00:00:00.000 Welcome to Topcast and to part 2 of my discussion of chapter 7, a conversation about justification
00:00:08.880 from the fabric of reality by David Deutsch. Now actually this is the third-ish episode
00:00:16.080 on this particular chapter. I've inserted a supplementary episode which is a discussion about
00:00:23.920 some material in realism and the aim of science which is a very long book by Karl Popper written
00:00:31.040 some time after the logic of scientific discovery which was his first book and in that book in realism
00:00:36.960 in the aim of science, among other things he talks about something called corroboration. Now I bring
00:00:43.040 that up and I found it relevant to the present episode and to the episode before that which was
00:00:48.880 kind of an introduction to the conversation that we're going to have today, the literal
00:00:52.640 conversation about justification. Because you can see Popper there trying to escape from the
00:00:59.360 prevailing view of epistemology. The prevailing view of epistemology at the time of course was
00:01:04.400 well you observe staff and you observe staff repeatedly and you become more certain. You confirm
00:01:10.080 your theory in some way by repeatedly observing things and you think that that theory is more
00:01:14.960 likely to be true, something like that. And Popper rejected this whole idea. Popper rejected the
00:01:19.520 idea that theory became more likely to be true or you could assume they were more likely to be
00:01:25.680 true based upon your confirming instances based upon you, seeing what you expected to see.
00:01:31.600 The whole all swans are white thing and so you're observing white swans and you become more
00:01:36.560 convinced apparently. The more white swans you see that in fact it's true that all swans are white.
00:01:41.920 Of course this particular process of predicting the color of swans and prediction in general
00:01:47.920 is not even the aim of science which is what Popper's whole project was. He was trying to explain
00:01:53.680 something completely new, something hitherto that no one had really understood about what the
00:01:58.880 creation of knowledge was and what science was all about. So you can see him trying to escape
00:02:04.560 the miring which he finds himself, the philosophical mire of his contemporaries and of the people that
00:02:10.720 went before in trying to explain stuff about explanation. And in trying to get away from what they're
00:02:17.440 saying, he's still using their language and you can see that he's, well in my opinion, trying to
00:02:24.160 fill a void in our language in our vocabulary which doesn't need to be filled. You can do away
00:02:29.920 with it, you can do away with so many of these words, buy his own lights, buy the way in which
00:02:34.480 he explains epistemology and how it works. By the fact that what we're after criticisms, we're
00:02:39.600 using this critical method to try and refute bad ideas, leaving us with a best existing idea.
00:02:45.360 It's either that or we've got nothing at all. We're not in the position of trying to
00:02:49.600 wait different ideas or to justify a particular idea as more likely to be true than some other
00:02:54.560 idea. We're just not in that position. It just doesn't come up. It's not a problem to be solved.
00:03:00.240 Hence, the whole problem of induction is a climber itself. It didn't need a solution as such.
00:03:05.360 The solution is, it was never a problem in the first place because induction isn't a thing.
00:03:10.480 There is no whole at the center of science where we need to try and become confident in our
00:03:15.680 theories because the king of knowledge is something like mathematics or logic or something like that
00:03:21.040 and what we're trying to start with are absolutely certain foundations and on that building an
00:03:25.520 edifice of ever more certain claims about reality. That's just not what's going on.
00:03:30.480 Pop us entire conception turns that on its head. What we're doing is we're
00:03:33.840 conjecturing, we're guessing, we're creating explanations that can be tested against reality and
00:03:39.920 can be shown to be wrong. If we can't show them to be wrong, we're lucky for us. We've learned
00:03:44.560 something about reality. We keep hold of that bit of stuff, that bit of information about reality
00:03:49.920 and we call it knowledge, something we know about reality. This is what we explain here at
00:03:54.960 TopCarsis, what David Deutsch explains about the beginning of infinity. Why it should be relevant
00:04:00.080 for me to go all the way back to Karl Popper and the realism in the aim of science and talk about
00:04:04.880 its corroboration stuff is because David uses the language of corroboration here in this chapter
00:04:10.640 and I think quite rightly once again, speaking in the language of people who are going to disagree
00:04:15.040 and there are still people who disagree today. Bazingism we have to admit is still a far more popular
00:04:20.240 and an ascendant even epistemology. People still have this idea that what science is about is trying
00:04:27.440 to predict the future and you're trying to be more and more sure or confident or certain that
00:04:33.360 this particular theory is actually a true account of reality or something like that. So if you've
00:04:39.120 ever been in a conversation with someone who disagrees with popular inner epistemology,
00:04:44.080 it can sometimes be difficult to get across what you're saying because just simple things like
00:04:48.960 claiming to know something. In the Papurian framework means you have an explanation. It could mean
00:04:55.280 you have an intuition but in any case you have some idea that solves your problem. That's what it
00:05:00.960 means to know something. You've got a solution of some kind. Sometimes you can know how to ride a bike
00:05:06.400 without having an explicit explanation of what's going on. What you've got is in explicit knowledge
00:05:12.000 and a whole bunch of implicit assumptions about exactly how you might body moves and that kind of
00:05:16.400 thing but you know how to ride a bike. It's a form of knowledge even if you can't explain it in words
00:05:22.080 or you can't explain it fully in words. Poppets account of knowledge takes account of this kind
00:05:26.240 of thing because among other things it's knowledge that can be improved over time. You never need to
00:05:30.880 justify as true this bit of knowledge that you have about how to ride a bike or how to swim or
00:05:35.840 any of this other in explicit stuff. It's stuff that in explicit knowledge just isn't stuff that
00:05:40.800 performs a part of anything like Bayesian epistemology or induction and whatever the case when we're
00:05:46.240 talking about explicit explanations. The rules don't need to change. Not much. We don't move from
00:05:54.000 having guesses about how to do stuff like ride a bike or guesses about what reality is like
00:05:59.360 into an area of certainty. That's not what's going on. So here in this chapter before us,
00:06:04.640 chapter seven, a conversation about justification. David is going to have an imaginary conversation
00:06:10.240 with someone who doesn't think that they're inductive. So they're called the crypto inductive us
00:06:14.560 but they have an induction shaped hole in their way of thinking about the world. They don't
00:06:18.880 understand why we should rely on our existing theories, our best existing theories of reality,
00:06:24.480 our scientific theories. They think the crypto inductive us thinks there needs to be some process by
00:06:29.440 which we justify, by which they presumably mean justify us true, any given theory that we have at
00:06:35.520 a particular time. And David is rejecting that whole idea as a proper indeed rejected that whole
00:06:41.200 idea. But our language, our common language between popurians and non-popurians and just person on the
00:06:47.680 street thinking about stuff. Well, we share this language, we call English and we have a whole
00:06:52.640 bunch of words, some of which can be quite misleading. Justification is one of them. What does
00:06:57.600 justified really mean? Well, it could mean something like showing us true, given some assumptions
00:07:04.400 you already know are true. Now, never mind how you know those assumptions are true. Of course,
00:07:08.800 you don't. Putting that aside, as David said and as I'll emphasize again now in the introduction
00:07:14.320 to the audiobook version of this, he said there that when he uses the word justified or justification,
00:07:20.800 he in the fabric of reality, what he intends is the normative claim. In other words, saying,
00:07:26.720 not that our best theories are justified is true, but rather you are justified in the sense of
00:07:31.440 you should use your best theory. What else could you do? You want to explain some sort of
00:07:35.680 cosmological phenomenon like gravitational lensing. Well, the only thing you can do, it is right,
00:07:40.400 you are justified in using the general theory of relativity. You're justified in using
00:07:45.440 evolution by natural selection in order to go through the process of explaining exactly
00:07:50.480 how it is that certain species exists in reality right now on planet Earth. And presumably,
00:07:56.080 any life that exists out there elsewhere in the universe, you're also justified in assuming
00:08:00.800 that that life there has also evolved by a process of natural selection because we know of
00:08:06.240 no other process and all we can go on if we have problems like, let's say, one day we observe
00:08:12.320 alien life and once we explain where did that come from, our best theory is that all life,
00:08:17.120 no matter where it is and of what form has evolved via a process of evolution by natural
00:08:21.840 selection. Now, could that be wrong? Of course, it could be wrong. So we're not justified in thinking
00:08:25.440 our theory is true, literally true, but we're justified in using it. Why? In order to solve our
00:08:31.680 problem, the best that we can at any given time and solutions to problems don't need to be the final
00:08:38.560 ultimate optimal solution for all time and everywhere. They just need to solve your problem now
00:08:43.440 to your satisfaction to the scientist satisfaction to the satisfaction of being able to say
00:08:49.520 problem solved and move on to the next best problem, the next most interesting problem, the next
00:08:54.160 thing that is important to you. Okay, so I'm not going to read the entire dialogue today here
00:09:01.840 in chapter seven because it's a long dialogue and I think that listeners to talk cast
00:09:08.320 if you've made your way through the beginning of infinity, you'll have a solid grounding in
00:09:14.240 an understanding of what epistemology is all about. So I just want to highlight sections of
00:09:18.640 the conversation that can be illuminating to us and just to of course enjoy the clarity of David
00:09:24.320 Deutsch's writing on this. I'm also going to pause at certain places where I think that
00:09:28.800 as I did in the last episode about this, about this particular chapter proper, where I was saying
00:09:35.120 that, well, we could probably rephrase this in a slightly different way in light of David's own
00:09:39.920 more recent work on this in light of what David says in the beginning of infinity and amongst
00:09:45.120 other things in his paper, his even more recent paper than the beginning of infinity,
00:09:49.120 that paper being the logic of experimental tests. And I think that that now, as far as I'm aware,
00:09:56.160 is the pinnacle of our understanding of exactly what science is about and the role of things
00:10:02.080 like observations, experiments, in scientific discovery. That is right now our most refined
00:10:09.120 understanding of things. And so we can draw a line if you like, all the way from prepopurian stuff
00:10:15.200 to what we might say is early pop or an in realism in the aim of science,
00:10:19.280 pop or admits he's got an early version of himself. And the early version of himself in the
00:10:22.960 logic of scientific discovery was saying things like and he admits he admits himself that this
00:10:27.920 was an error, saying things like theories can have a degree of confirmation. Now he was wrong
00:10:33.280 about that and he admits he was wrong about that. And so he moves on to degree of corroboration
00:10:38.880 because he thinks that some theories are more probable than other theories. And we, of course,
00:10:44.720 now think this is wrong following the work of David Deutsch. So David Deutsch has done stuff in
00:10:49.680 epistemology since the fabric of reality, which kind of does away with some of the ways of phrasing
00:10:55.040 these things. And so there's a more refined way of understanding some of this stuff,
00:10:59.360 certainly in the beginning of infinity, and certainly in something like the logic of experimental
00:11:04.640 tests, his academic paper on this. And if you're interested in that, it's kind of hard going,
00:11:10.160 it's focused on, well, the full title of the paper is the logic of experimental tests,
00:11:15.440 particularly of ever retying quantum theory. And so it can get quite technical in places. Now I've
00:11:20.240 written a guide about this, and it's on my website, just called the philosophy of science. And
00:11:25.520 there is a podcast all about devoted to this as well. If you're interested in that, you can go all
00:11:31.040 the way back to topcast episode 22, the logic of experimental tests where for a little over an hour,
00:11:38.000 I go through my understanding of and my reading of the logic of experimental tests, certainly just
00:11:44.160 excerpts thereof. And so that too conform a supplement to something like this episode, where we
00:11:50.000 are looking at the debate between Papyrians and Papyrian epistemology. And well, the rest of people
00:11:58.480 who think about how it is that science accomplishes what it does or what the purpose of science is.
00:12:03.760 And why we should rely upon our best explanation at any given moment in time. Why this is a debate
00:12:10.400 from our point of view is a mystery. After all, what else can you do but rely upon your best
00:12:16.880 existing theory? And if you can't show your best existing theory as actually finally once and
00:12:22.640 for all true in some way, who cares? You've got nothing else to rely on. And I guess push come to
00:12:31.360 shove a reasonable base in would say the same thing, but one wonders what they're really engaged in
00:12:36.160 if they're trying to increase their confidence, increase their credence in a particular claim.
00:12:41.840 Okay, without further ado, now I did think to myself, I would have further ado, I say, and then
00:12:46.320 he goes on to make more of an adieu. I did think to myself, should I get one of my friends or someone
00:12:54.240 else to play the role of David or play the role of crypto inductivist? And thereby I have a
00:13:00.960 proper dialogue. And after much thought about this, I thought, no, I will just read both parts.
00:13:06.640 So this might become a little bit irritating. And I'm not going to try and be a voice actor for this.
00:13:11.520 I'm just going to read it dry and pause where I think there are interesting things to talk about.
00:13:17.200 Okay, so let's go. This is the beginning of the actual dialogue, the actual conversation about
00:13:22.800 justification that David has in the fabric of reality chapter seven. And he begins with himself
00:13:28.480 asking the question of the crypto inductivist. David says, quote, since I read what
00:13:34.960 Papa has to say about induction, I have believed that he did, indeed, as he claimed, solve the problem
00:13:41.200 of induction. But few philosophers agree. Why? And the crypto inductivist replies, because
00:13:47.440 Papa never addressed the problem of induction as we understand it. What he did was present a
00:13:52.400 critique of inductivism. Inductivism said that there is an inductive form of reasoning,
00:13:58.400 which can derive and justify the use of general theories about the future given evidence in the form
00:14:04.320 of individual observations made in the past. It held that there was a principle of nature,
00:14:09.600 the principle of induction, which said something like observations made in the future,
00:14:14.080 are likely to resemble observations made under similar circumstances in the past.
00:14:19.520 Attempts were made to formulate this in such a way that it would indeed allow one to derive or
00:14:24.000 justify general theories from individual observations. They all failed. Papa's critique, though,
00:14:30.800 influential among scientists, especially in conjunction with his other work elucidating the
00:14:35.360 methodology of science, was hardly original. The unsanness of inductivism had been known almost
00:14:41.680 since it was invented, and certainly since David Hume's critique of it in the early 18th century,
00:14:47.760 the problem of induction is not how to justify or refute the principle of induction,
00:14:52.880 but rather taking for granted that it is invalid, how to justify any conclusion about the future
00:14:59.200 from past evidence. And before you say that one doesn't need to, and David interjects,
00:15:05.680 one doesn't need to. The crypto inductivist then comes back with, but one does. That is what
00:15:10.320 is so irritating about you, Papyrians, you deny the obvious. Obviously, the reason why you are not
00:15:16.240 even now leaping over this railing is in part that you consider it justified to rely upon our
00:15:22.400 best theory of gravity and unjustified to rely on certain other theories, of course, by
00:15:28.000 our best theory of gravity. In this case, I mean more than just general relativity. I am also
00:15:33.120 referring to a complex set of theories about such things as air resistance, human physiology,
00:15:38.240 the elasticity of concrete, and the availability of mid-air rescue devices end quote.
00:15:43.600 So remember the context here. What's going on as the two characters, David and the crypto
00:15:48.160 inductivist, are standing at the top of the Eiffel Tower. And the whole point of the dialogue is
00:15:53.440 that David is trying to explain why he is justified in relying upon the best existing theories
00:16:00.720 as to why he shouldn't jump over the railing. And the crypto inductivist is saying,
00:16:06.000 well, no, you haven't properly justified that claim. There's no good reason, so to speak,
00:16:12.560 for relying on this best theory. So there's some back and forth at this point between
00:16:17.920 David and the crypto inductivist. And I'll just leap to the point where the crypto inductivist
00:16:22.640 tries to summarize David, David's argument. And what the crypto inductivist says is, quote,
00:16:27.840 so to summarize, you believe that the evidence currently available to you justifies the prediction
00:16:33.200 that you would be killed if you leapt over the railing end quote. Now, do you listen to?
00:16:38.080 If you hear something like that, let me just say it again. The crypto inductivist is saying
00:16:41.920 to David, you believe that the evidence currently available to you justifies the prediction that
00:16:46.480 you would be killed if you leapt over the railing. Does anything spring to mind?
00:16:50.160 It perhaps should. The evidence doesn't justify anything. What's the point of evidence?
00:16:56.640 The point of evidence, the purpose of evidence, why one gathers evidence in science at all,
00:17:02.000 is to decide between theories already guessed. It's not there. It's function is not to justify
00:17:10.080 as true or as more true, a particular theory. Instead, it just rules out some whole bunch of theories
00:17:17.600 and is only explained by the soul standing theory. So the crypto inductivist is wrong to summarize
00:17:24.000 David in that way. Also, the crypto inductivist is absolutely focused on prediction. Now, in this
00:17:29.360 particular case, I guess that being focused on prediction is what it's all about. After all,
00:17:33.600 there's a baiting whether or not one is justified in leaping over the railing and thinking they are
00:17:38.160 or are not going to be killed, given the best current explanation. But the justification for
00:17:43.040 all or not leaping over the railing from top of the Eiffel Tower is all about the fact that
00:17:49.760 not the general relativity and those other theories that were mentioned earlier are absolutely
00:17:54.160 true or anything like that. It's the only theory you've got. It's the only theory you've got.
00:17:59.280 Everything else has already been ruled out by various other experiments. And so that theory
00:18:03.440 allows you to derive certain predictions, there are certain consequences that follow.
00:18:07.840 One of which is, well, anything that goes over that railing is going to hit the ground and
00:18:12.160 be destroyed upon impact. That's just what happens. A prediction, I say, is something you logically
00:18:19.760 derive from a given explanation, a given good scientific explanation. It's just a consequence. It's
00:18:25.760 one of the things that follows from assuming all else equal, which is often a hard thing to do.
00:18:31.120 You can't always assume all else being equal because in our real world, people's choices,
00:18:37.680 human creativity comes into play rather often. So only in very well-controlled environments,
00:18:44.880 like in a laboratory, can you do properly controlled experiments? And this is why
00:18:50.800 prediction is valid there. We could say, if you like, the predictions made in laboratory where you
00:18:56.800 have carefully controlled your variables and all of the conditions and you understand the
00:19:02.160 functioning of the equipment well, then you are justified on relying upon the predictions.
00:19:06.480 But outside of the laboratory, things get more tricky. Things get more tricky, especially in the
00:19:13.680 realm of human affairs. And that includes things like anything happening on the earth,
00:19:19.440 which could be geological or meteorological or climate related, or to do with the extinction
00:19:26.240 of animals or not, the evolution of life now on earth is very much affected by the choices
00:19:32.800 that people make. And so our capacity to predict the behavior of systems here on earth depends upon
00:19:40.160 knowing what people will do to impact those systems. And we can't predict what people will do,
00:19:46.880 especially people over the future. We don't know what knowledge they will have or will create.
00:19:51.360 And so this is why prediction outside of the carefully controlled laboratory, and by the way,
00:19:55.280 that includes the carefully controlled laboratory of deep space, which presumably there aren't
00:19:59.920 people out there affecting things just yet. But places where you can ignore the choices and
00:20:06.560 effects of people and knowledge creation, if you can do that, then you've got a set of valid
00:20:12.160 predictions. Otherwise, what do we say? We say we've only got prophecies, wild guesses where we
00:20:18.960 presume that we know the content of our future theories. And often when it comes to prophecy,
00:20:25.440 people are assuming they know the content of those future theories to be exactly the same as
00:20:30.320 the content of theories today. In other words, whatever problems we have today will still be there
00:20:35.120 in a hundred years because we won't have created the knowledge to fix the problem then, or something
00:20:39.760 like that. When you hear about people talking about the distant future and some existential
00:20:45.600 catastrophe that's on the way, talking about that today, they're engaged in pure prophecy.
00:20:52.000 They don't know what the next generation is going to do. They don't know what they are going to do
00:20:56.560 20 years from now, which could have an effect on the very problem they are so agitated in being
00:21:01.120 worried about right now. And I think it's good for people to be worried about certain problems today
00:21:06.400 that could have an effect in decades to come. But making prophecies about this particular thing
00:21:12.800 actually happening, well, that's wrong. So that's the distinction between prediction and prophecy.
00:21:17.760 But here we're focused on prediction. So let me go back and recap what the crypto inductive
00:21:23.840 is just said and and going with what David responds to. So the crypto inductive is just said,
00:21:28.320 quote, so to summarize, you believe that the evidence currently available to you
00:21:32.080 justifies the prediction that you would be killed if you leapt over the railing. David says,
00:21:36.320 no, it doesn't. And the crypto inductive says, but damn it, you are contradicting yourself.
00:21:40.640 Just now you said that prediction is justified. And David says, it is justified.
00:21:46.080 But it was not justified by the evidence, if by the evidence, you mean all the experiments
00:21:52.720 whose outcomes, the theory correctly predicted in the past. As we all know, that evidence is
00:21:59.520 consistent with an infinity of theories, including theories predicting every logically
00:22:04.640 possible outcome of my jumping over the railing. The crypto inductive says, so in view of that,
00:22:09.600 I repeat, the whole problem is to find what does justify the prediction. That is the
00:22:15.760 problem of induction. And David says, well, that is the problem, pop a solved end quote.
00:22:22.400 And quite right. And what I'll say here, because I'm just going to skip over a whole bunch of
00:22:26.320 things is if you cast science in terms of being a project about trying to predict the future and
00:22:33.280 almost nothing but this process of generalization or extrapolation that a set of observations in the
00:22:41.840 past should continue off into the future, because you can draw a straight line through a
00:22:48.160 particular trend. Without an explanation, you're assuming an explanation, the explanation you're
00:22:53.120 assuming is that that observation in the, that set of observations of past is somehow necessarily
00:23:00.640 required by laws of physics or something like that. And so therefore there is a law of physics or
00:23:04.880 something that ensures that what you've seen in the past will continue to be seen in the future,
00:23:10.960 which isn't much of an explanation. It's just kind of assuming there's an explanation beneath
00:23:15.200 all this. But you're not searching for explanations. You're searching for generalizations,
00:23:20.400 extrapolations, the ability to make predictions. This is the whole reason induction is just so
00:23:26.080 vacuous, pointless. A misguided attempt to understand what is going on with science,
00:23:33.040 science, like every other domain that is of interest to people in academic circles,
00:23:38.880 or just in problem-solving in general, it's actually about creating explanations, about understanding
00:23:44.880 the world. And once you have some understanding, then if you're lucky, you might be able to make
00:23:50.560 a prediction about something. Sometimes not. Sometimes even our best explanations don't allow
00:23:55.280 to make predictions. It's kind of in the physical sciences. And even then, not all the physical
00:23:59.600 sciences, but in the physical sciences, we can make some predictions sometimes under some circumstances,
00:24:05.040 as I already said. One of the trope examples I'd like to use is, well, we have these theory of
00:24:10.800 acids and bases. There are these two kinds of chemicals that exist in the world. One creates
00:24:17.200 hydronium ions when dissolved in water. These things are called acids. To a first approximation,
00:24:22.640 you say it's the hydrogen ion, H plus, that goes floating around in solution. It's not quite
00:24:27.840 like that, by the way. It actually creates these things called hydronium, which is H3O plus.
00:24:33.200 Never mind that. Whatever substance dissolved in water that produces such ions, we call
00:24:38.080 an acid. And on the other hand, we have these things called bases, which when dissolved in water
00:24:42.640 produce the hydroxide ion. The hydroxide ion is OH minus. It's a negative charge. And so things
00:24:49.680 like sodium hydroxide or potassium hydroxide, these are bases. And they can be used when combined
00:24:55.840 with fats to make soap, but never worry about that. Now, all of the soluble bases we have this
00:25:00.160 word called alkali. And so alkali is just a name for a base which will dissolve in water because
00:25:05.760 there are certain bases that don't dissolve in water. Now, bases or alkali is dissolved in water
00:25:10.400 produce the hydroxide ion, OH minus. And to a first approximation, the acids when dissolved in
00:25:15.600 water produce H plus, the hydrogen ion. Now, when you mix an acid and a base, so they call
00:25:22.240 concentration equal amounts and correct morality and all this sort of stuff, what you get, the products
00:25:28.560 that you get from the reactants, the reactants are acid plus base, is something called a salt plus water.
00:25:34.400 Now, where did the water come from? The water came from. The hydrogen ion over here from the acid,
00:25:40.320 H plus the hydroxide ion from the base, OH minus over there, H plus OH gives you H2O
00:25:48.880 water. And so that's where the water comes from when you mix an acid in the base. So you combine
00:25:52.720 an acid and a base. What am I telling you all that for? Well, that's a good explanation about what's
00:25:57.520 going on chemically with stuff and it allows you to make specific predictions. That's the general
00:26:03.040 universal rule that anywhere in the universe that you happen to have an acid dissolved in water
00:26:08.720 and a base dissolved in water. Then you will get a salt and water being produced as the products.
00:26:15.920 And so you take a specific example, in my trope example, I like to use this because it's the one I
00:26:20.240 can easily do in my head, is hydrochloric acid, which is HCl. That's the chemical formula
00:26:27.280 for it. And then sodium hydroxide, NaOH. This is probably the easiest reaction of all.
00:26:33.920 Equal concentrations and equal amounts of that will produce as products, HCl plus NaOH.
00:26:40.720 You will get sodium chloride NaCl, that's plain old table salt, and H2O. That's a prediction.
00:26:49.120 So if you give me any acid and you give me any base, which can dissolve in water,
00:26:53.760 then I should be able to tell you what the salt is, what the specific salt is that is going to be
00:26:58.960 produced, as well as being water. That's a prediction. I can make that kind of prediction. And in
00:27:05.040 fact, any reasonably competent high school chemistry student would also be able to tell you
00:27:10.080 precisely what the concentrations are required. And the volumes are required. This is the
00:27:14.320 process of titration, by the way. You can figure out the concentration of an unknown base,
00:27:19.440 given a certain concentration of a particular acid. And you combine these things together
00:27:23.600 and you can do quite simple calculations. You pull out your periodic table and your pocket
00:27:27.120 calculator and you can calculate this stuff. So you can make even more precise quantitative
00:27:31.760 predictions about this kind of stuff. This is physical chemistry, an organic chemistry
00:27:36.480 allows you to do the same thing. This is the wonderful thing about chemistry. It allows these
00:27:40.640 precise, very precise predictions exceeded perhaps only by physics itself. But then
00:27:47.120 sticking within the realm of physical staff, we only have to move to geology.
00:27:51.920 And suddenly, the systems become too complex for us to make precise predictions. We understand
00:27:59.920 some stuff, but we don't understand enough to make the predictions that we would like to,
00:28:04.800 if only we could. For example, we can't predict earthquakes and we can't predict volcanic
00:28:11.280 eruptions. Not to the precision and accuracy that we would like to. We'd like to be able to have
00:28:16.880 warnings well ahead of time that a particular earthquake is going to happen in a major city,
00:28:21.760 and do whatever it takes in order to protect people and structures and that kind of thing.
00:28:26.240 But instead what happens is the earthquake happens, the buildings fall down, people get hurt,
00:28:30.320 and then we come to have a better understanding after the fact of how earthquakes happen.
00:28:35.120 But we don't have the capacity just yet, just yet, to make these predictions ahead of time.
00:28:41.520 And that's a shame. And that's just geology. Rocks moving around. Well,
00:28:45.600 tectonic plates, but you get my point. These are by comparison, simple systems as compared to
00:28:52.400 well, anything involving human affairs, one day we'll be able to predict when earthquakes are happening
00:28:58.000 are going to happen one day, presumably. It must just come down to physical forces and it must
00:29:02.880 just come down to how much stress is being placed upon these tectonic plates, which we don't have
00:29:08.000 the ability to measure precisely just yet. But perhaps one day we must be able to one day.
00:29:13.280 And then we'll be justified in the future, relying upon the prediction from that good explanation.
00:29:19.840 We just don't have a good enough explanation just yet. We are right. We are justified in predicting
00:29:26.480 following a good explanation of how acids and bases interact, exactly what, for example,
00:29:32.640 concentration of base we need to neutralize a particular acid. That's another prediction that we
00:29:37.680 can make and we're justified in relying on certain calculations. It's the normative things to do.
00:29:43.920 We should use chemistry in order to solve those kind of problems in titration, let's say.
00:29:50.000 And if someone's talking about jumping over the railing of the Eiffel Tower, we are justified
00:29:55.760 in predicting they're going to hit the ground at such a velocity that it'll kill them.
00:30:00.400 Now, why? Because our best explanation, the combination of our best explanations, but
00:30:04.960 chief among general relativity says that you're going to accelerate towards the ground at a particular
00:30:08.880 rate, slowed only a little by things like air resistance. But your skull's going to hit the ground
00:30:15.280 and you're going to be dead. And that conclusion is justified in the sense that it follows from
00:30:20.480 the assumption that all those other theories are good explanations about what's going to happen.
00:30:25.600 They don't have to be true in any final sense, but they capture some truth about reality,
00:30:30.400 some truth. And some of the truth is that they're going to accelerate towards the ground.
00:30:36.080 So, once again, we have more back and forth between David and the crypto inductivist.
00:30:40.880 And I'm going to pick it up where the crypto inductivist says, quote,
00:30:44.720 Now, listen carefully, because you have just said something which is not only
00:30:48.960 provably untrue, but what you yourself conceded was untrue only moment ago. You say that the
00:30:55.920 outcomes of experiments, quote, refuted all the rival theories end, quote, but you know very well
00:31:02.960 that no set of outcomes of experiments can refute all possible rivals to a general theory. You
00:31:09.520 said yourself that any set of past outcomes is, I quote, consistent with an infinity of theories,
00:31:16.160 including theories predicting every logical possible outcome of my jumping over the railing end,
00:31:20.880 quote, it follows inexorably that the prediction you favor was not justified by the experimental
00:31:27.280 outcomes, because there are infinitely many other rivals to your theory, also unrefuted as yet,
00:31:33.280 which make the opposite prediction. And David responds, quote, I'm glad I listened carefully,
00:31:39.200 as you asked for now I see that at least part of the difference between us has been caused by
00:31:45.120 misunderstanding over terminology. When Papa speaks of rival theories to a given theory, he does
00:31:51.520 not mean the set of all logically possible rivals. He means only the actual rivals. Those proposed
00:32:00.240 in the course of a rational controversy that includes theories proposed purely mentally by one person
00:32:06.800 in the course of a controversy within one mind, end quote, and this is just me talking now. Yes,
00:32:12.320 of course. Now, it is impossible. It would be impossible to enumerate all the logically possible
00:32:17.360 theories. There wouldn't be a set. It would be some sweet class of things. And if we take the
00:32:24.720 worldview of the beginning of infinity seriously, then we must presume there are an infinite number
00:32:30.560 of theories better than our best existing theory yet to be generated. That's part of the beginning
00:32:36.560 of infinity that we will get ever closer and closer to describing reality without getting
00:32:42.480 ever a final description of reality. Because reality itself has this infinitely complex character.
00:32:50.560 The universe is vast beyond our imagination. And every time we discover something new about it,
00:32:57.040 it reveals a whole bunch of new phenomena that we are only just scratching the surface of.
00:33:02.400 And we're always just scratching the surface. So there must be this in theory infinite number of
00:33:08.320 theories yet to come. We're never going to get to the end all of our process of scientific discovery.
00:33:14.240 No matter what people say, no matter what books are written about the end of science,
00:33:19.440 no matter what other podcasts to say that we're almost there with a completed science of this
00:33:24.000 that or the other. No, we are coming to understand objective reality ever better over time.
00:33:30.720 But the end is nowhere near inside. Isn't that fun? That's great. That's cool. That means
00:33:34.800 there's always something new to do. And you can start anywhere and point yourself in any direction
00:33:40.400 and there'll be problems and you'll be able to make progress. So it's not possible to rule out
00:33:45.680 all of the theories. Because there's a whole bunch of theories and infinite numbers, I just said
00:33:50.160 that are better than your existing theory. Okay, you're yet to think of them. No one has yet
00:33:54.400 sort of them. But they will. They will think of them because they will need those to solve
00:33:59.280 problems yet to be encountered. Problems that will only be encountered once certain problems have
00:34:04.800 been solved now with our existing theories. So, you know, the existing theories that are certain
00:34:10.400 problems with, for example, the meshing of quantum theory and general relativity. You know,
00:34:14.800 what is the ultimate nature of reality? Is it discrete or continuous? Is it some other hitherto
00:34:19.600 unimaginable third thing? I don't know. But the fact is, we don't know the ultimate constituents
00:34:25.440 of matter. We may never know the ultimate constituents of matter because there might not be
00:34:29.440 ultimate constituents of matter. We just keep finding smaller and smaller, different and different
00:34:34.160 things that themselves are made up by other things. So I'm skipping ahead and the crypto
00:34:40.400 inductive is objects to being called and inductive is at all because they don't believe in
00:34:45.120 induction. As they say, as the crypto inductive says, quite offended. Quite. It really is
00:34:50.400 perverse to call a person and inductivist. If that person's whole thesis is that the invalidity
00:34:56.640 of inductive reasoning presents us with an unsolved philosophical problem, in David says,
00:35:02.240 I don't think so. I think that that thesis is what defines and always has defined an inductivist.
00:35:09.840 But I see that Popper has at least achieved one thing. Inductivist has become a term of abuse.
00:35:15.200 Anyway, I was explaining why it's not so strange that the reliability of a theory should depend
00:35:21.280 on what false theories people have proposed in the past. Even inductivists speak of a theory
00:35:26.640 being reliable and not given certain evidence. Well, Papurians might speak of a theory being the
00:35:32.160 best available for use in practice, given a certain problem situation. And the most important
00:35:38.320 features of a problem situation are, what theories and explanations are in contention,
00:35:44.080 what arguments have been advanced, and what theories have been refuted.
00:35:48.480 Corobration is not just the confirmation of the winning theory. It requires the experimental
00:35:54.960 refutation of rival theories confirming instances in themselves have no significance.
00:36:00.720 End quote. Yes. So this is one of the motivations for my immediate prior episode to this one,
00:36:08.560 where it's titled corroboration, because I think Popper was trying to escape from this kind of
00:36:15.120 language. And I think I can see here, and of course I can't speak for David, but I can see here
00:36:19.920 that David doesn't use this kind of language in the beginning of infinity, or much at all
00:36:24.720 since the fabric of reality, because I think it's kind of superfluous to our needs.
00:36:29.600 As he says there, quote, corroboration is not just the confirmation of the winning theory.
00:36:35.600 Now what I would say, and I think even Popper kind of admitted, is that, well, confirmation
00:36:40.800 is not a thing. What does confirmation mean? To confirm something means to show us true in some way?
00:36:48.000 Or it just means the observation is consistent with the theory, but we've already said here in
00:36:52.080 this chapter already that, you know, there are an infinite number of theories that could be
00:36:55.440 consistent with any set of observations. Fine. So we don't need to worry about confirmation.
00:37:01.600 And I think for similar reasons, we don't need to worry about corroboration, either. All of these
00:37:07.840 things are kind of on the positive side of the ledger, trying to support a theory in some way.
00:37:14.800 But we don't need to support a theory at all if it's the only existing theory.
00:37:19.760 And this is our situation in science, and everywhere we have our theory. It is exceedingly rare.
00:37:26.800 Exceedingly, I'm just laughing at myself because I think I've said this phrase so often recently.
00:37:32.160 It is exceedingly rare to have multiple theories of anything at all. You know, at the moment,
00:37:37.600 you look at something like dark energy, and people might very well say to me, well, this is a
00:37:42.320 problem, but you look up the Wikipedia article, and there's all sorts of theories about dark
00:37:46.000 energy, and I would just say no, no. None of them count as theories in the sense that I'm talking
00:37:52.080 about them. What a theory is in my usage of the word here. Now, of course, theory can just
00:37:58.480 mean any old wild guess that you like. But it should mean good explanation. And this is what we're
00:38:04.880 really talking about when we're talking about having multiple competing theories that make a
00:38:10.800 claim about reality. What we're saying is, multiple competing good, either two, good explanations
00:38:18.080 of reality, hard to vary, hard to vary accounts of the world, postulating the existence of
00:38:24.800 specific physical things in the case of Newtonian gravity, the postulation of a physical force,
00:38:31.680 an action at a distance that travels instantaneously between bodies. That's postulating a real
00:38:37.360 physical thing that we can experimentally test for, and in fact, fails the test. And on the other
00:38:42.880 hand, you've got general relativity postulating the existence of a true fabric of space time that
00:38:48.240 can bend and warp in the presence of mass and energy. And we can test for that too. And in fact,
00:38:53.680 passes the test. Now, not to say that it's confirmed as true or anything like that. It's just,
00:38:58.160 it's gone unrefuted. And if you want to explain stuff like gravitational lensing or
00:39:02.720 the bending of light during an eclipse or what's going on with a black hole, you've only got one
00:39:08.320 explanation. Now, forget support. We don't need to support it. Like, why what's the point of what
00:39:12.720 what help is support in that situation? Let's say we had a hundred points of support. Great. Well,
00:39:18.560 what's the rival? There's no rival. It's not like there's something else over there with
00:39:22.000 90 points of support. There's nothing. The next best thing you've got is Newton's theory of gravity.
00:39:27.040 It's already refuted. It can't do any of this other stuff. It can't explain precisely what's
00:39:33.120 going on with gravitational lensing. It can't explain black holes in the same way. It can't explain
00:39:38.320 gravitational waves, someone and so forth, just so much stuff now. There's only one thing. And this is
00:39:43.840 true across science. I mentioned earlier, the theory of acids and bases. The modern understanding
00:39:50.640 of that, whatever the chemists call it now, there is no rival theory. There used to be different
00:39:55.920 rival theories about what an acid and a base were was and how they pay. But the history of science
00:40:02.880 shows that there was this gradual process of incremental improvement where the previous theories of
00:40:08.320 acids and bases were ruled out leaving us with the modern understanding. The same as the history
00:40:12.880 of the atoms and other interesting one where people just didn't know they kind of had the idea
00:40:16.880 there was an atom. But what was the structure of an atom? No one knew was it kind of like this
00:40:20.160 plum pudding model where the protons and neutrons and electrons were just mashed together.
00:40:24.960 It took a while for us to escape from that view to moving towards something that looked kind of
00:40:30.240 like a solar system to today where you've got this set of fungible instances of an electron
00:40:36.720 in all birds with scare quotes around the world. But the nucleus where the nucleus is made of
00:40:43.920 protons which are really made of quarks and the electrostatic force and the Heisenberg uncertainty
00:40:49.520 principle is sort of intention. And this is why you have atoms. So this is our modern understanding
00:40:54.400 and it is refuted all the previous theories of atomic theory. So is true at acids and bases.
00:41:01.920 So is true of gravity. And so this is our situation. And by the way, not just in science. You can
00:41:08.960 talk about theories of history. You can talk about theories of aesthetics. You can talk about
00:41:14.000 theories of morality. Now sure in some of these areas there are still debates raging. But
00:41:19.600 generally speaking, broadly speaking, we have our theory and explanation and there are no rivals.
00:41:27.120 When a problem arises then creativity needs to start. Then you need to start thinking up new
00:41:32.720 explanations. And if you're really lucky and you try really hard then you'll find the explanation,
00:41:38.640 the explanation. You want to, it's not you create too. You'll be in a very fortunate position if you
00:41:44.160 were able to create two rival solutions to the problem that you have. But that usually isn't what
00:41:50.080 happens. You've got an existing theory. You make an observation which doesn't seem to fit with
00:41:56.240 your existing theory. So you've got this problem. So then you create another theory. And so
00:42:00.800 then you figure, ah, this theory, this theory explains why that observation doesn't fit with
00:42:07.120 that theory. But my new theory, it does explain the observation. Then you've got, you're back to one
00:42:11.200 again. You've gone from one to one. This is what happens. Okay. So yeah, never mind corroborating,
00:42:17.600 never mind confirming, never mind saying that one theory is more probable than any other. You've
00:42:22.720 just, you've got the one, you've got the one. Okay. Let me keep on going. The crypto inductive
00:42:27.760 says having just been told basically that by David, quote, crypto inductive is here saying,
00:42:34.000 very interesting. I now understand the role of a theory's refuted rivals in the justification
00:42:38.720 of its predictions under inductivism. Observation was supposed to be primary,
00:42:44.800 skipping a little. And he goes on to say in the popularion picture of scientific progress,
00:42:49.200 it is not observations, but problems, controversies, theories and criticism that are primary,
00:42:56.560 end quote, perfect, wonderful. That's exactly right. The crypto inductivist has got it. He goes
00:43:03.600 under say, experiments are designed and performed only to resolve controversies. Therefore,
00:43:08.880 only experimental results that actually do refute a theory and not just any theory. It must have
00:43:13.840 been a genuine contender in a rational controversy, constitute corroboration,
00:43:19.760 pausing their my reflection. Yeah, I don't need the corroboration bit, right? So you've got
00:43:25.280 therefore only experimental results that actually do refute a theory and not just any theory,
00:43:30.560 it must have been a genuine contender in a rational controversy. We don't need to say that
00:43:36.640 constitutes corroboration, although as I said in my last episode, unless of course you think
00:43:40.480 this corroboration thing is a synonym for refutation, which would be a bit weird, we don't need
00:43:47.520 both. We can just say we've got these competing theories, one of which gets refuted leaving us
00:43:52.800 with only one. Now, you could call that process. Well, we've corroborated this one existing theory.
00:43:59.600 It just gets a little bit misleading. That's all I would say. It suggests that we've confirmed
00:44:05.440 there's more likely true, something like that. It contains more truth. It's closer to describing
00:44:12.000 reality than the one that has been refuted, of course. It's necessarily the case. Skipping a bit
00:44:17.360 and the crypto inductivist goes on to say. Suppose that a theory has passed through this whole
00:44:23.760 process. Once upon a time it had rivals, then experiments were performed and all the rivals were
00:44:29.040 refuted, but it itself was not refuted. Thus it was corroborated. What is it about being corroborated
00:44:36.400 that justifies ever lying on it in the future? David says, since all its rivals have been refuted,
00:44:43.040 they are no longer rationally tenable. The corroborated theory is the only rationally
00:44:48.320 tenable theory remaining, pausing the MRI reflection. Again, I think this might be the last time
00:44:52.720 I just go back and say this sort of thing. Again, it would seem to me that we can just do away
00:45:00.080 with that way of saying things. Instead of saying the corroborated theory is the only rationally
00:45:04.560 tenable theory remaining, you just say the unrefuted theory is the only rationally tenable
00:45:10.000 theory remaining. Again, if corroborated means unrefuted very well, the crypto inductivist goes on
00:45:16.720 to say it. That only shifts the focus from the future import of past corroboration to the
00:45:22.640 future import of past refutation. The same problem remains why exactly is an experimentally
00:45:28.560 refuted theory not rationally tenable? Is it that having even one false consequence implies
00:45:35.520 that it cannot be true? David says, yes, and the crypto inductivist goes on to say, but surely
00:45:41.840 as regards the future applicability of the theory that is not a logically relevant criticism.
00:45:47.920 Admittedly, a refuted theory cannot be true universally, in particular it cannot have been true in
00:45:54.400 the past when it was tested, but it could still have many true consequences, and in particular
00:45:59.120 it could be universally true in the future. And David responds, this true in the past and true in
00:46:04.960 the future terminology is misleading. Each specific prediction of a theory is either true or false.
00:46:11.760 That cannot change. What you really mean is that though the refuted theory is strictly false,
00:46:17.440 because it makes some false predictions, all its predictions about the future might know that
00:46:22.160 less be true. In other words, a different theory, which makes the same predictions about the future,
00:46:27.200 but different predictions about the past might be true. The crypto inductivist says,
00:46:31.760 if you like. So instead of asking why a refuted theory is not rationally tenable,
00:46:36.160 I sure strictly speaking of asked, why does the refutation of a theory also render untenable
00:46:42.000 every variant of the theory that agrees with it about the future? Even a variant that has not been
00:46:47.920 refuted, David says. It is not that refutation renders such theories untenable. It is just that
00:46:54.640 sometimes they already are untenable by virtue of being bad explanations. And that is when
00:47:03.120 science can make progress. For a theory to win an argument, all its rivals must be untenable.
00:47:08.560 And that includes all the variants of the rivals, which anyone has thought of. But remember,
00:47:13.520 it is only the rivals which anyone has thought of that need be untenable. For example, in the case
00:47:18.880 of gravity, no one has ever proposed a tenable theory that agrees with the prevailing one in all its
00:47:24.320 tested predictions, but differs in its predictions about future experiments. I am sure that such
00:47:30.080 theories are possible. For instance, the successful to the prevailing theory will presumably be one of
00:47:35.840 them, but if no one has yet thought of such a theory, how can anyone act upon it? And the
00:47:41.440 crypto inductive says, what do you mean no one has yet thought of such a theory? I could easily
00:47:46.320 think of one right now. And David says, I very much doubt that you can. And the crypto inductive
00:47:51.520 says, of course I can. Here it is. Whenever you day the jump from higher places in ways
00:47:56.560 that would, according to the prevailing theory, kill you, you float instead. Apart from that,
00:48:01.520 the prevailing theory holds universally end quote. Okay. I am not going to read the next few pages
00:48:10.000 of, well, I'll read a little bit. I'll read a little bit, but basically the whole idea is
00:48:14.960 here that the crypto inductive is claims to have invented a theory on the spot,
00:48:18.400 namely that the prevailing view of gravity holds always all the time universally,
00:48:23.520 except that in this particular case, just ad hoc, he floats. Now, I say, as David will go on
00:48:29.760 to say, of course, this is not a theory. This is not an explanation. This is not genuinely
00:48:35.440 solving any problem at all. It's just an ad hoc modification with an assumption that comes out of
00:48:43.200 nowhere that the whole purpose of science is a problem solving exercise. That's what we're doing.
00:48:48.480 This solves no problems. In fact, it introduces problems. It breaks the existing theory. And this
00:48:53.200 is what David says. So let me just read the relevant part, because just as with David, it was David's
00:48:59.840 character. The crypto inductive is an irritating person. These people almost exist in real life.
00:49:08.320 Anyway, so the crypto inductive says, what's wrong with this theory? Why can't I just
00:49:12.720 to make this ad hoc modification, you know, what fault, what mistake have I made? David explains
00:49:19.440 quote, just about every fault in the Perparian book, your theory is constructed from the prevailing
00:49:25.120 one by appending an unexplained qualification about me floating. That qualification is in effect
00:49:32.000 a new theory, but you have given no argument either against the prevailing theory of
00:49:36.880 my gravitational properties or in favor of the new one. You have subjected your new theory to no
00:49:43.280 criticism other than what I'm giving it now, and no experiments testing. It does not solve,
00:49:48.560 or even purport to solve any current problem. Nor have you suggested a new interesting problem
00:49:54.880 that it could solve. Worst of all, your qualification explains nothing but spoils the explanation
00:50:01.520 of gravity that is the basis of the prevailing theory. It is this explanation that justifies our
00:50:06.720 relying on the prevailing theory and not on yours. Thus, by all rational criteria, your proposed
00:50:12.960 qualification can be summarily rejected, end quote. I think that, you know, this is the point at
00:50:20.640 which the judge comes along and hammers the gavel, that's case closed kind of thing. But still,
00:50:27.360 the crypto inductive is going on and on and on trying to say the ways in which, well, you know,
00:50:33.360 you're still not justified. You get to a point the crypto inductive is where he says, well, I could use
00:50:38.480 this new verb on X floats to describe situations in which you might fall to the ground, but in other
00:50:47.840 cases, you just happen to float unsupported so I can invent new language and stuff like that.
00:50:53.360 So he says, crypto inductive says, but when the theory is translated into my language,
00:50:57.840 no qualification is manifest. And on the contrary, a manifest qualification appears in the very
00:51:02.800 statement of the prevailing theory and David says, so it does, but not all languages are equal.
00:51:08.560 Languages are theories. In the vocabulary and grammar, they embody substantial assertions about
00:51:14.160 the world. Whenever we state a theory, only a small part of its content is explicit. The rest
00:51:20.960 is carried by the language. Like all theories, languages are invented and selected for their
00:51:26.400 ability to solve certain problems. In this case, the problems are those of expressing other theories
00:51:33.040 in forms in which it is convenient to apply them and to compare and criticize them.
00:51:37.840 One of the most important ways in which languages solve these problems is to embody implicitly
00:51:43.520 theories that are uncontroversial and taken for granted, while allowing things that need to be
00:51:49.120 stated or argued about to be expressed succinctly and cleanly. And the crypto inductive accepts
00:51:55.520 that. And David goes on to say, thus it is no accident when a language chooses to cover the
00:52:00.400 conceptual ground with one set of concepts rather than another. It reflects the current state of
00:52:05.760 the speaker's problem situation. That is why the form of your theory in English is a good indication
00:52:11.680 of its status, vis-Ã -vis the current problem situation, whether it solves problems or exacerbates
00:52:17.600 them, but it is not the form of your theory that I'm complaining about. It is the substance.
00:52:23.200 My complaint is that your theory solves nothing and only exacerbates the problem situation.
00:52:29.200 This defect is manifest when the theory is expressed in English, and implicit when it is expressed
00:52:34.720 in your language, but it is no less severe for that. I could state my complaint equally well in
00:52:39.360 English or in scientific jargon or in your proposed language or in any language capable of
00:52:44.160 expressing the discussion we have been having. It is a superior maxim that one should always
00:52:48.560 be willing to carry on the discussion in the opponent's terminology end quote. Now, I would just
00:52:54.000 say there is a lot there that is extremely useful for day to day life. If you are engaged in
00:52:59.200 a discussion with someone, especially in philosophical discussions where people tend to like to
00:53:04.720 just philosophise in completely abstract terms, divorce from physical reality or any other kind of
00:53:12.480 reality for that matter, a thought experiment, a trolley problem, and that kind of thing,
00:53:17.120 what you want to say to them in these situations is a simple question. Ask a simple question,
00:53:22.960 what problem are you solving? What problem are you solving? Okay, and let them go down the road,
00:53:29.760 and if it's a completely abstract problem about, you know, all these people are tied to a
00:53:34.000 railway and you have a lever and say, well, hold on, that is an imaginary situation. Can we bring
00:53:40.160 it back to something in the real world? And let's talk about something in the real world and let's
00:53:43.680 get to, let's say, moral principles or what's physically possible in the real world. And what one
00:53:48.560 would really do in the real world because that's imaginary. That's never happened before. And
00:53:54.160 insofar as things like that might have happened somewhere at some point, we call those things edge
00:53:58.960 cases. It doesn't affect the general rule, the general approach to life that people have.
00:54:04.480 And also in these situations and what David has just hinted at there is that language is there
00:54:08.960 to solve a problem. And so again, once more, you listen to certain philosophers on certain podcasts
00:54:14.880 at times, people get interviewed, and they invent words, they invent terms of vocabulary,
00:54:20.800 and they think that by inventing these words, they're solving some problem. But in fact,
00:54:24.000 what they're doing, they're exacerbating the problem. They're generating issues that weren't there
00:54:30.960 before, the prevailing existing view, the way of talking about these things, solve certain
00:54:37.360 problems, might have certain open problems, but their invention of new words, it just introduces
00:54:42.240 yet more problems without solving anything existing. They're often trying to say, I've discovered
00:54:46.640 something because I've invented a term. I hear I've got a piece of jargon, and that's
00:54:51.440 solved. It doesn't dissolve the problem at all. You're describing the problem using new bits of
00:54:57.200 language and inventing that's proliferation of language, neologisms is just something I always irritate.
00:55:03.440 Sometimes it's unavoidable. Of course, when a scientist, when I think out, really truly does
00:55:09.280 come up with a solution because they postulate the existence of entities that either
00:55:14.880 two we didn't know about. Of course, you need a new word for that. You need the word
00:55:18.000 electron. When someone has figured out that there is this little particle that carries this
00:55:22.160 little negative charge, when you need a word for that, fine, fine. We accept that. But there are
00:55:26.960 many cases where the proliferation of language seeks to obscure what's really going on,
00:55:32.720 rather than solving an actual problem. Why's ask? What problem are you trying to solve?
00:55:38.160 You're introducing this new term. What is the specific problem you're trying to solve? Let's talk
00:55:42.160 about solutions rather than inventing words that are only describing a problem that's out there.
00:55:48.880 I'll hop off my hobby course now. I'll pick it up where David is going on to just
00:55:55.040 hammer his point home to the crypto inductivist. He says, quote, your theory asserts the existence
00:56:00.480 of a physical anomaly, which is not present according to the prevailing theory. The anomaly is
00:56:06.240 my alleged immunity from gravity. Certainly, you can invent a language which expresses this anomaly
00:56:11.680 implicitly so that statements of your theory of gravity need not refer to it explicitly.
00:56:16.720 But refer to what they do arose by any other name would smell a sweet. Suppose that you,
00:56:22.080 indeed, suppose that everyone were a native speaker of your language and believed your theory of
00:56:26.800 gravity to be true. Suppose that we all took it entirely for granted and thought it's so natural
00:56:32.080 that we use the same word for X for to describe what you or I would do if we jumped over the
00:56:37.520 railing. None of that alters in the slightest degree. The obvious difference there would be between
00:56:43.600 my response to gravity and everything else's. If you fell over the railing, you might well envy
00:56:49.440 me on the way down. You might well think if only I could respond to gravity as David does,
00:56:54.960 rather than in this entirely different way that I do, and perfectly right. Remember what's
00:57:01.040 going on? The guy is claiming the crypto inductive is claiming that if David jumps over you know
00:57:05.360 he's going to float. But everywhere else, every other time anything else happens, the prevailing
00:57:10.640 theory just operates as normal. This is one of those situations where you can imagine a philosopher
00:57:16.720 making this argument. He's like, well, why are you even asking me, what problem are you solving?
00:57:21.280 They're saying, well, I want to figure out exactly why we should justify the predictions of our
00:57:26.320 existing explanation. Well, I'm telling you, there is only one explanation. And then they go down
00:57:30.480 this long road where it even requires the invention of a new language in order to try and
00:57:36.720 explain a problem that doesn't exist. It's just not a problem. It just is an issue that David is
00:57:42.240 going to float and everything else is going to fall. That's not the situation we're in. If it was,
00:57:48.240 then that would be a genuine problem as David says there. Then everyone would be wondering,
00:57:52.160 why is it that the laws of gravity appear to have selected David Deutsch out of everyone in the
00:57:57.600 entire universe for special treatment? What's going on there? That would be a problem,
00:58:02.800 and there would need to be an account of that. But that's not the situation we're in.
00:58:06.320 It's all in the crypto inductive. It's head, kind of like a trolley problem, right? It's all in
00:58:10.720 your head that you need to push the fat man off the bridge in order to save the others. It's in
00:58:16.480 your head. That's not happening. Do we talk about something real? Then that's where the
00:58:22.400 interesting philosophy comes in. Philosophy is most interesting. Not when you're talking about
00:58:26.800 to my mind, the things like Descartes Demon and the simulation hypothesis and a far distant future
00:58:35.600 existential threats. It's the here and now. It's problems right now that need addressing right now,
00:58:42.080 and which people are struggling to find answers to. And there may be some people just give up
00:58:45.680 and they go, oh, that's too hard. I'm not going to worry about trying to deal with the issue
00:58:50.320 right now of things like coercion and existence society. Let me talk about the far distant future.
00:58:57.760 I can deal with that. I can deal with the science fiction reality of the year 3000 when
00:59:04.080 these, the AIs are going to take over. That's more fun to talk about.
00:59:07.600 This is philosophy though. This is sometimes what passes for philosophy. And the crypto inductive
00:59:17.440 is still just rattling on. I encourage you to read the entire chapter. I'm not going to, as I say,
00:59:22.800 so I'm skipping a number of pages and I'll just pick it up where David says, quote,
00:59:26.880 theories postulating anomalies without explaining them are less likely than their rivals to make
00:59:32.880 true predictions. More generally, it is a principle of rationality that theories are postulated
00:59:38.320 only in order to solve problems. Therefore, any postulate which solves no problem is to be rejected.
00:59:44.960 That is because a good explanation qualified by such a postulate becomes a bad explanation.
00:59:50.320 End quote. Yes. And as I was just saying about, you know, philosophers who like to consider
00:59:55.280 problems of the distant future, what's going to happen in the year 2500 when the AGI are finally
1:00:03.040 here? Are they going to take over all of humanity? It's like you're inventing a problem of the
1:00:07.840 future. We're not there yet. That's not our problem right now. You know, how we should start
1:00:13.760 preparing for war with the AGI or with the aliens that we haven't detected yet. This kind of
1:00:20.160 stuff is solving no problem. It's not really part of rational discussion. I mean, it's fun,
1:00:26.800 but it's more in the realm as I like to say of science fiction. Now, science fiction can be useful.
1:00:32.640 And, you know, this is kind of a fun thing to do late on a Friday night after you've had a
1:00:38.960 beer or wine with a friend to discuss the possibilities of the encounter between alien life and
1:00:45.680 humanity and what problems might be involved in. But it's not our problem situation right now.
1:00:50.320 It's this is not the problem we have right now. There are problems right now. I'm not saying anyone
1:00:54.640 should not necessarily think about those things, but those things can't be fun. What I'm saying is
1:00:59.360 that that's not a part of serious philosophical discussion and rationality. As much as
1:01:06.400 people get out there onto podcasts and start talking about these things, you're not solving any
1:01:09.680 problems right now. There are problems right now. There are problems right now that a philosopher
1:01:14.320 should direct their attention to. People love to listen to stories and hypotheses about what
1:01:19.840 UFOs are and people speculate about what these different things are. But in so far as it's a problem,
1:01:24.880 it's not a serious problem. If the Star Destroyer populated by stormtroopers appears in the sky and
1:01:30.000 is as big as an entire city, then we've got a problem. Let's talk about it then. Should we prepare
1:01:35.040 for that? No. Why not? Why shouldn't we prepare for the distant future of things? Because
1:01:39.680 there is an infinite number of things we can imagine that are terrible off into the
1:01:43.040 distant future. But learned people all tend to agree that this particular problem about catastrophic
1:01:49.120 AI, apocalypse is the most important problem. I don't care. I don't care that there's so many of
1:01:54.160 these people talking about that particular thing. It's not our situation now. I think that those
1:01:58.960 people have perhaps too much time on their hands. They should be focusing their brainpower on
1:02:05.600 problems of today, right now, rather than being concerned about existential threats of tomorrow.
1:02:13.440 But of course, people can do what they like. Okay, I want to pick up something that I mentioned
1:02:20.240 my preface in the immediate previous episode of Topcast when I was talking about corroboration.
1:02:26.160 And the issue is this, let me just read apart here. And the crypto inductivist is saying that
1:02:32.480 he's beginning to agree with David. And he says, quote, now that I understand there really is
1:02:37.680 an objective difference between theories which make unexplained predictions and theories which
1:02:42.400 don't, I must admit that this does look promising as a solution to the problem of induction.
1:02:46.880 You seem to have discovered a way of justifying your future alliance on the theory of gravity,
1:02:50.560 given only the past problem situation, skipping a little in David says, it was not AI who discovered
1:02:56.240 this. And the crypto inductivist says, well, I don't think Papa did either. For one thing,
1:03:00.960 Papa did not think that scientific theories could be justified at all. You make a careful distinction
1:03:06.560 between theories being justified by observations as inductive as think and being justified by argument.
1:03:13.040 But Papa made no such distinction. And in regard to the problem of induction,
1:03:16.960 he actually said that although future predictions of a theory cannot be justified,
1:03:20.880 we should act as though they were. And David says, I don't think he said that exactly.
1:03:26.560 If he did, he didn't really mean it, pausing their just my reflection. Yeah, I kind of get this
1:03:31.440 impression as well from my reading of Papa, especially recently, and as I said in the last episode,
1:03:36.400 Papa just wrote so many thousands of words, thousands of pages, thousands of words.
1:03:43.280 And so it gets hard. It's like me talking here on talk guys, you kind of, it's hard to keep track
1:03:49.760 of what you did say in the past. I feel as though I've got a reasonably coherent worldview,
1:03:54.960 and people ask me about aspects of it, and I can respond near immediately to what I think is the case.
1:04:01.520 But I can certainly imagine miss speaking now and again, or not quite constructing the sentences,
1:04:06.720 precisely in the same way that I did in the past as I will in the future, because it's just hard
1:04:13.120 to keep track of everything. We're human. Now, Papa was operating in a time where he's
1:04:16.960 writing by hand or he's typing or whatever it happens to be the case and trying to keep track of
1:04:21.360 everything he's written. Indeed, he says of himself that he changed his mind between the logic
1:04:26.320 of scientific discovery through realism in the aim of science and various other things.
1:04:30.720 And I do get the impression that he said, well, you know, he didn't really mean to say things
1:04:34.800 like, you can confirm your theory. Okay, there are confirming instances of the theory. That's one
1:04:40.560 thing he actually says. He actually rejects a version of himself in realism in the aim of science.
1:04:47.360 And so, let me just repeat what David says there, or basically on the same topic.
1:04:52.080 The crypto inductive has suggested, in regard to the problem of induction, he, Papa,
1:04:56.640 said that although future predictions of a theory cannot be justified, we should act as though
1:05:00.240 they were. David says, I don't think he said that exactly. If he did, he didn't really mean it.
1:05:06.560 The crypto inductive says, what? And David says, or if he did mean it, he was mistaken.
1:05:12.400 Why are you so upset? It is perfectly possible for a person to discover a new theory.
1:05:17.680 In this case, popularing epistemology, but nevertheless, to continue to hold beliefs,
1:05:22.720 that contradicted the more profound the theory is, the more likely this is to happen.
1:05:28.240 Crypto inductive says, are you claiming to understand Papa's theory better than he did himself?
1:05:34.160 And David says, I know the no nor care. The reverence that philosophers show for the
1:05:39.120 historical sources of ideas is very perverse, you know. In science, we do not consider
1:05:45.760 the discovery of a theory to have any special insight into it. On the contrary,
1:05:52.160 we hardly ever consult original sources. They invariably become obsolete, as the problem
1:05:58.880 situations that prompted them are transformed by the discoveries themselves. For example,
1:06:05.520 most relativity theorists today understand iron science theory better than he did.
1:06:11.120 The founders of quantum theory made a complete mess of understanding their own theory.
1:06:15.520 Such shaky beginnings are to be expected. And when we stand upon the shoulders of giants,
1:06:20.960 it may not be all that hard to see further than they did. But in any case, surely it is more
1:06:27.040 interesting to argue about what the truth is than about what some particular thinker, however
1:06:33.040 great, did or did not think end quote. Isn't that marvelous? That's perfect. I agree entirely
1:06:39.760 with that. No one learning science learns from the original research papers. No one tempted to learn
1:06:46.000 special relativity. You should go back to iron science original paper, presumably in German,
1:06:50.320 by the way. No one consults the original struggling attempts to understand the photoelectric
1:06:57.200 effect and the beginnings of quantum theory. No, you don't even go to ever its papers. You go to
1:07:02.800 the beginning of infinity. You go to modern accounts of how to understand this stuff, modern textbooks
1:07:08.640 in many, many cases, explaining so much better. Sometimes the scientists, people can be brilliant
1:07:15.680 scientists and make great insights and yet be terrible communicators of their ideas. It takes
1:07:21.280 other reasonably good scientists who are better communicators to distill out what is actually being
1:07:26.800 said. And then perhaps textbook writers who are a different kind of person again to distill out what
1:07:33.120 the good communicators of science are actually saying about the truth of science. And then you end up
1:07:39.200 with textbooks, which can be quite good accounts of our best existing theories. And then eventually
1:07:44.880 after all of that, you might get popular accounts of what's going on, which can be even better.
1:07:50.080 Sometimes worse, sometimes misleading. Sometimes they introduce yet new misconceptions. But
1:07:54.960 you know, this is if you want to understand something is why you should read widely a whole bunch
1:08:00.160 of different things. Certainly this has been my experience of reading some of the research papers,
1:08:05.840 reading some of the textbooks and then reading some of the popular science books and being able to
1:08:11.040 see what they all have in common where they agree on what the experiments are saying, what actually
1:08:16.800 makes rational sense, what actually accounts for what's going on, what has the best explanations,
1:08:25.200 which is one of the reasons I am so fixated on David Deutsch's worldview, which talks so
1:08:30.640 much about explanations, puts explanations at the center of rational understanding that if you
1:08:36.560 don't understand something, that's a problem. But it might not be you that's the problem.
1:08:41.840 It could be the person ostensibly doing the explaining, the textbook, the lecturer, whatever.
1:08:48.720 They could just be non understanding the phenomena themselves and giving you a non explanation,
1:08:54.320 being evasive. And so you have to somehow or other find the person that knows the stuff and
1:08:59.600 is able to explain it well. What you don't want in physics, for example, is an instrumentalist.
1:09:04.400 And yet they are a dime a dozen. They're out there and they just say, well, let's just crank
1:09:09.120 through the formula. Let's just figure out how to do the calculations. There you go, you're a physicist.
1:09:14.880 Well, you know, you might be competent at the mathematics at turning the handle,
1:09:20.160 which at one end you put in your initial conditions and out the other end you get your final
1:09:24.320 conditions, you get your solution. And you don't really understand what's going on. You don't
1:09:28.240 understand what the turning of the handle is really all about. You may not appreciate what
1:09:33.760 reality really consists of. In fact, you may have even been told, you can't understand what reality
1:09:39.760 consists of, not even approximately. There is no understanding of reality. The best you can do
1:09:44.640 is to predict the outcome of experiments. And this is all wrong. Okay, this is all wrong.
1:09:50.320 But people who first discover a theory, they're struggling to understand stuff. And so
1:09:55.360 they're throwing stuff at the blackboard and the whiteboard and, you know, writing papers and
1:10:00.720 trying to get other people to see if they can understand it as well and know if they agree with
1:10:04.720 the results and what their conclusions might be and people are hypothesizing things. And it takes
1:10:08.800 some time for the dust to settle, so to speak. Okay, so I'm going to pick it up where the crypto
1:10:13.840 inductivist almost gets the idea and then falls back into bad ways of thinking because they say,
1:10:20.720 quote, look, you David have justified a theory about the future, the prevailing theory of gravity,
1:10:27.600 as being more reliable than another theory, the one I proposed. Even though they are both consistent
1:10:33.360 with all currently known observations, since the prevailing theory applies both to the future and
1:10:39.200 to the past, you have justified the proposition that, as regards gravity, the future resembles the past.
1:10:45.680 And the same would hold whenever you justify a theory as reliable on the grounds that it is
1:10:50.400 corroborated. Now, in order to go from corroborated to reliable, you examine the theory's
1:10:56.560 explanatory power. So what you have shown is that what we might call the principle of seeking
1:11:02.160 better explanations, together with some observations, yes and arguments, imply the future will,
1:11:08.160 in many respects resemble the past, and that is a principle of induction. If your explanation
1:11:15.760 principle implies a principle of induction, then logically it is a principle of induction.
1:11:21.120 So inductivism is true after all, and a principle of induction does indeed have to be
1:11:25.920 postulated, explicitly or implicitly, before we can predict the future. And David says,
1:11:30.560 oh dear, this inductivism really is a virulent disease. Having gone into a mission for only a
1:11:36.720 few seconds, it now returns more violently than before. And the crypto inductivist says,
1:11:41.920 does perpyrean rationalism justify ad hominem arguments as well? I ask for information only,
1:11:48.720 David says, hi, apologies. Let me go straight to the substance of what you said. Yes,
1:11:54.240 I have justified an assertion about the future. You say this implies that the future resembles the
1:12:00.320 past. Well, evacuously, yes, in as much as any theory about the future would assert that it
1:12:06.880 resembled the past in some sense. But this inference that the future resembles the past is not
1:12:13.440 the sort after principle of induction. For, we could neither derive nor justify any theory or
1:12:18.880 prediction about the future from it. For example, we could not use it to distinguish your theory
1:12:23.840 of gravity from the prevailing one for they both say in their own way that the future resembles
1:12:28.880 the past. Yes, end quote. And by the way, he, an explanation that sort of suggests that the
1:12:36.480 part resembles a future, as David says there, well, that's a vacuous claim. It's like a universal
1:12:41.600 theory of gravity or anything else is universal, which means the future resembles the past, the
1:12:47.680 past resembles the future, the near resembles the far in this respect of obeying that law. That's
1:12:54.560 all. It's like, you know, any claim in relativity about the constancy of the speed of light, the
1:12:59.520 speed of light is constant. The speed of light was constant at the beginning of the universe. The
1:13:03.360 speed of light was constant 10 billion years ago. The speed of light was constant one billion years
1:13:07.520 ago. The speed of light is constant today and the speed of light will be constant tomorrow.
1:13:10.320 That's not a principle of abduction. That's just because you're saying that these things are the
1:13:14.640 same is not going to justify some theory. It's just a logical conclusion, a logical thing that
1:13:22.960 you can say that falls out of what we understand about relatively our best explanation of space,
1:13:29.600 time, and light in this particular case. As David goes on to say, and I'm skipping it, but he
1:13:34.160 says, quote, nothing in the concepts of rational argument or explanation relates the future to the
1:13:39.760 past in any special way. Nothing is postulated about anything resembling anything. Nothing of that
1:13:45.520 sort would help if it were postulated in the vacuous sense in which the very concept of
1:13:50.720 explanation implies that the future resembles the past, it nevertheless implies nothing specific
1:13:56.480 about the future. So it is not a principle of induction. There is no principle of induction.
1:14:01.760 There is no process of induction. No one ever uses them or anything like them. And there is no
1:14:07.120 longer a problem of induction. Is that clear now? And the crypto inductive says, yes, please excuse
1:14:14.480 me for a few moments while I ingest my entire worldview. And David goes on to explain, well,
1:14:20.880 let me read part of his explanation. He says, quote, in response to the crypto inductive.
1:14:26.240 As we have agreed, your theory consists objectively of a theory of gravity, the prevailing theory,
1:14:32.960 qualified by an unexplained prediction about me. It says that I would float unsupported,
1:14:39.120 unsupported means without any upward force acting on me. So the suggestion is that I would be
1:14:44.400 immune to the force of gravity, which would otherwise pull me down. But according to the
1:14:47.520 general theory of relativity, gravity is not a force, but a manifestation of the curvature of space
1:14:52.480 time. This curvature explains why unsupported objects like myself and the earth move closer together
1:14:58.320 with time. Therefore, in the light of modern physics, your theory is presumably saying there is an
1:15:03.840 upward force on me as a required to hold me at a constant distance from the earth. But where does
1:15:09.360 that force come from? And how does it behave? End quote. David does go on. But yeah, we can just see
1:15:14.400 here that this unsupported assertion about reality that the crypto inductive has introduced creates
1:15:21.440 more problems. It doesn't solve anything. It creates more problems. And our project is in science
1:15:26.480 and everywhere else to solve problems so that we can move on to better problems. Not introduce
1:15:32.000 new problems that weren't there before, so that we've actually ruined the existing solution.
1:15:36.720 No, that's not what we're supposed to be doing. And this, by the way, is why, you know,
1:15:40.800 there's this trope idea among physicists that retired engineers tend to go into proving Einstein
1:15:48.960 wrong. You know, using simple algebra, they show you that, well, the speed of light can't be
1:15:54.720 constant. Or you can travel fast in the speed of light. Or any number of things that there's no
1:15:59.120 such thing as the relativity of similar to the relative of time. And all this kind of stuff.
1:16:02.880 So the problem with these approaches for any physicists to receive such a letter from a retired
1:16:10.800 engineer to be done fair to these poor retired engineers. I'm sure they're not all retired engineers,
1:16:14.800 but physicists know what I'm talking about. They receive emails and letters of this kind,
1:16:19.360 you know, such and such has got the new theory of relativity. They're never solving any problem.
1:16:27.280 They're never solving an existing problem in science. What they're doing is they're objecting
1:16:31.920 on common-sense grounds to some conclusion of relativity, let's say. They don't like the idea that
1:16:39.040 there should be a relativity of similarity that things appear to happen at the same time or not,
1:16:47.280 depending upon whether you're moving or not with those things that are happening at the same time.
1:16:52.880 Or they don't like the idea that lengths can contract. If you move faster, then things get shorter.
1:17:00.720 They don't like the idea that the speed of light is constant. Any number of things like this.
1:17:05.360 And so because they don't like it, they invent a new theory that far from solving any problems
1:17:10.880 breaks the existing theory, the theory of relativity and introduces new problems. So then it means
1:17:16.160 that, well, we can't explain what's going on if we were to use the retired engineer's
1:17:20.480 explanation of reality. We can't explain what's going on in places like the Large Hadron Collider.
1:17:26.720 We can't explain what's going on with the GPS system, which we can explain perfectly well
1:17:32.240 using general relativity and in many cases, special relativity as well. And so this is the issue here.
1:17:37.360 You aren't solving a real problem. What you're doing in those situations is you don't understand
1:17:45.120 something and you're trying to understand something and rather than trying harder to understand
1:17:50.160 something by asking questions of people who might know more, who in a patient way will explain it to you,
1:17:56.560 you just say, I'm throwing it all out and I'm telling you what the truth is.
1:18:01.280 Being a little dogmatic, let's say, after a little more back and forth, we again come back to
1:18:05.680 David saying, quote, your additional postulate is not just superfluous, it is positively bad.
1:18:11.040 In general, perverse but unrefuted theories, which one can propose off the cuff for roughly
1:18:17.200 into two categories. There are theories that postulate unobservable entities such as particles
1:18:22.720 that do not interact with other matter. They can be rejected for solving nothing,
1:18:27.120 or comes razor if you like. And there are theories like yours that predict and explain
1:18:32.400 observable anomalies. They can be rejected for solving nothing and spoiling existing solutions.
1:18:38.560 It is not, I hasn't to add, that they conflict with existing observations. It is that they
1:18:43.760 remove the explanatory power from existing theories by asserting that the predictions
1:18:49.280 of those theories have exceptions, but not explaining how, end quote, the crypto inductor
1:18:54.240 that says, I see that now. Now will you give me help in adjusting my worldview? David says,
1:19:00.480 well, have you read my book The Fabric of Reality? But what the crypto inductor this thing goes on
1:19:05.680 to ask of David? Well, he says, quote, what I cannot understand is, where in that raw material
1:19:11.840 past observations, the present problem situation and timeless principles of logic and rationality,
1:19:16.960 none of which justifies inferences from the past into the future, the justification of future
1:19:21.920 predictions has come from. There seems to be a logical gap. Are we making a hidden assumption
1:19:27.360 somewhere? And David responds, no, there is no logical gap. What you call our raw material does
1:19:32.800 indeed include assertions about the future. The best existing theories, which cannot be abandoned
1:19:37.600 lightly because they are the solutions of problems, contain predictions about the future,
1:19:42.080 and these predictions cannot be severed from the theories of the content as you tried to,
1:19:47.040 because that would spoil the theories explanatory power. Any new theory we propose must,
1:19:52.320 therefore, either be consistent with these existing theories, which has implications for what the
1:19:57.600 new theory can say about the future, all contradict some existing theories, but address the problems
1:20:03.520 thereby raised, giving alternative explanations, which again constrains what they can say about the
1:20:09.120 future and the crypto inductor says. So we have no principle of reasoning, which says that the
1:20:14.320 future will resemble the past, but we do have actual theories which say that. So do we have
1:20:20.640 actual theories which imply a limited form of inductive principle? David says, no. Our theory
1:20:26.960 simply assert something about the future. Vacuously, any theory about the future implies that
1:20:32.320 the future will resemble the past in some ways, but we only find out in what respects the theory
1:20:38.480 says that the future will resemble the past after we have the theory. You might as well say that
1:20:44.000 since our theories hold features of reality to be the same throughout space, they imply a spatial
1:20:49.520 principle of induction to the effect that the near resembles the distant. Let me point out that in
1:20:54.800 any practical sense of the word resemble, our present theory say that the future will not resemble
1:20:59.920 the past. The cosmological big crunch, for instance, the recolapse of the universe to a single
1:21:03.920 point, is an event that some cosmologists predict, but which is just about as unlike the present
1:21:09.360 epoch in every physical sense as it possibly could be, the very laws from which we predict its
1:21:14.080 currents will not apply to it. End quote. And exactly the same argument applies with not
1:21:18.240 know of cosmology today. The far distant future apparently given what we know now, and this is
1:21:23.680 very much open to change given future observations in future theories. Of course,
1:21:29.920 is that this dark energy is going to continue to accelerate the expansion of space such that
1:21:34.400 everything gets ripped apart in this kind of big rip event where even atomic nuclei and fundamental
1:21:40.080 particles get torn apart and everything turns into photons. And at that point, if you listen to
1:21:45.200 someone like Roger Penrose, you have at the very end of time, after this happens,
1:21:50.400 it's something that looks very much like the beginning of time, namely a universe or a space that
1:21:56.720 contains no matter in nothing but photons, which is precisely what the big bang was like. Anyway,
1:22:02.080 that's off topic. But the point is that right now, the present doesn't resemble the past,
1:22:06.560 the big bang, and it doesn't resemble the future, which is this far distant big rip future.
1:22:11.280 Who knows? Now I'm skipping past a bit where they get into a discussion about what justifies
1:22:17.760 being true the principles of logic and how these justifications for logic isn't perfectly
1:22:24.000 secure, and David agrees it's not perfectly secure. We can't expect it to be what we want,
1:22:29.280 our good explanations, of course. We want explanations about why and how rationality works.
1:22:34.240 They come across the Turing principle, and the crypto inductor says, how we know it's true,
1:22:38.880 and David says, we don't know, of course. In fact, let me just read what he says on that,
1:22:43.920 when he's confronted with, is the Turing principle true? David says, we don't know, of course,
1:22:48.960 if it's true or not. But you are afraid, aren't you, that if we can't justify the Turing principle,
1:22:54.000 then we shall once again have lost our justification for relying on scientific predictions,
1:22:58.960 and the crypto inductor says, yes, David says, but we have now moved on to a completely different
1:23:04.400 question. We are now discussing an apparent fact about physical reality, namely that it can
1:23:09.200 make reliable predictions about itself. End quote, that's in the form of us. We are the thing that
1:23:14.640 makes reliable predictions about the future. So we are part of physical reality that makes reliable
1:23:20.080 predictions about physical reality. Continuing, David says, we are trying to explain that fact,
1:23:26.640 to place it within the same framework as other facts we know. I suggested that there may be a
1:23:31.920 certain law of physics involved, but if I were wrong about that, indeed, even if we were entirely
1:23:37.360 unable to explain this remarkable property of reality, that would not detract one jot
1:23:42.960 from the justification of any scientific theory for it would not make the explanation in such a
1:23:47.920 theory one jot worse. And the crypto inductor says, now my arguments are exhausted.
1:23:53.200 Intellectually, I am convinced, yet I must confess that I still feel what I can only describe
1:23:58.320 as an emotional doubt. And David says a few things, but then gets to the meat of the matter
1:24:04.960 and says, quote, the misconception is about the very nature of argument and explanation.
1:24:10.320 You seem to be assuming that arguments and explanations, such as those that justify acting on a
1:24:15.680 particular theory, have the form of mathematical proofs, proceeding from assumptions to conclusions.
1:24:22.720 You look for the raw material, axioms, from which our conclusions, theorems are derived.
1:24:29.520 Now, there is indeed a logical structure of this type associated with every successful
1:24:34.640 argument or explanation, but the process of argument does not begin with the axioms and end
1:24:40.400 with the conclusion. Rather, it starts in the middle, with a version that is riddled with
1:24:45.760 inconsistencies, gaps, ambiguities and irrelevancies. All these faults are criticized,
1:24:51.680 attempts are made to replace faulty theories. The theories that are criticized and replaced
1:24:56.960 usually include some of the axioms. That is why it is a mistake to assume that an argument begins
1:25:03.600 with, or is justified by, the theories that eventually serve as its axioms. The argument ends
1:25:10.720 tentatively when it seems to have shown that the associated explanation is satisfactory.
1:25:16.400 The axioms adopted an unultimate, unchallengeable beliefs. They are tentative,
1:25:22.160 explanatory theories. End quote. And all I would say there, just in other words,
1:25:28.400 is the argument ends tentatively when it seems to have solved the problem. Solve the problem
1:25:35.680 to the satisfaction of whoever had the problem, and that could be the community of scientists,
1:25:40.160 or an individual scientist working on a particular issue. So that's the way things go,
1:25:45.120 and it doesn't just have to be science, of course. This applies universally to all
1:25:49.280 problem situations and all kinds of knowledge creation. The crypto inductivist finally agrees,
1:25:54.240 and he says, quote, I see, argument is not the same species of thing as deduction,
1:26:00.800 or the non-existent induction. It is not based on anything or justified by anything,
1:26:06.160 and it doesn't have to be, because its purpose is to solve problems, to show that a given
1:26:13.680 problem is solved by a given explanation. And David says, welcome to the club.
1:26:19.200 And the crypto inductivist has now been retitled, X inductivist, and he says, all these years I
1:26:27.920 have felt so secure in my great problem, I felt so superior both to the ancient inductivists,
1:26:33.760 and to the upstart popper, and all the time without even knowing it, I was a crypto inductivist
1:26:38.640 myself. Inductivism is indeed a disease, it makes one blind. And there are a few more
1:26:45.680 marketing in there, but we may as well also throw into the bin here, the Bayesian, the modern incantation
1:26:52.240 of inductivist. They have this induction shaped hole in their epistemology. Almost everyone
1:26:59.680 who's not a perpyrean does, so-called objectives do as well, they seek a principle of induction,
1:27:06.080 this secure foundation from which they can derive explanations in the same way that mathematical
1:27:12.240 proofs are derived. It's as if, as David says elsewhere in the fabric of reality, there's this
1:27:17.280 hierarchy of knowledge creation, or of argumentation, where mathematics is the king of all
1:27:23.760 knowledge, and everything seeks to aspire to be like mathematics, to start with the secure
1:27:29.520 foundation, with axioms that are self-evident, that are absolutely true, that cannot be denied.
1:27:34.240 And from that, you just use your rules of inference, which presumably also are completely
1:27:38.640 inherent, that can only produce truth from truth, and you get, thereby, true conclusions.
1:27:44.800 This way of thinking about reality and the way of constructing knowledge is completely
1:27:49.200 misconceived, among other things, who says and who can prove, and why should we believe that
1:27:54.880 the axioms are themselves absolutely true? We don't know if the process is this method of
1:28:01.040 deduction in order to generate true conclusions, if that's the process, then where for the axioms,
1:28:07.680 how did we get these true axioms in the first place? As David says there, you start in the middle,
1:28:11.680 we start with the problem, and it's a messy kind of process of creating an explanation. No one
1:28:17.280 really knows how it happens yet. We don't have an explanation for creativity yet. We just know that
1:28:22.320 we come up with these creative explanations of reality. They are hard to vary, because each part of
1:28:29.680 the explanation corresponds to some part of reality, postulating into existence some physical thing
1:28:36.080 when it comes to science, something that is accounted for by the explanation. So that's where we'll
1:28:41.440 leave chapter 7 of the fabric of reality. Next time, once I do come back to the fabric of reality,
1:28:46.240 we'll be on to the significance of life, which chapter 8, all about evolution by natural selection,
