit within the environment and the assumption of an existing environment is consistent with realism.” Okay, so that was all quoting Bruce Caldwell. Now, again, from Caldwell's article, the goal of analysis is to subject theories to an optimal amount of criticism, but most important of all, the level of criticism will depend on the problem to be solved and the nature of the material under investigation.

[00:29:03]  Bruce: Empirical theories are the strongest, and whenever possible, they should be used. Furthermore, it is generally the case that immunizing strategies should be avoided, but at least in the special case of what we talked about situational analysis and the rationality principle, one is able to criticize more severely and obtain more fruitful criticisms if one blames the model rather than the rationality principle whenever a falsification occurs. So that's why it's okay in some cases to use something like the rationality principle, even though it's known to be false, even though it's known to be ad hoc saved. It's because we're trying to apply this idea that we want to maximize criticism and subject our theories to as much criticism as possible, and in this case, that's what you do to be able to maximize criticism. By the way, I really like the fact that Caldwell says that the ideal is still falsificationism.

[00:29:56]  Bruce: That's something that I want to get back to in some future podcast. Now, here's the problem. Defining critical rationalism in this way leads to a new problem, namely that whereas criticism by repeatable experiment are an objective thing, because they're intersubjective, anyone can go repeat the experiment, criticism by mere criticism becomes subjective. This is something that I've accused Deutschians of in the podcast more than once. So Hands, that's a philosopher, he puts it this way. He says, “The real problem of critical rationalism, when understood in this way, is not that one can say very much against it, but rather that one cannot say very much with it. Critical rationalism is a view which seems to be palatable by virtue of its blandness. The epistemological analog of the ethical mandate to quote, ‘live the good life.’” “According to this objection, it is not enough”—this is Caldwell now—“not enough to say that the theories must be criticizable and that scientists must exhibit a critical attitude.

[00:31:05]  Bruce: At a minimum, criteria must be provided so that one can tell when such conditions are being met.” Okay, many good theories are defended dogmatically. “Furthermore”—this is Caldwell again—“it is often hard to judge who is being dogmatic in a debate. Besides one's opponents, how does one identify what constitutes a genuine critical attitude versus, say, a curmudgeonly one? When it comes to criticizability, similar problems appear. Aren't all theories criticizable? After all, even the uncriticizable theories can always be criticized on the grounds that they aren't criticizable. And just how is criticizability to be identified anyway? Perhaps more important, are there any criteria by which the effectiveness of a criticism might be assessed? How does one decide what constitutes an optimal amount of criticism? And when or when a criticism has been successful?

[00:32:02]  Bruce: Clearly not all criticisms should carry the same weight. Otherwise, a proof of logical inconsistency would have to be considered on equal footing with the following criticism provided by one of the referees:” “I do not like the Dr. Fell. The reason why I cannot tell. But this I know and I know it well. I do not like the Dr. Fell,” which I thought was funny. So Campbell [name as spoken] is basically saying, okay, I'm accepting that Popper has a generalization of critical critical rationalism that you don't have to use falsification and that economic can fall under that. And instead we're trying to subject to maximum criticism rather than trying to falsify it. But he's being very clear that this is a problem still, because now you've moved into this realm of subjectively deciding what criticisms matter if they even were valid criticisms or not.

[00:32:59]  Bruce: And Caldwell unfortunately kind of leaves the paper here. He admits that he doesn't have a solution to this problem, that if you look at critical rationalism in this way, then you really just have to kind of do your best. You're going to do criticisms, everybody criticizes, and you do your best to be open to criticism, but really it just sort of becomes subjective what the best theory is, whether you have really properly opened yourself to criticism or not. I mean, nobody who's dogmatic thinks they're dogmatic, right? A person who's being dogmatic thinks they're open to criticism. They just don't think any of the criticisms have been very good. And if we're in this subjective realm of what counts as a good criticism, there's no way to really refute that point of view, right? They can always just say, no, this is why I think this criticism is wrong, et cetera.

[00:33:53]  Bruce: Okay, so falsification’s weakness compared to critical rationalism, understood this way, is that it's too stringent for many kinds of knowledge and theories and ideas. So going back to Caldwell now: “Falsification is anything but complacent, always seeking out new ways to prove a theory wrong. It is not content with even high corroboration. Its weakness is that, in its zeal to avoid complacency, it lays out criteria that even true theories may not be able to satisfy, given all the problems that are associated with hypothesis testing. There is the danger, too, that falsificationism might be applied selectively to defeat one's opponents. This is why its best use is probably as a check on one's own enthusiasm, rather than as a weapon for the elimination of one's rival's theories.” So Caldwell's conclusion is that social sciences and economics will follow critical rationalism rather than falsificationism, and that's okay, even if it is not as good, and it's inferior in some way.

[00:35:06]  Bruce: And this is really, at least the way I read the paper, kind of where Caldwell leaves things. I thought it was a fascinating paper, and I actually would recommend people go get a copy of it and read it, see if they agree with my summary of it or not. And I feel like this is actually a jumping-off point for several other really important things I would like to say that we don't have time to do right now that we'll do as a separate podcast someday. But I think that it gets to really, to the heart of what I see as the biggest problem with critical rationalism is exactly what Caldwell says, that it comes in two forms, falsificationism, which is quite stringent and very objective, and then this non-falsificationist version, which it's hard to figure out how to use in real life or to even use to say that you've eliminated competitors. And it's just too general.

[00:36:03]  Bruce: And therefore, it's difficult to know what to do with it. I kind of agree with Hands that it just doesn't say anything useful in this format, or at least arguably you could say that in some cases.

[00:36:17]  Unidentified speaker: I think we just talked about the rationality principle and how it actually is saying something useful in some cases.

[00:36:22]  Bruce: I don't mean to be dismissive of it. I just mean there's a very real problem there with this version of critical rationalism that really has never been resolved. Now, Peter, you brought up an article critical of Bruce Caldwell's article.

[00:36:37]  Unidentified speaker: Can I ask you one question first?

[00:36:39]  Unidentified speaker: Yeah.

[00:36:40]  Unidentified speaker: Big picture.

[00:36:41]  Peter: Would you say Caldwell is concluding that critical rationalism is very useful in social economics and other social sciences?

[00:36:50]  Unidentified speaker: Yes.

[00:36:50]  Peter: Or would you say he's not?

[00:36:52]  Unidentified speaker: Okay.

[00:36:52]  Unidentified speaker: I'd say he is.

[00:36:53]  Peter: Say he more is. [Brief reply unclear.]

[00:36:54]  Bruce: He seems to be entirely in the critical rationalist camp, but he's critical of falsificationism as the goal of some fields.

[00:37:01]  Unidentified speaker: I see.

[00:37:03]  Bruce: You brought up, and we'll put a link in the show, to the criticisms. We want to include the criticisms. You initially, at least, Peter, thought it was a fairly good critical article of Caldwell's article, but I found it difficult to understand what their criticism was.

[00:37:22]  Peter: Yeah, I didn't know if I would quite say it was—reading it again, especially—if it was great. [Hesitation unclear.] I was more curious what you would think about it, I think, more than anything.

[00:37:35]  Bruce: It seemed like what they were saying was, oh, Caldwell's just saying things we already knew. From that standpoint, I think I can understand where they're coming from, because Caldwell's kind of revealing this, like he's come up with this whole idea that critical rationalism can be split into falsificationism or a more generalized version, and they're kind of saying, oh, people worry too much about falsificationism. People, of course, really, it's about criticism, not falsification. Look, I can show you people who said that well before Caldwell said it. And it seemed like a lot of the criticism was based around that. And yet when I actually looked at the examples that they were quoting, I mean, I didn't read the papers that they were quoting, so maybe it's more clear if I were to go to the actual articles. I didn't see where they made the same distinction, where they're saying, look, there's a stronger form and a weaker form.

[00:38:28]  Bruce: It seemed almost like the person writing the article was unaware that there's a stronger form of critical rationalism and a weaker form of critical rationalism, which I think is absolutely necessary in understanding a lot of the confusion that exists around critical rationalism. And from that standpoint, I didn't think that the criticism hit home for me. I mean, if Caldwell's making a fair point that there's this stronger and weaker version and making points about when to use one versus the other and making points about why we should prefer falsification, but it's okay to ignore it if you can't use it and do the best you can with criticism, I didn't see where that was being said previously. And even if it was being said, even if Caldwell's just reinventing the wheel, I don't see how it's wrong. And therefore it seems like this article is still worth a read, if that makes any sense.

[00:39:28]  Bruce: And I couldn't put my finger on any like specific, like what I would have expected, considering the tone of the article, it was like they were revealing, oh, don't look at the man behind the mirror [as spoken], there is no point in reading this article. And yet I would be at a loss to be able to say, because they're arguing that this theory is better instead. So if Caldwell's theory that falsification can be made distinct from critical rationalism, falsification is the stronger version, critical rationalism is the weaker version, but it's more general, it applies to more things. If they're offering an alternative theory, I could not tell you what it was, because I never saw them state it. And because of that, the criticisms just sort of fell flat for me, because I almost just don't care. Like, if you're offering me all these problems with an existing theory, even if I completely agree with you, if you're not offering me an alternative theory, I don't know what to do with your criticisms.

[00:40:27]  Bruce: They may be valid, I may need to figure out how to tweak the existing theory, I may need to find a new theory, but if you're not offering it to me and this is still my best theory, then it's still my best theory, and I'm going to go with it until I find a better theory.

[00:40:42]  Peter: I should add that we have not clarified who the author is. It's not actually clear to me from the website who wrote this, but it's on the rathouse.com.

[00:40:53]  Bruce: Yeah, there was no name on the article.

[00:40:58]  Peter: Yeah, I'm just trying to determine that right now. And I can't.

[00:41:01]  Bruce: So I felt like this was one of the most useful papers that I've read in a while, and it helped me clarify. I mean, I had a lot of these thoughts bouncing around in my head, and, of course, a lot of these things are David Deutsch brings up, so it's not like I hadn't read about some version of it. But I felt the way he put it clarified for me, oh, it makes sense to try to... One of the things I think I've done is when I talk to critical rationalists, I often find that they're doing something, they're doing an epistemology very different than mine, even though I call myself a critical rationalist and so do they. And I've come to the realization it's because I'm doing falsificationism, and they're doing what Caldwell calls critical rationalism. So I'm doing the strong form where I expect everything to be testable, and they're doing the weak form where you just use criticisms.

[00:41:49]  Bruce: And I think that that was a big breakthrough for me, was the realization, the reason why every critical rationalist on the internet disagrees with you, Bruce, almost constantly, is because you're doing falsificationism, you're expecting testability in theories, and they're not doing falsificationism. They're doing the weaker form of critical rationalism. And for me, that made me start to realize, okay, what's going on here? I mean, I think about a lot of the theories I've criticized on the podcast that came out of the fans of David Deutsch. How is it that they get to these theories when they're doing critical rationalism, and these are clearly not testable theory. They're often very open, very open about the fact that this theory is not testable, right? And it always puzzled me, why do they not see that as a problem when critical rationalism is rather explicitly from Popper about falsification and about testability?

[00:42:51]  Bruce: And I think this article at least explains how they're looking at it differently than me. I still have a very valid criticism I wish to offer, which I'll do in a separate podcast. But it allowed me to start to say, okay, they're understanding critical rationalism in the weaker sense, and I'm understanding it in the stronger sense, which I think in other podcasts I've said, I see Popper really as saying, yes, it's okay to do it in the weaker sense, but you should be striving for the stronger sense. Whereas I'm not sure my other friends that are fans of David Deutsch on the internet see it that way. And that may be what the actual disagreement between me and them is.

[00:43:32]  Peter: So the idea is they're kind of just using their creativity to come up with conjectures.

[00:43:40]  Unidentified speaker: Yes.

[00:43:41]  Peter: But they're not taking the next step.

[00:43:44]  Bruce: Yes, which is how do I turn this into a testable theory? How do I then try to refute the theory through an experiment?

[00:43:51]  Unidentified speaker: I see.

[00:43:52]  Bruce: And since you can't always do that, you can kind of see where they're coming from, right? You can kind of see, well, you can't make experiments for Popper's theory, so it must be okay for me to make this a metaphysical theory. I know I've picked on Brett's theory of intelligence a lot, and I don't hate his theory. I actually think his theory has a lot of verisimilitude. I've always seen the mistake that Brett and the defenders of his theory made as they just did not take seriously the idea that this theory that they're advancing, the theory that all intelligence can be explained by interest and time, they just did not take seriously the idea that that can easily be turned into an empirical testable theory. And in fact, they preferred not to turn it into an empirical testable theory and instead wanted to say, no, we're still open to criticism, so we're still doing critical rationalism, but we're just declaring this a metaphysical theory.

[00:44:55]  Bruce: When in fact, this is not a metaphysical theory, this is a scientific theory that the assumption that all intelligence is explained through interest and time, I can immediately falsify it by saying, what about an older person who's starting to get older so that their ability to, how fast they can process and come up with knowledge is starting to go away? Popper himself complained about this. He talked about how he felt this was equivalent to a mental, age was equivalent to a mental illness, and that he just couldn't remember things as well, and he was glad he could still do his math, but it was much harder for him to do it. And this is an example of someone who now clearly spent more time and interest than the younger version of himself, and yet clearly we now have a problem due to something physical. This refutes Brett's theory. It does, right?

[00:45:49]  Bruce: There is now a problem with Brett's theory, period, end of story, and as a critical rationalist, they should be taking that problem seriously. Now, that doesn't mean that they're necessarily abandoning the theory entirely, but some sort of tweak is now going to be necessary where we make an exception for, yes, there can be physical problems, say, due to age.

[00:46:10]  Peter: It seems to me you could say that for a lot of theories, even subscribed to by critical rationalists, for just throwing one example out there, the universal explainer hypothesis. I mean, how I understand it, at least, is it's something that kind of just makes a lot of sense. There's definitely some reasons behind it, but it's not clear to me how you would move that to become an empirical theory, or if people are even trying to do that. Is that an equivalent example?

[00:46:49]  Bruce: It is an equivalent example. I think that's why we get some of these somewhat weird theories that come out of the fans of David Deutsch, like animals don't feel things. I'm not denying, I don't know if animals feel things or not. That's not the issue.

[00:47:04]  Peter: It's an intriguing issue.

[00:47:06]  Bruce: Yeah, they could be right. But the moment they start declaring, this is the best theory because the other theory, we've made these criticisms of it. And then it's just silly at this point. No, of course it's not a best theory. It's a conjecture and that is it. And in fact, there have been attempts to test this, which we've not done a podcast on this, but there have been attempts to test whether animals feel things or not. We used the example in a past podcast of Animal Grief, the fact that the theory that animals don't feel things has to explain animal grief in a testable way, and it can't as of today. That's a giant problem for this Deutsch fan theory that they don't even recognize as a problem, right? Because they've misunderstood the difference between the weaker form of critical rationalism and the stronger form of critical rationalism, the fact that we should be moving towards the stronger form if we can.

[00:48:04]  Bruce: And so it causes them to decide, universal explainership, I think it means animals don't feel things. That's a best theory, so I could feel I can now start to treat animals don't feel things as if it's a best theory, which under the Deutsch way of looking at that, I now embrace it as if it's true, even though it's only tentatively. And you end up with this way of going about things that allows you to endorse theories that just don't belong in the discussion as best theory. And that scientists will never treat as best theories because scientists are holding to the strong form of critical rationalism, the one where you have to actually come up with something testable before you get to even advance your theory. And this, in a nutshell, is one of my main criticisms of the Deutsch viewpoint. I need to probably expand this out a little into a separate podcast.

[00:48:57]  Bruce: I'm kind of doing this on the fly.

[00:48:58]  Peter: It sounds like, in a sense, you're really arguing for just a little more epistemic humility.

[00:49:05]  Unidentified speaker: I am.

[00:49:08]  Bruce: Yeah, it's not that I know the animals feel things. I'm arguing that they haven't gone to the point where they get to announce this as a best theory. And the fact that they often do, that that is absolutely objectively wrong, and that they have violated both the strong and weak form of critical rationalism when they do that. And it's just wrong. I don't know if they're right about animals, but I know that their pronouncement that this is now a best theory is objectively wrong. That's what I know, right? And this is fairly consistent with how critical rationalism actually works. You never assess, is the theory true? You assess, what is the state of the critical argument? Because that's assessable [possibly “accessible”], whereas whether this theory is true or not is not. So I almost don't care, do animals feel things or not? What I care about is, is that they're making a false pronouncement about the state of the critical discussion.

[00:50:04]  Bruce: And that's what I wish to correct, if that makes any sense.

[00:50:08]  Peter: Yeah, it does to me.

[00:50:09]  Bruce: And it is equivalent to epistemic humility. They're not showing the humility that actually is demanded of a critical rationalist, where you really get down and dirty about how testable is this. And if it's not testable, then we don't start pronouncing it as a theory superior to a testable theory. We just don't do that, right? So, okay, that's my soapbox.

[00:50:35]  Unidentified speaker: I've got to go.

[00:50:37]  Unidentified speaker: Okay.

[00:50:41]  Bruce: The Theory of Anything Podcast could use your help. We have a small but loyal audience, and we'd like to get the word out about the podcast to others so others can enjoy it as well. To the best of our knowledge, we're the only podcast that covers all four strands of David Deutsch's philosophy as well as other interesting subjects. If you're enjoying this podcast, please give us a five-star rating on Apple Podcasts. This can usually be done right inside your podcast player. Or you can Google The Theory of Anything Podcast Apple. Or something like that. Some players have their own rating system, and giving us a five-star rating on any rating system would be helpful. If you enjoy a particular episode, please consider tweeting about us, or linking to us on Facebook or other social media to help get the word out. If you are interested in financially supporting the podcast, we have two ways to do that.

[00:51:26]  Bruce: The first is via our podcast host site, Anchor. Just go to anchor.fm slash four dash strands, F-O-U-R dash S-T-R-A-N-D-S. There's a support button available that allows you to do reoccurring donations. If you wanna make a one-time donation, go to our blog, which is fourstrands.org. There is a donation button there that uses PayPal.

[00:51:51]  Unidentified speaker: Thank you.