1 00:00:00,240 --> 00:00:06,610 Hi, everyone. I'm Marianne Talbot. Some of you know me and I know some of you, but not all of you. 2 00:00:06,610 --> 00:00:12,810 And I didn't know anyone who's watching the podcast. So hello to everyone watching the podcast. 3 00:00:12,810 --> 00:00:21,570 Okay, I'm going to. I've given six lecture titles for this series, but I'm going to be changing a few at the end. 4 00:00:21,570 --> 00:00:27,900 So you're going to get everything that you've signed up for, but you might get it in a different order or something. 5 00:00:27,900 --> 00:00:31,290 But today we're going to look at these things. 6 00:00:31,290 --> 00:00:34,980 We're going to, first of all, look at why is causation important? 7 00:00:34,980 --> 00:00:41,160 Because when I've talked over the summer to people about the lectures I'm writing, I'm writing lectures on causation. 8 00:00:41,160 --> 00:00:46,530 Lots of people said, what's causation, wife? Why is this important? 9 00:00:46,530 --> 00:00:55,230 So I think it's very important, actually, to have a look first. That's why we're bothering to think about causation. 10 00:00:55,230 --> 00:00:58,500 Then I'm going to look about how philosophers think about causation, 11 00:00:58,500 --> 00:01:05,640 because there's a difference to the way philosophers think about causation, to the way that scientists, for example, think about causation. 12 00:01:05,640 --> 00:01:10,950 And we'll have a little look at but what that way is and why it's different. 13 00:01:10,950 --> 00:01:16,350 Then I'm going to look at Hume's regularity theory. And the reason I'm going to look at that is modern. 14 00:01:16,350 --> 00:01:23,070 Thinking about causation started with Hume. You cannot not start with Hume. 15 00:01:23,070 --> 00:01:26,640 Well, you could. I mean, a people like me have been doing it for years. 16 00:01:26,640 --> 00:01:32,790 Think of boring old Hume. But but of course, Hume isn't boring. Hume is really, really interesting. 17 00:01:32,790 --> 00:01:38,460 And we are going to start with him and his regularity theory. 18 00:01:38,460 --> 00:01:45,240 And so I'm going to explain what the regularity theory is. Then I'm going to look at problems for the regularity theory. 19 00:01:45,240 --> 00:01:55,170 And I'm going to finish today with the canonical statement of RTC, the regularity theory, because it's gone through many changes since Houdin. 20 00:01:55,170 --> 00:02:00,120 And I'll say which one? Those who adopt the regularity theory. 21 00:02:00,120 --> 00:02:04,050 And that's many, many philosophers still. Which statement of it. 22 00:02:04,050 --> 00:02:11,200 They would accept today. So that's what we're going to do today. And I'm going to start with why is causation important? 23 00:02:11,200 --> 00:02:16,180 And I would start off looking at why it's important for everyone, for everyone here. 24 00:02:16,180 --> 00:02:23,040 And what then I'm going to look at why it's important in particular for philosophers. And so for a start. 25 00:02:23,040 --> 00:02:31,260 Nearly every explanation you offer of almost anything you're trying to explain is going to be a causal explanation. 26 00:02:31,260 --> 00:02:39,960 And there are philosophers, David Lewis, whom you'll be introduced to next week, is someone who believes that all explanation is causal explanation, 27 00:02:39,960 --> 00:02:48,370 because all explanation gives some information about the causal history of an event. 28 00:02:48,370 --> 00:02:53,190 OK, so if you're thinking, well, why did that happen? 29 00:02:53,190 --> 00:02:56,280 You look for the cause of whatever that is. 30 00:02:56,280 --> 00:03:03,900 So if you're wondering about the explanation of B, if you can think that A causes B or Albie's are caused by A, 31 00:03:03,900 --> 00:03:08,010 then and there wasn't a minute ago then you've got your explanation of B. 32 00:03:08,010 --> 00:03:15,870 So explanations appeal to the relation of causation, the relation of between two events. 33 00:03:15,870 --> 00:03:20,070 And we'll look at whether that's whether they are indeed events later. 34 00:03:20,070 --> 00:03:25,710 But the relation of causation is very important for the purposes of explanation. 35 00:03:25,710 --> 00:03:32,730 And human beings go in for explanation a lot. That's what we do, really. 36 00:03:32,730 --> 00:03:39,090 Also, you cannot really predict things without appealing to the causal relation. 37 00:03:39,090 --> 00:03:47,970 So if you know that A is cause B and you see the occurrence of an A or you've got reason to believe whatever it is that an A is coming. 38 00:03:47,970 --> 00:03:58,920 Then you'll have reason to predict a B, because knowing the T and B causally related enables you to predict to be on observation of an A. 39 00:03:58,920 --> 00:04:03,150 So again, it causation is is vital for explanation. 40 00:04:03,150 --> 00:04:10,890 Causation is vital also for prediction. So we can predict things so well, for example, eclipses. 41 00:04:10,890 --> 00:04:15,600 I mean I don't know about you, but I got up at three o'clock the other night and wrapped myself in a duvet and went and sat 42 00:04:15,600 --> 00:04:21,150 in the garden because they told me that there was going to be a lunar eclipse and there was, 43 00:04:21,150 --> 00:04:27,210 you know, amazing, isn't it? But that's because we know about causal relations. 44 00:04:27,210 --> 00:04:34,740 So causation is essential to prediction in the same way that it's essential to explanation. 45 00:04:34,740 --> 00:04:43,410 And then if we want to manipulate anything, if we know the A's caused B, if we bring about any, will also bring about A, B, won't we? 46 00:04:43,410 --> 00:04:48,600 So so if we want to bring about a B, the way we can do it is to bring about an A. 47 00:04:48,600 --> 00:04:55,530 So we can manipulate the world in virtue of the fact that we know about causal relations. 48 00:04:55,530 --> 00:05:00,210 So causation is absolutely vital. It really is. 49 00:05:00,210 --> 00:05:06,210 Will say a bit more about that later. For philosophers, causation is every bit as vital. 50 00:05:06,210 --> 00:05:09,540 I mean, we have causal theories of knowledge. 51 00:05:09,540 --> 00:05:17,670 So if anyone of you know about the Getty problems, the causal theory of knowledge is supposed to explain the Gachet problems. 52 00:05:17,670 --> 00:05:20,360 Now, I'd better explain that for those who don't know, 53 00:05:20,360 --> 00:05:32,340 the Gachet problems says that the traditional theory of knowledge, according to which knowledge, is justified. 54 00:05:32,340 --> 00:05:36,090 True belief. OK. 55 00:05:36,090 --> 00:05:44,400 So let's say Chris has a belief about me that I only tie to Yaris and he's justified 56 00:05:44,400 --> 00:05:49,380 in believing that because he's seen me driving around town in a Toyota Yaris. 57 00:05:49,380 --> 00:05:56,640 OK. What's more, I do own a Toyota Yaris. Nice little silvery green thing. 58 00:05:56,640 --> 00:06:02,550 Now, there's a problem with this because actually the Toyota to Chris Kris's seen me driving around town in isn't mine. 59 00:06:02,550 --> 00:06:07,590 It belongs to Bob. And it's it's a horrible Reds thing. 60 00:06:07,590 --> 00:06:20,400 So the Toyota, that makes true his belief that I own a Yaris and the Toyota that justifies his belief that I own a Yaris come apart, don't they? 61 00:06:20,400 --> 00:06:24,440 Does he know that I own a Yaris? 62 00:06:24,440 --> 00:06:33,980 Most people would say, no, he doesn't, because the conditions justifying his belief come apart from the conditions that make true his belief. 63 00:06:33,980 --> 00:06:35,600 How do we solve this? 64 00:06:35,600 --> 00:06:44,900 Well, we we try and make a causal relationship between the conditions that make his belief true and the conditions that justify his belief. 65 00:06:44,900 --> 00:06:51,050 And so that's the causal theory of knowledge. We appeal to causation to explain what knowledge is. 66 00:06:51,050 --> 00:06:54,800 And we also explain to causation, to appeal what content is. 67 00:06:54,800 --> 00:06:58,130 So how do I know what the content of your belief is? 68 00:06:58,130 --> 00:07:10,730 Could you entertain the concept of read unless a red object has caused you to have an experience of red at some time in your past? 69 00:07:10,730 --> 00:07:14,190 Do you think. No. Okay. 70 00:07:14,190 --> 00:07:19,140 If he if you think not. Then there's the causal theory of content. 71 00:07:19,140 --> 00:07:26,760 There's got to be causal relations involved in the identification of content of your beliefs and perceptions. 72 00:07:26,760 --> 00:07:31,830 And, of course, the causal theory of perception. And Macbeth. 73 00:07:31,830 --> 00:07:39,840 Is this a dagger I see before me? Well, no, it wasn't, because there is no dagger causing the experience he is having. 74 00:07:39,840 --> 00:07:48,510 He was having an experience as of a dagger, but there was no dagger in the causal history of that experience. 75 00:07:48,510 --> 00:07:58,380 And so he wasn't perceiving a dagger. So, again, do you see that causation is absolutely fundamental to the idea of what it is that your cause? 76 00:07:58,380 --> 00:08:09,180 What it is that you're perceiving? So David Hume called causation the cement of the universe. 77 00:08:09,180 --> 00:08:14,610 Causation is the relation that holds together all the events in the universe. 78 00:08:14,610 --> 00:08:21,090 Every event has a cause. Perhaps not the beginning of the universe itself and the causes of God. 79 00:08:21,090 --> 00:08:26,160 But but otherwise, everything has a cause, is an effect. 80 00:08:26,160 --> 00:08:32,130 And it's the cement of the universe. As David Hume and John Carroll. 81 00:08:32,130 --> 00:08:38,610 And you'll find all the references, by the way. You'll find on the handouts that I've got here that I'm not handing out now because 82 00:08:38,610 --> 00:08:43,300 I don't want you to have all the answers to the questions I'm going to ask. But you can take the hand out. 83 00:08:43,300 --> 00:08:48,630 So actually, you don't need to make Coke copious notes unless it helps you understand. 84 00:08:48,630 --> 00:08:51,330 It does me. So I understand if it does, you. Okay. 85 00:08:51,330 --> 00:08:59,520 So John Carroll says, with regard to our total conceptual apparatus, causation is the centre of the centre. 86 00:08:59,520 --> 00:09:04,860 You just would not understand anything if you didn't understand causation. 87 00:09:04,860 --> 00:09:14,640 The philosopher Kant thought causation was one of the concepts that we have innately were born with the concept of cause. 88 00:09:14,640 --> 00:09:23,220 This is very boring. Yes, you will. 89 00:09:23,220 --> 00:09:28,880 Keep the colour scheme and don't show this message again. That sounds right, but where's the cursor gone? 90 00:09:28,880 --> 00:09:38,700 If you turn the volume down, we can't actually see it on the screen. So. 91 00:09:38,700 --> 00:09:44,580 We extended the screen, so it was sort of keep the don't show this message again. 92 00:09:44,580 --> 00:09:49,550 OK, we'll get one more chance. Julia, good. Right. 93 00:09:49,550 --> 00:09:55,880 I didn't realise you couldn't see it, but I was also obscuring what I could see on my screen. 94 00:09:55,880 --> 00:10:02,720 OK. So so causation is the centre of the centre of our conceptual scheme. 95 00:10:02,720 --> 00:10:07,430 So let's go on to how do philosophers think about causation? 96 00:10:07,430 --> 00:10:13,310 Well, and we want to know first what causation is. 97 00:10:13,310 --> 00:10:17,270 I mean, if you're going to ask does something it says. Does God exist? 98 00:10:17,270 --> 00:10:23,390 You need to know what it is about, whose existence you're what your questioning. 99 00:10:23,390 --> 00:10:28,340 So what is God? What is this thing that's you're asking whether it exists or not? 100 00:10:28,340 --> 00:10:37,310 And the same thing is true of causation. And if we want to know whether causation exists and what it is, we need to know first what causation is. 101 00:10:37,310 --> 00:10:44,090 So we start by analysing the concept and to analyse the concept is to. 102 00:10:44,090 --> 00:10:48,440 For example, we reason using the concept of causation. 103 00:10:48,440 --> 00:10:57,380 So we say if A causes B and an A occurred, we can follow on that by saying a B will occur. 104 00:10:57,380 --> 00:11:05,240 That's the prediction I was talking about earlier. So we look at how the we reason with the concept of causation, what it entails. 105 00:11:05,240 --> 00:11:11,540 What we infer from it and so on. That's what it is to analyse a concept. 106 00:11:11,540 --> 00:11:15,350 But then we want to know whether it exists. So we turn to metaphysics. So. 107 00:11:15,350 --> 00:11:18,950 Is there anything that satisfies this role? 108 00:11:18,950 --> 00:11:29,540 The role that something plays in our reasoning. So if God is omnipotent, omniscient and omnipresent, I forgot that. 109 00:11:29,540 --> 00:11:35,210 Right then we want to know whether there is anything that satisfies those three things. 110 00:11:35,210 --> 00:11:42,770 And if we think that the problem of evil shows that that it's a contradiction that anything could suffer to satisfy those things. 111 00:11:42,770 --> 00:11:47,840 We're going to conclude that God doesn't exist. And it's the same with causation. 112 00:11:47,840 --> 00:11:53,180 We will look at its role in reasoning and then we'll ask whether it exists. 113 00:11:53,180 --> 00:11:57,890 So we are interested in the nature of causation of causation itself. 114 00:11:57,890 --> 00:12:06,440 But we start by looking at the concept of causation, the way we work with causation in our reasoning. 115 00:12:06,440 --> 00:12:12,110 So let's have a have a little go at that. Okay, so imagine my tomato plants have died. 116 00:12:12,110 --> 00:12:19,340 They look terrible, don't they? And if the queen had watered my tomato plants, they would not have died. 117 00:12:19,340 --> 00:12:27,110 I mean, this is true. Well, the queen is not watering my tomato plants, therefore caused them to die. 118 00:12:27,110 --> 00:12:30,650 Did it not? And here's the queen failing to water my tomato plants. 119 00:12:30,650 --> 00:12:37,700 I couldn't find one with her watering tomato plants. But what do you think about that? 120 00:12:37,700 --> 00:12:44,760 I mean, it is true, isn't it, that if the queen had watered my tomato plants, they wouldn't have died? 121 00:12:44,760 --> 00:12:52,260 It's also true that if I had watered them, they wouldn't have died. But if the queen had watered them, given that I didn't, they wouldn't have died. 122 00:12:52,260 --> 00:13:01,820 So is the queen the cause of the death of my tomato plants? 123 00:13:01,820 --> 00:13:09,790 Well, yes, but that's true, isn't it? If the queen had watered them, that it wouldn't have died. 124 00:13:09,790 --> 00:13:16,480 Yes, you could you could substitute anyone else in there. Anyway, we won't worry about that now because we will look at that later. 125 00:13:16,480 --> 00:13:26,950 But that's the sort of thought experiment that we might go into to say, well, OK, if that's true and that's true, then isn't that true? 126 00:13:26,950 --> 00:13:31,690 And if it isn't true, none of us think it's true. Why isn't it true? 127 00:13:31,690 --> 00:13:36,350 What's the difference between the Queen's failing to water my tomato plants and my failing to 128 00:13:36,350 --> 00:13:43,520 most of them or perhaps the the neighbourhood promise to water them or something like that? 129 00:13:43,520 --> 00:13:51,780 Whoops. We're going around. So in analysing our concept of causation, we hope to do one of three things. 130 00:13:51,780 --> 00:13:57,150 So we either hope to reduce causation to non causal relations and matters of fact. 131 00:13:57,150 --> 00:14:06,300 So we hope to understand causation by cashing it out in terms that we do understand that don't appeal to causation. 132 00:14:06,300 --> 00:14:12,540 Otherwise we'd have something circular or we hope to eliminate causation. 133 00:14:12,540 --> 00:14:19,980 Maybe causation doesn't really exist. But there's some other relation that that does the job that we think is causation. 134 00:14:19,980 --> 00:14:25,140 Or we might want to admit that causation is primitive. 135 00:14:25,140 --> 00:14:33,180 In other words, that it can't be reduced and it can't be eliminated. Causation is part of the furniture of the universe. 136 00:14:33,180 --> 00:14:38,160 It's it. It is sui generous. It exists in and of itself. 137 00:14:38,160 --> 00:14:43,050 And for example, can you think of anything else that might be of that kind? 138 00:14:43,050 --> 00:14:46,750 Truth is, maybe suy generous. 139 00:14:46,750 --> 00:14:58,790 But so we do one of these things will either reduce it or will eliminate it or will admit it as primitive as a result of our conceptual analysis. 140 00:14:58,790 --> 00:15:07,160 Okay, now let's have a look at how science thinks about causation. Scientists use the concept of causation all the time. 141 00:15:07,160 --> 00:15:13,280 They're relying on the concept of causation, but they very rarely think about causation. 142 00:15:13,280 --> 00:15:21,200 That's not their job. The science says that think about causation tend to be the sciences that think about thinking. 143 00:15:21,200 --> 00:15:31,340 So psychology, for example, is very interested in the concept of causation because I, for example, do babies have a concept of causation. 144 00:15:31,340 --> 00:15:36,530 So you'll have seen some of these videos of babies that, you know, 145 00:15:36,530 --> 00:15:44,480 you do something and try and work out whether the baby realises that one thing causes another. 146 00:15:44,480 --> 00:15:50,390 So psychologists are interested in causation. Neuroscientists are interesting causation. 147 00:15:50,390 --> 00:16:02,270 They like to know what it is in the brain that realises thoughts about causation and applied robotics is particularly interested in causation. 148 00:16:02,270 --> 00:16:07,850 So there's a paper by Judith Perl on your reading list, which is very large. 149 00:16:07,850 --> 00:16:13,380 I mean, if you want for instance, if you want a domestic robot or a caring robot, 150 00:16:13,380 --> 00:16:18,710 then you want it to understand if it leaves the child's skateboard in the middle of the floor, 151 00:16:18,710 --> 00:16:22,910 the person its caring for is likely to trip over it or something. 152 00:16:22,910 --> 00:16:28,040 So you'll you want to give the robot some idea of causation. 153 00:16:28,040 --> 00:16:35,930 And actually that's that's you'll see if you watch the video by due to Air Pearl, that's much easier said than done. 154 00:16:35,930 --> 00:16:41,360 So those in sciences that are really concentrating on causation. 155 00:16:41,360 --> 00:16:50,600 But of course it's it does appear in every science. How can it not if it's so central to explanation and prediction? 156 00:16:50,600 --> 00:16:59,410 OK. His very handsome chap, Hume, Scottish philosopher 1711 to 1776, 157 00:16:59,410 --> 00:17:08,470 and he's the author of The Regularity Theriot's Hume's Regularity Theory that really started off thinking about causation. 158 00:17:08,470 --> 00:17:09,430 It's not entirely true. 159 00:17:09,430 --> 00:17:17,320 I mean, Aristotle thought about causation, too, and I'd give new references on your handout as to who thought about causation before Hume. 160 00:17:17,320 --> 00:17:26,380 But Hume really started the modern discussion of causation, and it was in this book that he claimed fell still from the press. 161 00:17:26,380 --> 00:17:28,660 In other words, nobody was interested in this book. 162 00:17:28,660 --> 00:17:37,120 So he wrote this book, which is basically this book with a few things added and a few things taken away. 163 00:17:37,120 --> 00:17:44,740 And it's in there that he discusses causation. And again, you'll get all the references and you can also find these texts. 164 00:17:44,740 --> 00:17:52,670 All his texts are available freely online. You can look at what he actually said himself. 165 00:17:52,670 --> 00:18:00,980 Okay. Right. We're going to look first at the traditional interpretation of Hume and you'll see 166 00:18:00,980 --> 00:18:06,200 later that it becomes important that there are different interpretations of Hume. 167 00:18:06,200 --> 00:18:11,510 Then we're going to look at two key problems for the traditional interpretation of Hume, 168 00:18:11,510 --> 00:18:18,500 and then we're going to look at the solutions to these problems, leading to the version of Hume's theory that's used today. 169 00:18:18,500 --> 00:18:25,640 So we'll end, as I said before, with the canonical statement of the regularity theory of causation. 170 00:18:25,640 --> 00:18:35,360 But before I go on to its traditional interpretation of Hume, any questions about what I've said so far? 171 00:18:35,360 --> 00:18:44,110 The particular example you said about the care home of leaving escapable in the middle of the room as a potential cause of potential effect. 172 00:18:44,110 --> 00:18:49,350 You have not mentioned the probability. At all that is that it was. 173 00:18:49,350 --> 00:18:54,040 There is a possibility. Well, 174 00:18:54,040 --> 00:18:59,570 I shall see that we could probability does become important later and causation 175 00:18:59,570 --> 00:19:06,170 could be either always deterministic or it could be sometimes deterministic, 176 00:19:06,170 --> 00:19:13,070 sometimes probabilistic or always probabilistic. And there are those three possibilities. 177 00:19:13,070 --> 00:19:18,320 And there are philosophers who who adopt each of those views. 178 00:19:18,320 --> 00:19:22,730 And you're going to see this is actually less important than you might think, 179 00:19:22,730 --> 00:19:30,890 because although the traditional interpretation of Hume treat cities deterministic, because in those days they did. 180 00:19:30,890 --> 00:19:37,910 You'll see that there is a probabilistic version. And ditto for every other theory of causation that we're going to be looking at. 181 00:19:37,910 --> 00:19:48,080 So A causes B might be A necessitates B if determinism is true or A makes B more probable. 182 00:19:48,080 --> 00:19:52,700 If causation is probabilistic. OK. Any other. 183 00:19:52,700 --> 00:20:03,100 David, you said A causes B, you said that if you find to be somewhat. 184 00:20:03,100 --> 00:20:07,480 If he calls his BS and if if an A occurs, then a B will occur. 185 00:20:07,480 --> 00:20:18,030 Yes, that's see with some probability. Yes, but it could be different because that said, it's not necessary. 186 00:20:18,030 --> 00:20:23,550 No, because if A causes B doesn't necessarily mean the AIDS the only cause of B. 187 00:20:23,550 --> 00:20:29,930 So there might be a B without any. If A causes B, but there won't be any without a B. 188 00:20:29,930 --> 00:20:35,030 Did I say that the right way round, I could suddenly hear myself getting confused? 189 00:20:35,030 --> 00:20:47,760 But OK, distinction. Maybe more in psychology between causation and. 190 00:20:47,760 --> 00:20:54,320 Ability of a human mind to connect? No, no, no causality causation. 191 00:20:54,320 --> 00:21:00,650 They use pretty much interchangeably. I'm sure I could think of a distinction if you really want me to, 192 00:21:00,650 --> 00:21:10,850 but I don't think I want to in your possible versions of what you might mean by understanding causation. 193 00:21:10,850 --> 00:21:16,040 One thing which seems to be missing is the possibility that there is no such unified 194 00:21:16,040 --> 00:21:22,340 concept discrimination against a bunch of different things we be lumped together. 195 00:21:22,340 --> 00:21:33,070 Yes. OK. So. So I said that in analysing the concept of causation, we hope hope either to reduce it or to eliminate it or to an. 196 00:21:33,070 --> 00:21:36,640 Make it primitive. Yes, I suppose there is a fourth. 197 00:21:36,640 --> 00:21:44,120 The fourth one is that what we show is that there's there's there are different relations of causation that is not a univocal relationship, 198 00:21:44,120 --> 00:21:48,940 that there are different types of causation. So. Yes. 199 00:21:48,940 --> 00:21:54,490 Yes. If that's a forth, then there is a fourth. Okay. 200 00:21:54,490 --> 00:21:59,440 Let's move on. Let's go on to the traditional interpretation of human. 201 00:21:59,440 --> 00:22:08,710 OK. There is. I don't usually believe in putting lots of words up on a screen, 202 00:22:08,710 --> 00:22:26,160 but but having not have a look at that, because that's the quotation that starts it all. 203 00:22:26,160 --> 00:22:33,570 Okay, so that's what Humes says about causation, but how should we understand that? 204 00:22:33,570 --> 00:22:38,790 Well, let's look at Hume's negative argument to start off with. 205 00:22:38,790 --> 00:22:44,280 We'll look at what he thinks causation isn't and then we'll have a look at what he thinks. 206 00:22:44,280 --> 00:22:55,200 Causation is so huge. Negative argument. The first premises are idea of causation seems to be the idea of necessary connexion. 207 00:22:55,200 --> 00:23:01,410 So if we think that A causes B, while you've already mentioned probably ism, but but in fats, 208 00:23:01,410 --> 00:23:12,330 that we tend to think that if A causes B, then A is sufficient for B, i.e. if any happens, then A B will happen. 209 00:23:12,330 --> 00:23:18,120 So that was the start of Hume's argument. And we'll have a look later. 210 00:23:18,120 --> 00:23:25,350 I've got a little sort of movie that will show you more about what that means. 211 00:23:25,350 --> 00:23:36,180 Premise two is and this is an important premise as empiricists, we should accept that all our ideas come from impressions. 212 00:23:36,180 --> 00:23:42,450 And again, I'll explain this more later. But the an idea is a concept. 213 00:23:42,450 --> 00:23:51,750 A concept is a constituent of thought. So if you're thinking Marianne's wearing blue, you're entertaining your concept of blue. 214 00:23:51,750 --> 00:23:55,510 Now, I want you to imagine that this is yellow. Okay. 215 00:23:55,510 --> 00:24:01,040 You doing that or you're entertaining your concept of yellow at that point? 216 00:24:01,040 --> 00:24:06,010 And you must be entertaining your concept because it's not yellow, is it? It's so. 217 00:24:06,010 --> 00:24:10,040 So you're thinking about this. You're not seeing that the shirt is yellow. 218 00:24:10,040 --> 00:24:14,620 Your your imagining that the shirt is yellow. 219 00:24:14,620 --> 00:24:21,010 But on the other hand, you can see that it's blue. So you have a perception of my shirt. 220 00:24:21,010 --> 00:24:27,520 And you also have a thought about my shirt. And one of them entertainment uses an impression. 221 00:24:27,520 --> 00:24:32,620 You see that it's blue and the other one uses an idea. 222 00:24:32,620 --> 00:24:35,320 You imagine that it's yellow. 223 00:24:35,320 --> 00:24:44,170 So, for example, and you're all thinking about elephants right now, but there isn't an elephant anywhere near this room that I know of. 224 00:24:44,170 --> 00:24:52,690 And even if there is, you can't see it. And so your entertain your concept of elephant, but not your percept of elephant. 225 00:24:52,690 --> 00:25:00,140 For that, you'd have to go to a zoo. Okay, so I'll say something more about that later on. 226 00:25:00,140 --> 00:25:06,740 And premies three, says Hume, we do not and cannot. 227 00:25:06,740 --> 00:25:18,350 He says have any experience of necessary connexion and so experience and impressions go together because when we're having an experience 228 00:25:18,350 --> 00:25:30,710 where we're having impressions of something and so his conclusion is our idea of causation is not an idea of necessary connexion. 229 00:25:30,710 --> 00:25:36,590 And I think you'll agree that that argument is is a good argument. 230 00:25:36,590 --> 00:25:45,560 The conclusion follows from the premises. And so if the premises are true, then the conclusion will also be true. 231 00:25:45,560 --> 00:25:50,030 Is everyone happy with that? Does anyone want to as I sound and say more about this later. 232 00:25:50,030 --> 00:25:57,820 But does anyone want to ask anything about that? As it stands. 233 00:25:57,820 --> 00:26:03,770 As usual, with a standard scientific view of evolution. 234 00:26:03,770 --> 00:26:10,470 Well, can we leave that to. Because that's a substantive question rather than a question of clarification. 235 00:26:10,470 --> 00:26:16,410 Okay. So. That's humour, negative argument. 236 00:26:16,410 --> 00:26:24,170 Now let's have a look at premiss, too, so premise two got what premise two was its premise to us empiricists. 237 00:26:24,170 --> 00:26:31,820 We should accept that all our ideas come from impressions. OK, well, Hume's theory of ideas. 238 00:26:31,820 --> 00:26:36,100 Let's have a look at that, because that's absolutely crucial to his theory of causation. 239 00:26:36,100 --> 00:26:40,160 Hume is an empiricist. I mentioned Kant earlier. 240 00:26:40,160 --> 00:26:46,310 Kant isn't an empiricist. Kant is a nativist. Kant thinks that we have innate ideas. 241 00:26:46,310 --> 00:26:52,760 We're born with our minds already stocked with certain concepts. 242 00:26:52,760 --> 00:27:04,340 Hume and Locke and Berkeley, the three British empiricists all believed that we are born tabula rasa as Hume, as Locke put it. 243 00:27:04,340 --> 00:27:10,550 So we're born with our minds as blank slates. We have nothing in our minds when we're born. 244 00:27:10,550 --> 00:27:17,360 Instead, we acquire all our thoughts and all our ideas from experience. 245 00:27:17,360 --> 00:27:27,960 And people have a great tendency these days to be empiricists because they tend to think that that's the scientifically respectable thing to be. 246 00:27:27,960 --> 00:27:33,410 But we're going to leave that open. But Hume was definitely empiricist. So that's an impression. 247 00:27:33,410 --> 00:27:38,460 Let's pretend. Okay. So. So you're experiencing a cat. 248 00:27:38,460 --> 00:27:45,290 A very handsome ginger cat. And as a result of experiencing things like that. 249 00:27:45,290 --> 00:27:48,620 And black ones and ginger ones. And three legged ones. 250 00:27:48,620 --> 00:27:52,250 And tailless ones. And female ones. Male ones and so on. 251 00:27:52,250 --> 00:28:01,880 You form an idea of a cat. Okay. So you have a percept and this enables you to form a concept. 252 00:28:01,880 --> 00:28:08,580 And this. Sorry, this is me trying to be clever here. 253 00:28:08,580 --> 00:28:13,590 If it's going to work. No, it's not going to work. Dan? 254 00:28:13,590 --> 00:28:23,140 If I could play that, it would me, Al, because I want you to to understand that ideas are not always visual. 255 00:28:23,140 --> 00:28:27,120 A lot of your climbing, your idea of a corporator isn't visual or auditory. 256 00:28:27,120 --> 00:28:32,490 Is it your idea of feminism or austerity? 257 00:28:32,490 --> 00:28:37,560 None of those things are ideas that come directly from impressions. 258 00:28:37,560 --> 00:28:41,670 But the if you are going to have an idea of a cat, 259 00:28:41,670 --> 00:28:49,800 it's got to be an idea of something that could be Ginger or Blackhall to Dodge Dart, but also makes the certain sound and. 260 00:28:49,800 --> 00:28:54,390 Unfortunately, I can't play you the sound it would make. Okay. 261 00:28:54,390 --> 00:28:58,440 So that's that's his theory of ideas. 262 00:28:58,440 --> 00:29:05,040 Any questions about that? Before we move on, because that's a. 263 00:29:05,040 --> 00:29:14,940 Crucial. Element. No. And does Hugh specify any sort of medium? 264 00:29:14,940 --> 00:29:24,080 Through which the idea manifests itself can be. You know, it needn't be a picture at all. 265 00:29:24,080 --> 00:29:28,610 I mean, Locke thought it was a picture. But Hume did. 266 00:29:28,610 --> 00:29:31,260 I don't think Hume specified. I might be wrong about that. 267 00:29:31,260 --> 00:29:36,830 But but it's certainly the case that these days we would say it's not necessary for it to be a pitcher. 268 00:29:36,830 --> 00:29:44,750 I mean, you might or might not have a picture of what feminism means or what austerity means. 269 00:29:44,750 --> 00:29:52,160 And if you have a picture. But yours is different. And yet the two of you mean the same thing by austerity, let's say. 270 00:29:52,160 --> 00:29:58,450 But the pictures you associate with it, if you do associate any with it, might be different. 271 00:29:58,450 --> 00:30:04,480 Can you have something totally abstract, say something that's informed by equations or social? 272 00:30:04,480 --> 00:30:11,580 Yes, or love. That's pretty abstract justice. 273 00:30:11,580 --> 00:30:18,450 So Hume would have to say that you acquire the idea of justice by seeing instances of justice. 274 00:30:18,450 --> 00:30:26,460 So when a mother is fair to her children, for example, you see her being just when a teacher is fair to the pupils, 275 00:30:26,460 --> 00:30:41,070 you see the teacher being just and through experiences of justice like that, you form your idea of justice might not experience a string. 276 00:30:41,070 --> 00:30:45,360 Well, string theory is not not an idea, it's a multitude of ideas, isn't it? 277 00:30:45,360 --> 00:30:58,560 I mean, it's a theory, string theory. A theory is, if you like, a web of beliefs that go together in a coherent whole. 278 00:30:58,560 --> 00:31:03,810 OK. It may be a cold call. Some people would question that, but. 279 00:31:03,810 --> 00:31:11,310 So. So a theory is a multitude of beliefs. And each belief has a content that's made up of concepts and an idea is a concept. 280 00:31:11,310 --> 00:31:20,880 So an idea is a constituent of a thought. So every thought has content and every content will be made up of concepts. 281 00:31:20,880 --> 00:31:25,920 And string theory is made up of beliefs, multitude of beliefs. 282 00:31:25,920 --> 00:31:32,380 Each belief has a content in each content is made up of concepts. 283 00:31:32,380 --> 00:31:37,450 OK, so now we're going to look at his empiricism, the idea of causation. 284 00:31:37,450 --> 00:31:45,300 OK. Now here's a billiard ball, OK? And this billiard ball. 285 00:31:45,300 --> 00:31:49,680 Is going to go off when the first one hits it. 286 00:31:49,680 --> 00:31:56,630 But what what makes us think that it's going to do that? Because it could be. 287 00:31:56,630 --> 00:32:01,690 It does that, couldn't it? I mean, why would you expect? 288 00:32:01,690 --> 00:32:08,510 Because I bet you do expect the first thing to happen. Rawdon's second thing to happen. 289 00:32:08,510 --> 00:32:13,640 Is there a necessary connexion? Could it do the second thing or not? 290 00:32:13,640 --> 00:32:23,720 It could. OK. It said there's absolutely nothing necessary about the fact that the second ball will roll off when the first ball hits it, 291 00:32:23,720 --> 00:32:29,360 as opposed to turning purple and doing a little spirit pirouette. 292 00:32:29,360 --> 00:32:33,020 And that's what Hume notices. And he's actually very. 293 00:32:33,020 --> 00:32:39,440 Because if we think causation is necessary connexion, then we're saying that when the billiard ball hits, 294 00:32:39,440 --> 00:32:45,140 when the first billiard ball hits the second billiard ball, it's necessary that the second billiard ball rolls off. 295 00:32:45,140 --> 00:32:49,910 But Hume was the first one to notice that there is nothing necessary about it. 296 00:32:49,910 --> 00:32:54,980 It's certainly not logically necessary. We can imagine. 297 00:32:54,980 --> 00:33:14,060 I mean, there are logically necessary things. For example. A bachelor is an unmarried man. 298 00:33:14,060 --> 00:33:25,140 John is a bachelor. Therefore. John is an unmarried man. 299 00:33:25,140 --> 00:33:34,740 Yep. I mean, there's a logical necessity. These two sentences entail the conclusion an entailment is the relation of logical necessity. 300 00:33:34,740 --> 00:33:42,850 Well, there's no logical necessity that when this billiard ball comes in and hits the AASA that the other will roll off in the opposite direction. 301 00:33:42,850 --> 00:33:50,580 There's nothing logically necessary about that. You can imagine it doing all sorts of things like spinning, turning purple and spinning off. 302 00:33:50,580 --> 00:33:54,990 And so if we're going to say that there is a necessity of any kind. 303 00:33:54,990 --> 00:34:00,960 We've got to introduce a new notion of necessity. We've got to talk about metaphysical necessity. 304 00:34:00,960 --> 00:34:04,400 All we've got to talk about empirical necessity or something like that. 305 00:34:04,400 --> 00:34:10,440 We've got to introduce a new concept, if you like. 306 00:34:10,440 --> 00:34:17,880 But Hume says. But how do we experience something necessary? 307 00:34:17,880 --> 00:34:23,580 Think about that for a second. If something's necessary, it must happen, OK? 308 00:34:23,580 --> 00:34:27,390 Every time one billiard ball hits another, the billable will roll off. 309 00:34:27,390 --> 00:34:32,070 It must happen. So not just does it happen that way. It must happen that way. 310 00:34:32,070 --> 00:34:43,610 Coaching experience that that must. Do you think? 311 00:34:43,610 --> 00:34:50,310 But must not happen. What if you've been given anaesthetic? 312 00:34:50,310 --> 00:34:56,640 Well, obviously, there's some limiting conditions. But for most purposes, it will. 313 00:34:56,640 --> 00:35:01,920 For most purposes, it will hurt. Yes, fine. But we're talking about it must hurt. 314 00:35:01,920 --> 00:35:07,360 There's nothing necessary about it, is there? That. That's the thing that we're thinking. 315 00:35:07,360 --> 00:35:20,680 This is. For like something great. 316 00:35:20,680 --> 00:35:27,010 Is everything right? 317 00:35:27,010 --> 00:35:33,970 No. Okay. Same the same way if I strike a match, it'll only light if there's oxygen around. 318 00:35:33,970 --> 00:35:39,320 So it's got to be all sorts of. OK. We're going to be looking at this in some detail later on. 319 00:35:39,320 --> 00:35:42,680 But but at the moment, what I want to get hot is the idea, 320 00:35:42,680 --> 00:35:53,020 the fact that we cannot experience something necessarily something that must be the case or that can't be the case. 321 00:35:53,020 --> 00:36:02,020 We can experience something isn't the case or that it's not often the case or this is nearly always not the case. 322 00:36:02,020 --> 00:36:05,440 We can't experience that. It cannot be the case. 323 00:36:05,440 --> 00:36:13,270 So if if I use possible world talks to talk, just remember to say that something must be the case, to say that two plus two equals four. 324 00:36:13,270 --> 00:36:22,840 And that that's a necessary condition. I'm saying that there is no possible world in which two plus two don't equal four. 325 00:36:22,840 --> 00:36:28,990 Okay. I can do it in this case because that's a logical necessity. You've got the same condition. 326 00:36:28,990 --> 00:36:36,100 So the American system. Because otherwise you do. 327 00:36:36,100 --> 00:36:43,090 No, no, no, I don't have to specify that because two has a normal meaning in English and so does plus and so does the equals. 328 00:36:43,090 --> 00:36:51,310 And so just for. And so it's only because your requiring you're being clever and thinking if I if I interpret it differently. 329 00:36:51,310 --> 00:36:55,750 It wouldn't be. It wouldn't. There is a possible world in which two plus two equals four. 330 00:36:55,750 --> 00:37:02,090 But I'm assuming that you're not interpreting it differently. 331 00:37:02,090 --> 00:37:11,330 That's it. Now is arguably your second or third dynamics. Oh, Evie. 332 00:37:11,330 --> 00:37:19,730 Thank you. You bet you Unda undermined your what you said by adding in a physical well in a felt world like this. 333 00:37:19,730 --> 00:37:29,920 ET cetera. I mean, if there are worlds in which the second law of thermodynamics doesn't hold and might there be. 334 00:37:29,920 --> 00:37:35,320 I have to say yes than than I my rest, my case again. 335 00:37:35,320 --> 00:37:44,860 Necessity is a very strong relationship. If something is necessary, there is no possible world in which it doesn't happen. 336 00:37:44,860 --> 00:37:52,360 And if something is is possible, there is one possible world in which there is at least one possible world in which it does happen. 337 00:37:52,360 --> 00:37:58,210 So could we experience necessity? Answer no. 338 00:37:58,210 --> 00:38:04,990 Because we would have to experience every single possible world and we couldn't do that. 339 00:38:04,990 --> 00:38:11,980 So Cuba has a real problem with the idea of necessity. You cannot experience necessity. 340 00:38:11,980 --> 00:38:17,680 So as as logicians, we can believe that something is necessary. 341 00:38:17,680 --> 00:38:22,390 But the idea that you could experience it is a very different claim. 342 00:38:22,390 --> 00:38:29,470 So where do we get this idea of necessary connexion or where do we think we get it from? 343 00:38:29,470 --> 00:38:38,620 Okay, let's. Oh, it's not working as well as I thought. 344 00:38:38,620 --> 00:38:43,790 In fact, it's not working at all. Sold. It worked on my home computer. 345 00:38:43,790 --> 00:38:51,760 OK. What? What do you thinks that we get our idea of necessary connexion from is constant conjunction. 346 00:38:51,760 --> 00:38:55,810 So we see one billiard ball hitting another. And the other rolling off. 347 00:38:55,810 --> 00:39:00,340 And we see one billiard ball hitting another. And the other rolling off. And we see one billiard ball hitting another. 348 00:39:00,340 --> 00:39:05,290 And the other rolling off. We see this often enough that we start to expect it. 349 00:39:05,290 --> 00:39:10,960 And it's that expectation that gives us the idea of necessary connexion. 350 00:39:10,960 --> 00:39:18,640 So, OK, if our idea is idea of causation is not an idea of necessary connexion, what is it? 351 00:39:18,640 --> 00:39:26,800 An idea of. And secondly, why are we so certain this idea of causation is an idea of necessary connexion? 352 00:39:26,800 --> 00:39:31,150 Okay, I've sort of answered that. But, sir, let's let's go on. 353 00:39:31,150 --> 00:39:35,650 Okay. So can you see that his negative argument. This is what causation isn't. 354 00:39:35,650 --> 00:39:39,970 We think that causation is the idea of necessary connexion. 355 00:39:39,970 --> 00:39:49,120 If it is the idea of necessary connexion, this must have come from experience because all our ideas come from experience. 356 00:39:49,120 --> 00:39:55,600 But we can't experience necessity. Therefore, it can't be the idea of necessary connexion. 357 00:39:55,600 --> 00:40:00,760 So that's negative argument. I see. No, but I'd like to continue just for a minute. 358 00:40:00,760 --> 00:40:03,490 The negative argument leaves us with two questions. 359 00:40:03,490 --> 00:40:11,230 If Hume is right, if Hume's negative argument is right that the idea of causation is not an idea of necessary connexion, 360 00:40:11,230 --> 00:40:17,800 then we're left thinking, well, what is it then? What is this idea of causation that's so important? 361 00:40:17,800 --> 00:40:27,340 And would also like to ask, well, given that we think it is necessary connexion, why are we shows sure of that when you've just shown us. 362 00:40:27,340 --> 00:40:32,980 Hume. If you have shown us that it can't be. So those are the two questions. 363 00:40:32,980 --> 00:40:38,200 So his answer to question one is that our idea of causation is and wait for it. 364 00:40:38,200 --> 00:40:43,510 Because this is the regularity theory, temporal priority. 365 00:40:43,510 --> 00:40:48,270 So the cause comes before the effect. Okay. 366 00:40:48,270 --> 00:40:54,090 It's spatial contiguity, the scores and the effect are spatially contiguous. 367 00:40:54,090 --> 00:41:03,150 And if they appear not to be, there's something in between that makes them there's no action at a distance as cause that says Hume. 368 00:41:03,150 --> 00:41:15,090 And finally, its constant conjunction. When we see A and B constantly conjoined, we start to think that A causes B. 369 00:41:15,090 --> 00:41:22,680 So the correlation between A and B causes us to expect B when we see DNA. 370 00:41:22,680 --> 00:41:31,200 And that expectation is projected by us onto the world and becomes the idea of causation. 371 00:41:31,200 --> 00:41:38,760 So Hume thinks there is no more to our idea of causation than these three things. 372 00:41:38,760 --> 00:41:45,030 Now, that should be making you feel quite uncomfortable. But I'm not going to take any questions just at the moment. 373 00:41:45,030 --> 00:41:51,280 I'd like to just go on a little bit more. Okay. This is the regularity theory of causation. 374 00:41:51,280 --> 00:41:57,800 There's no more to causation than these three things. 375 00:41:57,800 --> 00:42:02,140 And Hume's answer to question two is it is just habit. 376 00:42:02,140 --> 00:42:16,030 It's because we see and be correlated and we form the expectation of seeing a B whenever we see any that we think that A and B causally related. 377 00:42:16,030 --> 00:42:24,700 And it's that expectation that we spread on the world, that we project onto the world uncool, necessary connexion. 378 00:42:24,700 --> 00:42:30,730 But in fact, there is no necessary connexion. All there is is a habit of mind. 379 00:42:30,730 --> 00:42:41,340 So in a way, Hume is saying, if you're saying that causation is necessary connexion, then Hume is saying there isn't any causation. 380 00:42:41,340 --> 00:42:48,330 But actually, what he's saying is not that causation is necessary connexion and there isn't any. 381 00:42:48,330 --> 00:42:59,550 He's saying causation isn't necessary connexion at all. Causation is just temporal priority, spatial contiguity and constant conjunction. 382 00:42:59,550 --> 00:43:11,060 Hence the regularity theory and regularities. Okay, let me see where I am because. 383 00:43:11,060 --> 00:43:21,580 Okay, the key characteristics of the regularity theory of causation are, firstly, it's reductive. 384 00:43:21,580 --> 00:43:27,020 Okay. So do you remember I said that what we hope to do as a result of. 385 00:43:27,020 --> 00:43:34,610 Conceptual analysis is either to reduce or to eliminate or to make primitive RTC is reductive. 386 00:43:34,610 --> 00:43:43,910 He's reducing the idea of causation to temporal priorities, spatial contiguity and constant conjunction. 387 00:43:43,910 --> 00:43:51,950 So it reduces causation to regularity and it prioritises causal regularities over singular causal relations. 388 00:43:51,950 --> 00:44:02,570 If you only ever see one billiard ball hitting another and the other rolling off once, you'll never think that the first caused the second, says Hume. 389 00:44:02,570 --> 00:44:05,900 Actually, you would, because you've seen so many other similar things. 390 00:44:05,900 --> 00:44:12,920 But if you only see one case, there can't be causation, according to Hume, in the individual case. 391 00:44:12,920 --> 00:44:18,590 Now, many philosophers have quarrels about that, but that's what Hume thinks. 392 00:44:18,590 --> 00:44:25,070 He reduces causation to regularity. And he also is a realist. 393 00:44:25,070 --> 00:44:29,900 So the regularity theory of causation is a realist theory about causation. 394 00:44:29,900 --> 00:44:36,470 Causation really exists. And that's because there really are regularities. 395 00:44:36,470 --> 00:44:42,230 There really are correlations between events. And that's what causation is. 396 00:44:42,230 --> 00:44:47,810 So there is something that satisfies the conditions that it would have to satisfy. 397 00:44:47,810 --> 00:44:53,240 If causation were to exist. So regularities are real. 398 00:44:53,240 --> 00:44:59,030 And their mind independent. Even if we weren't here to see these correlations, they would be there. 399 00:44:59,030 --> 00:45:07,460 So causation is realist, realistic. And finally, regularity theory of causation is austere. 400 00:45:07,460 --> 00:45:16,490 And what I mean by that is Hume rarely is saying there's no explanation. 401 00:45:16,490 --> 00:45:25,250 There is no relation that is this constant conjunction of sorry that underpins the constant conjunction or the regularities. 402 00:45:25,250 --> 00:45:32,810 All there is. Is the regularities. So if a scientist is trying to find out whether A causes B, 403 00:45:32,810 --> 00:45:38,960 he sees the constant conjunction between A and B, and he sees yes, A causes B, what explains that? 404 00:45:38,960 --> 00:45:47,840 He'll look. And if he's lucky, he'll find a deeper correlation. So so C causes A and D causes B. 405 00:45:47,840 --> 00:45:52,520 And that's why a bear can conjoined. Because see India conjoined. 406 00:45:52,520 --> 00:45:57,290 And then if he looks a bit harder, he might look a bit. And he'll find another correlation. 407 00:45:57,290 --> 00:46:04,280 And another correlation. And Hume will think that's all he'll ever find all the way down. 408 00:46:04,280 --> 00:46:12,500 So it's correlation all the way. There is no such thing as an metaphysically necessary connexion. 409 00:46:12,500 --> 00:46:17,840 That explains these constant conjunctions. And if that's making you very uncomfortable, don't worry. 410 00:46:17,840 --> 00:46:25,160 You're not alone. So it eliminates the idea of necessary connexion, is it? 411 00:46:25,160 --> 00:46:31,030 It says you want this idea of there being a metaphysical necessity there. 412 00:46:31,030 --> 00:46:37,400 Okay. You recognise it's not a logical necessity, but you want to introduce this idea of a metaphysical necessity. 413 00:46:37,400 --> 00:46:45,320 But you shouldn't because we can explain everything to do with causation without appealing to necessary connexion. 414 00:46:45,320 --> 00:46:49,880 Therefore, necessary connexion is redundant. We don't need it. 415 00:46:49,880 --> 00:46:56,810 And we should be austere in our theories of causation. 416 00:46:56,810 --> 00:47:05,640 So. OK. Well, before I look at the problems that everyone else has thought, what are the problems that you think? 417 00:47:05,640 --> 00:47:17,670 What? Is there anything you don't like about this theory? Mike, it's not so much from the studio, but I can't see that point of. 418 00:47:17,670 --> 00:47:22,910 Temporal aspect of we're seeing and others will not criticise the temporal aspect of the sea, 419 00:47:22,910 --> 00:47:28,090 and it does not seem to me at all his take example of colliding with billiard balls. 420 00:47:28,090 --> 00:47:39,100 You might as well say that for a stop, because Bobby's direction and direction, which is not okay. 421 00:47:39,100 --> 00:47:41,250 Well, it's interesting that you went for that one first. 422 00:47:41,250 --> 00:47:52,660 I mean, it's certainly true that the humans theory makes temporal priority of the cause over the effect, the analytical condition of causation. 423 00:47:52,660 --> 00:47:59,290 And we might say, well, is that true? I mean, could it not be that the effect comes before the cause? 424 00:47:59,290 --> 00:48:07,060 Couldn't there be backwards causation? And lots of philosophers have looked at the possibility of backwards causation. 425 00:48:07,060 --> 00:48:17,410 But Hume doesn't. I mean, that's one thing he did. That's actually probably the least examined from by humans of his theory. 426 00:48:17,410 --> 00:48:22,780 He just assumes that causes come before effects. And actually so to most of us. 427 00:48:22,780 --> 00:48:26,300 Most of the time. 428 00:48:26,300 --> 00:48:39,320 So if we see a correlation, it's that the prior event that we think of as the cause and the second event that we think of as the effect. 429 00:48:39,320 --> 00:48:48,080 Oh, sorry. They're two people talking. I'm sorry, David. I was talking to the gentleman further back just to start with. 430 00:48:48,080 --> 00:48:54,430 Of the concept of necessity, because it doesn't seem that we should have access. 431 00:48:54,430 --> 00:48:56,500 Well, that's a bit unfair. 432 00:48:56,500 --> 00:49:04,670 I mean, the question was, does Hume justify his use of the concept of necessity because he doesn't seem to have access to the concept of necessity. 433 00:49:04,670 --> 00:49:10,510 And what he thinks is that the knee jerk concept of causation? 434 00:49:10,510 --> 00:49:17,050 Is that a necessary connexion? And he claims to have shown that it's not necessary connexion. 435 00:49:17,050 --> 00:49:27,770 So I don't think it's Hume who needs to justify his use of the concept of necessitates everybody else. 436 00:49:27,770 --> 00:49:33,940 I mean, the to a human concept should really have never come into being. 437 00:49:33,940 --> 00:49:37,360 Well, no, because logical necessity is permitted. 438 00:49:37,360 --> 00:49:46,840 I mean, we all understand logical necessity and we logical necessities is explicable in the reasoning that we do. 439 00:49:46,840 --> 00:49:55,870 But what he's questioning is whether this the second sort of necessity, namely metaphysical or empirical necessity. 440 00:49:55,870 --> 00:49:59,410 And he's the one who shows that. Everyone assumes that. 441 00:49:59,410 --> 00:50:05,920 But actually, we don't need to. Also, he says. 442 00:50:05,920 --> 00:50:18,620 When we talk about a cause, most things that we can think of have multiple. 443 00:50:18,620 --> 00:50:23,600 Well, we'll look at that in more depth later. That's the same question as the one that came earlier and when. 444 00:50:23,600 --> 00:50:28,250 So when you strike a match, the strike of the match causes the lighting of the match. 445 00:50:28,250 --> 00:50:32,720 Well, it wouldn't if there weren't any oxygen in the room. So. 446 00:50:32,720 --> 00:50:37,490 So the strike hit. The match can't be a sufficient condition for the lighting of the match. 447 00:50:37,490 --> 00:50:41,990 You've got to take into account other things. So we'll have a look at that in a minute. 448 00:50:41,990 --> 00:50:53,530 Okay. Speak out this connexion. 449 00:50:53,530 --> 00:50:59,680 Call them, sir. All instances. But then he calls himself. 450 00:50:59,680 --> 00:51:04,020 Well, we may be dead. 451 00:51:04,020 --> 00:51:11,550 Well, no, we can see a constant conjunction because a constant conjunction doesn't necessarily mean an exception, this conjunction. 452 00:51:11,550 --> 00:51:16,470 Nor does it necessarily mean a conjunction that goes on forever in a day. 453 00:51:16,470 --> 00:51:28,600 It just means if I see a closing base, I think for a second about the concept of causation. 454 00:51:28,600 --> 00:51:40,530 If you're going to claim A causes B. Firstly, what's your evidence? 455 00:51:40,530 --> 00:51:51,440 Secondly, what would falsify it? 456 00:51:51,440 --> 00:51:53,570 Well, perhaps we'll just stick with those two at the moment. 457 00:51:53,570 --> 00:52:05,560 So if we can't say that, make a claim, not A causes B, what's going to be your evidence for this? 458 00:52:05,560 --> 00:52:22,810 Of what? I mean, you're directly perceiving me, but that doesn't give you evidence that A causes B. 459 00:52:22,810 --> 00:52:28,860 OK. And what is it? Well, you'll give me a shot. And that is a particular case. 460 00:52:28,860 --> 00:52:31,450 Can we go back to the everyday thought of causation? 461 00:52:31,450 --> 00:52:39,420 So when people are doing the psychological experiment with the children, seeing whether the children have the idea of causation, 462 00:52:39,420 --> 00:52:49,900 what do they tend to show children in the hope that they form the belief A causes B? 463 00:52:49,900 --> 00:52:57,040 Light switch in the light. Yes. Okay. So you switch the switch and the light goes on. 464 00:52:57,040 --> 00:53:00,610 You switch, switch the other way and the light goes off. You switch it that way and the light goes on. 465 00:53:00,610 --> 00:53:05,790 You switch that when the light goes off and so on. And the child will very quickly pick up. 466 00:53:05,790 --> 00:53:11,700 What to do? So what you're doing is you're showing a constant conjunction, aren't you? 467 00:53:11,700 --> 00:53:19,140 You're saying that putting the switch that way turns the light on and putting the switch that way turns the light off. 468 00:53:19,140 --> 00:53:29,970 And so the evidence for causation is always going to be a correlation or a constant conjunction. 469 00:53:29,970 --> 00:53:39,720 So Hume is certainly right from that point of view. We don't have evidence for causation other than correlations or constant conjunctions. 470 00:53:39,720 --> 00:53:52,070 And what would falsify the claim that A causes B? But what would you make you think it's not true that A causes B? 471 00:53:52,070 --> 00:53:56,810 So you see an A and not a B, so you break the constant conjunction. 472 00:53:56,810 --> 00:54:05,780 You break the correlation. If you think that A causes B and you see an A without a B, you know you've got something wrong. 473 00:54:05,780 --> 00:54:12,320 So, OK, it's not true that A causes B deterministically, you might think. 474 00:54:12,320 --> 00:54:24,380 OK. So you C, C, A and not be so you know that it's not the case. 475 00:54:24,380 --> 00:54:28,520 A causes B, there's something wrong with that. 476 00:54:28,520 --> 00:54:36,770 Maybe it's only a certain type of A's that cause B's. Maybe it's only a store's calls BS. 477 00:54:36,770 --> 00:54:43,130 So you when they say that you saw that isn't followed by a B actually isn't in a stall. 478 00:54:43,130 --> 00:54:54,140 Or maybe this is a case of probabilistic causation. Maybe A does cause B, but not deterministically, only with a certain probability. 479 00:54:54,140 --> 00:55:00,590 So there will be exceptions. The constant conjunction. It'll be only certain. 480 00:55:00,590 --> 00:55:06,110 Constant to a certain extent. So do you see that correlation? 481 00:55:06,110 --> 00:55:13,250 Constant conjunction is is absolutely essential to our idea of causation. 482 00:55:13,250 --> 00:55:17,690 But Hume, of course, is claiming that. That's all there is. 483 00:55:17,690 --> 00:55:23,030 There is no more to causation than constant conjunction. 484 00:55:23,030 --> 00:55:24,950 And that's quite counterintuitive. 485 00:55:24,950 --> 00:55:32,240 I remember when I first understood this theory as an undergraduate, I just thought, how could it possibly be the case? 486 00:55:32,240 --> 00:55:40,610 How could somebody really think that? And you might be thinking the same thing and thinking that you've misunderstood the theory, but you haven't. 487 00:55:40,610 --> 00:56:00,840 The theory really is that there is no more to causation than constant conjunction. 488 00:56:00,840 --> 00:56:13,720 So you're saying that that whatever it is is necessary for tuberculosis, okay, but it must also be sufficient for it to be a. 489 00:56:13,720 --> 00:56:26,980 We usually think of a cause being sufficient rather than being necessary. But it's sufficient in that it necessitates it. 490 00:56:26,980 --> 00:56:32,090 Then let's go in to look at Caltech. Two more quite quickly and then I'll go and look at the problems. 491 00:56:32,090 --> 00:56:45,280 What are you going to say? 492 00:56:45,280 --> 00:56:53,410 Well, it's not just humour, it's also the rest of us, because all of us take constant conjunction as evidence for causation. 493 00:56:53,410 --> 00:56:57,850 But some of us are more inductively bold than others of us. 494 00:56:57,850 --> 00:57:07,840 And those of us who are inductively bold may take just two cases and then extrapolate from there. 495 00:57:07,840 --> 00:57:12,640 Others of us will want to say, well, hang on, I've only seen two. 496 00:57:12,640 --> 00:57:21,160 I won't see a few more. And so on. So it's so it depends how inductively bold you are as to how. 497 00:57:21,160 --> 00:57:24,520 So one more done here. Sorry, I will take questions later. 498 00:57:24,520 --> 00:57:37,820 But one more one relies on this insight. 499 00:57:37,820 --> 00:57:42,040 It doesn't look like an experiment is quite obviously denies. 500 00:57:42,040 --> 00:57:46,030 Yes, absolutely. Yeah. If if we've seen A causing B, 501 00:57:46,030 --> 00:57:53,410 if we've seen A and B in constant conjunction forever and ever and always and everyone else we've ever spoken to has seen it, 502 00:57:53,410 --> 00:58:01,150 then we see an A without a B. We're much more likely to think that we can't believe our eyes because a miracle is like that, isn't it? 503 00:58:01,150 --> 00:58:10,360 A miracle is an exception to the laws of nature. And you don't make something a law of nature unless you've seen it happen very, very often. 504 00:58:10,360 --> 00:58:19,630 So the idea that you can say there's been an exception to it, just because you think you've seen an exception on one occasion is a big problem. 505 00:58:19,630 --> 00:58:24,520 Humour is very good on miracles because he thinks that, epistemologically speaking, 506 00:58:24,520 --> 00:58:32,470 you can never justify the claim that there's been a miracle because a miracle is an exception to a law of nature and the epistemology. 507 00:58:32,470 --> 00:58:43,150 The evidence that we need to make something, a law of nature could never be undermined by one experience of a of a country case. 508 00:58:43,150 --> 00:58:51,220 But we can talk about that more in the Question Time, if you like. Let's have a look at the problems that philosophers have found for this. 509 00:58:51,220 --> 00:58:58,090 Okay. One problem might be the whole thing relies on Hume's empiricism. 510 00:58:58,090 --> 00:59:07,330 Hume thinks that the idea of causation must come from experience and that the idea of necessary connexion cannot come from experience. 511 00:59:07,330 --> 00:59:15,490 So the idea of causation cannot be necessary connexion. But that relies on this empiricist idea. 512 00:59:15,490 --> 00:59:21,400 And if you're can't, you say, well, sod that, let's throw empiricism out. 513 00:59:21,400 --> 00:59:29,020 Let's say that we're born with the idea of causation. And then we can say that the idea of causation is the idea of necessary connexion. 514 00:59:29,020 --> 00:59:38,770 We don't need to be empiricists so we can reject Hume's bedrock theory, which is his theory of ideas, his empiricism. 515 00:59:38,770 --> 00:59:45,640 We don't have to accept that. We can easily accept another theory. 516 00:59:45,640 --> 00:59:54,370 There are other people who think, well, who might ask, is there really no impression from which we might get our idea of necessary connexion? 517 00:59:54,370 --> 00:59:58,510 So you mentioned my shot. Says is a psychologist. 518 00:59:58,510 --> 01:00:06,580 I think, who tried to show that we do see causation in the individual case. 519 01:00:06,580 --> 01:00:09,910 And he. Well, actually, there have been other philosophers. 520 01:00:09,910 --> 01:00:16,840 So when you cut a slice of bread, says Elizabeth Anscombe, when you're slicing the loaf of bread, 521 01:00:16,840 --> 01:00:22,960 can't you feel that you are causing the bread as you push the knife through the bread? 522 01:00:22,960 --> 01:00:28,390 That is not causation, something you're experiencing directly. 523 01:00:28,390 --> 01:00:35,590 That this is a necessary connexion. That the bread could not not be cut, as you slice it. 524 01:00:35,590 --> 01:00:41,470 And Davidson also thinks that we see causation in the individual case. 525 01:00:41,470 --> 01:00:46,930 So human. Do you remember I said the regularity theory? 526 01:00:46,930 --> 01:00:53,230 He thinks that causal regularities are prior to the cause, a causation in the individual case. 527 01:00:53,230 --> 01:00:58,450 Well, we can reject that. We can say that's not true. We do see causation in the individual case. 528 01:00:58,450 --> 01:01:05,740 When we see one with billiard ball hitting another, we actually see a case of causation. 529 01:01:05,740 --> 01:01:12,060 And we call that necessary connexion. So we can reject that. 530 01:01:12,060 --> 01:01:15,820 And we can say that correlations do not have a direction. 531 01:01:15,820 --> 01:01:22,150 I mean, this is what. Well, we're talking about spatial contiguity or temporal priority. 532 01:01:22,150 --> 01:01:28,600 If we see A and B constantly conjoined. There isn't that way. 533 01:01:28,600 --> 01:01:34,750 Then the correlation doesn't have a duration, does it? If A is correlated with B, then B is correlated with A. 534 01:01:34,750 --> 01:01:39,130 What makes us think that A causes B rather than the other way round? 535 01:01:39,130 --> 01:01:45,000 This is the point that Mike was making a minute ago. And. 536 01:01:45,000 --> 01:01:50,430 We might also ask, how can a relation that depends on similarity be objective? 537 01:01:50,430 --> 01:01:57,940 So do you remember that quotation I gave you from a Causes? 538 01:01:57,940 --> 01:02:02,500 Let's see if I can if I've got it written here. 539 01:02:02,500 --> 01:02:14,800 No, I haven't. He talks about a cause is where you see one event and another event and then similar events in the same constant conjunction. 540 01:02:14,800 --> 01:02:22,300 So you've got to notice that the relation that the events that are correlated are similar to each other. 541 01:02:22,300 --> 01:02:27,760 So A causes B, A causes B, A causes B, A and B are in constant correlation. 542 01:02:27,760 --> 01:02:35,560 You've got to have the idea of the similarity between the A's and the similarity between the B and the idea of similarities. 543 01:02:35,560 --> 01:02:40,330 A very human centred concept, isn't it? 544 01:02:40,330 --> 01:02:47,890 So we think of causation as objective. But how can it be objection objective, if that's what we're talking about. 545 01:02:47,890 --> 01:02:57,160 If we rely on similarity and the last two objections are absolutely key, 546 01:02:57,160 --> 01:03:07,000 surely regularity isn't sufficient for causation and surely with there are regularities that are not causal. 547 01:03:07,000 --> 01:03:13,810 So if a pi, if every time a pineapple has dropped from a tree, Marianne's coughed. 548 01:03:13,810 --> 01:03:19,180 Is that a causal regularity or is it just a coincidence? 549 01:03:19,180 --> 01:03:28,240 And if it's just a coincidence and we can imagine I mean, imagine in this room, every male in this room is his second son. 550 01:03:28,240 --> 01:03:35,260 Okay, that that is possible. It could be the case that just by coincidence, every male in this room is his second son. 551 01:03:35,260 --> 01:03:41,800 Would you think it's causal because there's a correlation between being a male in this audience and being a second son? 552 01:03:41,800 --> 01:03:44,830 Answer no. It's surely a coincidence. 553 01:03:44,830 --> 01:03:53,550 So if if there can be irregularities that are not sufficient for causation, then how can causation be irregularity? 554 01:03:53,550 --> 01:04:00,540 It can't be. And it gets worse, surely regularity is not necessary for causation. 555 01:04:00,540 --> 01:04:07,290 The Big Bang caused the universe. Well, there that only happened once, didn't it? 556 01:04:07,290 --> 01:04:11,120 And if it did only happen once. Well, okay. Oh, God. 557 01:04:11,120 --> 01:04:25,620 Multiverse. But if you think that causation can happen in the individual case, then regularity isn't necessary for causation. 558 01:04:25,620 --> 01:04:34,670 And it's also not sufficient. So over the years since Hume developed his regularity theory, there've been hundreds of objections to it. 559 01:04:34,670 --> 01:04:40,470 And and I've just given you a selection here. And these have been discussed ad nauseum. 560 01:04:40,470 --> 01:04:45,900 But we're going to look at that one in that one. 561 01:04:45,900 --> 01:04:49,860 As you see, I spent my summer learning powerful interest. 562 01:04:49,860 --> 01:04:55,620 OK, so let's let's have a look at the claim that regularities are not sufficient for causation. 563 01:04:55,620 --> 01:05:01,260 Okay. If every male in this audience has a second son, that's not enough to make it. 564 01:05:01,260 --> 01:05:12,090 I mean, are they were they cause to be a second son by being in this audience or is there being a second son causing them to be in this audience? 565 01:05:12,090 --> 01:05:19,560 No. Is the answer to that. Well, what about the barometer falling every time a storm is about to start? 566 01:05:19,560 --> 01:05:28,180 Well, there's a there's a correlation, isn't there? So does the falling off the barometer cause the storm to start? 567 01:05:28,180 --> 01:05:34,230 Does this storm starting cause the barometer to fall? 568 01:05:34,230 --> 01:05:39,370 Not really. No. It goes far. I mean, what you've got is a drop in the atmospheric pressure. 569 01:05:39,370 --> 01:05:46,650 Is that a drop in the atmospheric pressure, the causes, both the storm to start and the barometer to fall. 570 01:05:46,650 --> 01:06:01,640 So you've got one thing. So you've got something like this, a structure like this. 571 01:06:01,640 --> 01:06:07,250 A and B are correlated. But that's because both are caused by sea. 572 01:06:07,250 --> 01:06:15,420 So this kott this relation is a correlation, but not a cause. 573 01:06:15,420 --> 01:06:25,980 This is a cause and this is a cause, but this isn't do you see so a correlation cannot be sufficient for further causation, can it? 574 01:06:25,980 --> 01:06:28,510 So humans got to be wrong, surely. 575 01:06:28,510 --> 01:06:40,650 And so sometimes correlations are coincidental and sometimes coincidences, correlations come about because they are both the result of a common cause. 576 01:06:40,650 --> 01:06:47,460 So how before Hume's theory can really be taken seriously? 577 01:06:47,460 --> 01:06:57,610 We've got to exclude accidental generalisations and other non causal regularities from the regularity theory, of course, of causation. 578 01:06:57,610 --> 01:07:05,520 Do you see that? Okay. What about the claim that regularities are not necessary for causation? 579 01:07:05,520 --> 01:07:13,620 Well, some smokers don't get cancer. You know, we say that smoking causes cancer, but everyone can tell the story of my dad, 580 01:07:13,620 --> 01:07:21,780 which is that he didn't die till 84 and yet he smoked 60 cigarettes a day since he was 16. 581 01:07:21,780 --> 01:07:25,710 And I mentioned the existence of the universe brought about by a big bang. 582 01:07:25,710 --> 01:07:30,340 There seemed to be at least there are exceptions. 583 01:07:30,340 --> 01:07:37,830 Sorry, there are causation is correlations to which there are exceptions that we think of as causation. 584 01:07:37,830 --> 01:07:43,230 And also, we looked at miniskirt. Do we not observe causation in the individual case? 585 01:07:43,230 --> 01:07:50,670 And so we've got to include this type of non regular causation in our accounts of causation. 586 01:07:50,670 --> 01:07:57,890 And how can you do that if he says that causation is regularity? 587 01:07:57,890 --> 01:08:01,540 Okay, so the solution. Okay. 588 01:08:01,540 --> 01:08:09,020 So it's woo overdone. So going back to problem one, regularities are not sufficient for causation. 589 01:08:09,020 --> 01:08:17,680 How does Hume account for that? Okay, so. 590 01:08:17,680 --> 01:08:20,110 He appeals to the well. He doesn't. 591 01:08:20,110 --> 01:08:30,130 We appeal on his behalf to the laws of nature to distinguish between those regularities that are causal and those that aren't. 592 01:08:30,130 --> 01:08:37,140 So no law of nature ensures that any male in this audience is a second son. 593 01:08:37,140 --> 01:08:44,170 But there is a law that underpins the fact that every Charles Bolden with Down's syndrome has Trisomy 21, 594 01:08:44,170 --> 01:08:52,540 has a third or a partial third copy of chromosome 21. 595 01:08:52,540 --> 01:08:59,890 So it's by appeal to the laws of nature that we say actually it's only regularities that are underpinned by the laws of nature. 596 01:08:59,890 --> 01:09:04,990 Now, if you're thinking what's the law of nature, that's good. 597 01:09:04,990 --> 01:09:11,920 The laws of nature also underpin relations between fall in the atmospheric pressure and a falling barometer. 598 01:09:11,920 --> 01:09:22,150 And between the fall and atmospheric pressure and the onset of the storm, no separate law relates these two. 599 01:09:22,150 --> 01:09:31,510 So in order for a relation to for a correlation to be a causal correlation, it's got to be underpinned by a law of nature. 600 01:09:31,510 --> 01:09:37,600 That's what's so we can distinguish between accidental correlations and causal correlations. 601 01:09:37,600 --> 01:09:43,960 And we can get rid of the first really quite serious problem for Hume's theory. 602 01:09:43,960 --> 01:09:50,010 So it's only regularities underpinned by a law of nature that are causal. 603 01:09:50,010 --> 01:09:57,450 And for the second problem, the one that says that that regularity isn't necessary for causation. 604 01:09:57,450 --> 01:10:04,950 We can insist that it only appears to be the case that this causation without regularity. 605 01:10:04,950 --> 01:10:11,280 So Mitchell says and couple of you have already given voice to this objection. 606 01:10:11,280 --> 01:10:18,360 Whatever we identify as the cause of a given effect is in fact only part of the cause. 607 01:10:18,360 --> 01:10:25,740 So when we say that the match licked because it was struck. 608 01:10:25,740 --> 01:10:33,290 And you rightly point out that actually it wouldn't have lit even though it was struck if there was no oxygen around. 609 01:10:33,290 --> 01:10:38,460 Mill says quite right. The striking of the match was only part of the cause. 610 01:10:38,460 --> 01:10:45,240 An oxygen was another part of the cause. Another part of the cause was how hard you struck the match and so on. 611 01:10:45,240 --> 01:10:54,450 So everything that's needed to make the cause sufficient for the effect has got to be in there as part of the cause. 612 01:10:54,450 --> 01:10:59,280 So we're partially describing the cause. 613 01:10:59,280 --> 01:11:03,450 So we pick out something is the cause. We're not sure it isn't the cause. 614 01:11:03,450 --> 01:11:13,890 It's just part of the cause. And Davidson Donald Davidson says whenever we observe a case of causation, we always believe that there is a law. 615 01:11:13,890 --> 01:11:18,510 But we don't know what that law is. So we see one thing causing another. 616 01:11:18,510 --> 01:11:31,410 We see causation in the individual case. And we think that if a so you see an individual case of A causing B, 617 01:11:31,410 --> 01:11:39,480 but that makes you think that there is a law that somehow means that A was sufficient. 618 01:11:39,480 --> 01:11:45,330 Fibi. There's some law that underpins that's happening. 619 01:11:45,330 --> 01:11:52,290 If A did cause B, then there's some law of nature that makes it the case that if you were to repeat that a. 620 01:11:52,290 --> 01:11:59,610 Exactly. And the same exactly the same circumstances, you would again get a B. 621 01:11:59,610 --> 01:12:06,450 That's what causation is. So we see causation in an individual case or in a few cases, 622 01:12:06,450 --> 01:12:14,370 and then later through observation and experiment, i.e. science will discover a regularity. 623 01:12:14,370 --> 01:12:26,750 If it's true that A causes B, then if we look hard enough, we will find that is a regularity underpinning that a law of nature underpinning of it. 624 01:12:26,750 --> 01:12:31,280 Okay, now, how are we doing time wise? OK. 625 01:12:31,280 --> 01:12:39,470 Now, John Mackey argued that individual causes must be at least heinous conditions for their effects. 626 01:12:39,470 --> 01:12:44,420 OK. A cause is an illness condition. If and only if that's not a misprint. 627 01:12:44,420 --> 01:12:58,040 It means if and only if the former i.e. the cause is an insufficient but necessary part of an unnecessary but sufficient condition of the effect. 628 01:12:58,040 --> 01:13:07,250 Now, if you think about again, then the think of a short circuit causing a fire, which is one that he was using. 629 01:13:07,250 --> 01:13:14,360 OK. The short circuit was not sufficient for the fire because there also had to be the flammable 630 01:13:14,360 --> 01:13:18,020 material that was sitting around it and the oxygen in the air and things like that. 631 01:13:18,020 --> 01:13:28,370 Okay. So the short circuit was insufficient, but it was necessary in that had the short circuit not occurred, the fire would not have occurred. 632 01:13:28,370 --> 01:13:33,380 Okay. But it was an unnecessary but sufficient condition. 633 01:13:33,380 --> 01:13:38,350 It was sorry. Insufficient but necessary part of. 634 01:13:38,350 --> 01:13:48,370 Okay. So the short circuit was part of a set of conditions that were not themselves necessary because the fire might have started for other reasons, 635 01:13:48,370 --> 01:13:52,480 but that were, in fact, sufficient for the effect. 636 01:13:52,480 --> 01:13:57,310 Can you see how this would work out? I mean, it's very complicated trying to put it like that. 637 01:13:57,310 --> 01:14:06,280 But if you work through this yourself. So you say a short circuit caused a fire and you can work out why the short circuit is 638 01:14:06,280 --> 01:14:16,090 insufficient but necessary part of an unnecessary but sufficient condition of the fire. 639 01:14:16,090 --> 01:14:26,500 We can talk about that in a minute. So this is the canonical statement of RTC as used today. 640 01:14:26,500 --> 01:14:32,620 C is a cause of V if and only if for some time early as an E notice, 641 01:14:32,620 --> 01:14:37,410 we're getting the temporal priority in there cause this come before their effects. 642 01:14:37,410 --> 01:14:44,280 C belongs to a set of events that occurred at T. 643 01:14:44,280 --> 01:14:50,730 That non redundantly suffices for E! In other words, it is sufficient for the effect. 644 01:14:50,730 --> 01:15:01,620 And what's more, it was necessary for the effect. And that is a cause of causing to the regularity theory of causation. 645 01:15:01,620 --> 01:15:09,410 And next week, we're going to look at a completely different theory of causation that was for years considered to be a rival to the RTC. 646 01:15:09,410 --> 01:15:14,120 So lots of people didn't like the RTC. But this one came along instead. 647 01:15:14,120 --> 01:15:24,090 And what a relief. It's so much better. Or so we thought, OK, I'm going to stop there and let's. 648 01:15:24,090 --> 01:15:31,480 See if we can make sense of that. OK? Is that thoroughly bamboozled you? 649 01:15:31,480 --> 01:15:35,340 Oh, yes, that would that would be good, actually. 650 01:15:35,340 --> 01:15:45,050 Douglas is going to go around with a with a microphone if we can find the. 651 01:15:45,050 --> 01:15:52,520 Well, actually, there isn't a mike here. Don't worry. 652 01:15:52,520 --> 01:15:58,550 People asking questions just have to shout. We'll do that at the end. 653 01:15:58,550 --> 01:16:03,530 So just to recap, what we've done is we've looked at what the causation is, 654 01:16:03,530 --> 01:16:11,770 a concept that we have that is absolutely essential to our ability to explain, predict or manipulate the world. 655 01:16:11,770 --> 01:16:19,010 And so it's quite simple. What is this causation? What is this causal relation? 656 01:16:19,010 --> 01:16:26,060 Hume comes along and says, well, okay, if we look at the way we use the notion of causation in ordinary language, 657 01:16:26,060 --> 01:16:36,690 we treatises the concept of necessary connexion. So if A causes B, then A is sufficient for B, it necessitates B. 658 01:16:36,690 --> 01:16:45,890 And then Hume says, but it can't be because we must experience necessity in order to have this idea of metaphysical necessity. 659 01:16:45,890 --> 01:16:55,130 Well, if it can't be, what is causation? And he looks at the evidence for causation and what we think and points out that 660 01:16:55,130 --> 01:17:02,240 actually it's only a constant conjunction or correlation that we ever use. 661 01:17:02,240 --> 01:17:10,910 And we could ever use as evidence for causation. And he thinks there is no more to causation than correlation. 662 01:17:10,910 --> 01:17:22,670 No more than constant conjunction. There's no metaphysical underpinning of that constant cause, constant conjunction. 663 01:17:22,670 --> 01:17:30,280 Okay, let's let's have a few questions. Going to go back to your last slide, the last statement. 664 01:17:30,280 --> 01:17:39,530 And I wonder if you could just explain what you've done. That one. 665 01:17:39,530 --> 01:17:48,300 Yeah. Let's let's use it again with the short circuit. 666 01:17:48,300 --> 01:17:53,550 So there's a fire in a warehouse and the. 667 01:17:53,550 --> 01:18:05,220 Where's my pen? And the fireman comes in and he says, the short circuit. 668 01:18:05,220 --> 01:18:09,660 Caused the fire. OK. 669 01:18:09,660 --> 01:18:15,240 We can, we can. That makes sense to us. We would accept that, especially as the firemen said it. 670 01:18:15,240 --> 01:18:24,280 But the full and short circuit was that it was not itself sufficient for the fire. 671 01:18:24,280 --> 01:18:31,730 Okay. It was insufficient for the fire because if there hadn't been that that flammable material next to it. 672 01:18:31,730 --> 01:18:37,760 So that when the short circuit occurred, the flammable material caught fire and uncertain. 673 01:18:37,760 --> 01:18:42,590 Okay, so the short circuit caused the fire. But it was not sufficient for the fire. 674 01:18:42,590 --> 01:18:51,210 All these other things were needed as well. But it was necessary for the fire. 675 01:18:51,210 --> 01:18:56,970 In other words, had the show had the short circuit not to Kolff occurred. 676 01:18:56,970 --> 01:19:00,710 The FIA would not have occurred. Okay. On this day. 677 01:19:00,710 --> 01:19:07,870 So. So the reason the farm is saying the short circuit caused the fire is because it's true according to the farm. 678 01:19:07,870 --> 01:19:12,510 And that had the short circuit not occurred, the FA would not have occurred. 679 01:19:12,510 --> 01:19:23,460 So it's necessary. But it's a. 680 01:19:23,460 --> 01:19:30,220 Hang on. Have I got this the right way round? I'm actually going back to this one. 681 01:19:30,220 --> 01:19:35,350 Insufficient but necessary part of an unnecessary okay. 682 01:19:35,350 --> 01:19:44,780 The total calls. Was sufficient. 683 01:19:44,780 --> 01:19:50,860 OK. Because if you combine the short circuit carrying when it did the flammable material beside it, 684 01:19:50,860 --> 01:19:56,320 the fact there was oxygen in the air, cetera, all of that was sufficient. 685 01:19:56,320 --> 01:20:01,960 But the total cause wasn't necessary because the fire might have occurred if 686 01:20:01,960 --> 01:20:07,420 some vandals have got in and sorted it with a match or something like that. 687 01:20:07,420 --> 01:20:19,990 So when the fireman says the short circuit caused the fire, it's shorthand, if you like, for the short circuit was not itself sufficient for the fire. 688 01:20:19,990 --> 01:20:28,090 But it was necessary. And the thing that was sufficient for the fire was the totality of conditions that caused the fire, 689 01:20:28,090 --> 01:20:34,520 which, of course, were not necessary for the fire because the fire might have started in other ways. 690 01:20:34,520 --> 01:20:44,150 OK. So it's all getting very complicated, but it's only when you get all those conditions in. 691 01:20:44,150 --> 01:20:50,130 OK, said so moving. How did that get translated into this? 692 01:20:50,130 --> 01:20:59,750 Sometime earlier than see. So the short circuit caused the fire. The fire happened at TI and the short circuits caused at T minus one. 693 01:20:59,750 --> 01:21:05,390 Okay. So the short circuit occurred at T minus one. And the fire occurred at T. 694 01:21:05,390 --> 01:21:16,850 So this is set of events. The total cause that occurred t that non redundantly suffices for E. 695 01:21:16,850 --> 01:21:23,540 In other words the. Short circuit was necessary. 696 01:21:23,540 --> 01:21:39,700 It wasn't redundant. They don't redundantly crucifies is go with C or the set of events centres, C, C beholders. 697 01:21:39,700 --> 01:21:48,840 Oh, I see, occurring to see. Yes, it's the set of events that non redundantly suffices for tea. 698 01:21:48,840 --> 01:21:53,390 But don't worry too much about this canonical statement. 699 01:21:53,390 --> 01:21:58,680 And they only I mean, I used to get very fed up reading Macchi. 700 01:21:58,680 --> 01:22:04,040 The only thing that you need to note to remember really are these three things. 701 01:22:04,040 --> 01:22:08,720 Okay, let's make it really easy for you. Quite properly. 702 01:22:08,720 --> 01:22:13,770 There's no I'm not dumbing down. This is going up too far. 703 01:22:13,770 --> 01:22:18,600 You don't need it. Although I've got to say, it says three things you need. 704 01:22:18,600 --> 01:22:34,670 What are they? Okay, so Hume says that causation is regularity and also temporal priority and spatial continuity. 705 01:22:34,670 --> 01:22:47,300 But but regularity is the important thing. The problem is that regularity and no causation. 706 01:22:47,300 --> 01:22:54,130 And causation and no regularity. 707 01:22:54,130 --> 01:22:59,500 OK, that's causation isn't sufficient for regularity and causation isn't necessary for regularity. 708 01:22:59,500 --> 01:23:01,120 Sorry. All the way round. 709 01:23:01,120 --> 01:23:11,590 So if your insisting that there's an identity between those two things, then it shouldn't be the case that you have can have one without the other. 710 01:23:11,590 --> 01:23:25,630 Should it? Okay. And so the solution is regularity and no causation. 711 01:23:25,630 --> 01:23:34,120 The solution is the laws of nature. When you have a regularity that isn't causal, it's because it's a coincidence or something like that. 712 01:23:34,120 --> 01:23:37,930 So in order to be causal, it needs to be underpinned by laws. 713 01:23:37,930 --> 01:23:47,080 Law of nature and the other one. You've got to think of total causes. 714 01:23:47,080 --> 01:23:55,080 And Paul, partial causes, so you may think says causation and no regularity. 715 01:23:55,080 --> 01:23:59,820 Because the short circuit has occurred, not every short circuit causes a fire. 716 01:23:59,820 --> 01:24:04,710 So why do you think this short circuit causes the fire? It is supposed to be a regularity here. 717 01:24:04,710 --> 01:24:15,900 Well, answer. You've got to look at the total cause, the whole set of conditions that were sufficient for the effect. 718 01:24:15,900 --> 01:24:30,190 Does that help a bit? Surely it's a regular occurrence, RVO. 719 01:24:30,190 --> 01:24:39,110 Well, the man was. Well, excellent. 720 01:24:39,110 --> 01:24:43,330 You are absolutely right. And I said the point. 721 01:24:43,330 --> 01:24:49,480 I introduced it. I said, if you're wondering what a law of nature is, that's a it's a very good question. 722 01:24:49,480 --> 01:25:07,900 So if if we were appealing to laws of nature to distinguish causal correlations and non causal correlations, 723 01:25:07,900 --> 01:25:14,020 i.e. accidents or two effects of the same cause or something like that, well, 724 01:25:14,020 --> 01:25:19,000 then a law of nature ought to be something different from a regularity, oughtn't it? 725 01:25:19,000 --> 01:25:29,440 And interestingly, the type of account of law of nature doesn't really distinguish between this. 726 01:25:29,440 --> 01:25:40,000 What it does is it science will come in and say, okay, every male in this room is a second son and is there a causal relationship there? 727 01:25:40,000 --> 01:25:46,420 Now, if all they looked at was the correlation between second sons and here, you would have to say there is a causal relation. 728 01:25:46,420 --> 01:25:55,900 But what you would do is you say, well, okay, if there's a law of nature there, it must be the same outside this room. 729 01:25:55,900 --> 01:25:59,290 And so they would try and do it elsewhere. They would try and see. 730 01:25:59,290 --> 01:26:06,160 And so actually a law of nature has got to be implicated in our best deductive systems. 731 01:26:06,160 --> 01:26:13,150 That's what a law of nature is. So it's a correlation that works over and over again in different places of the world, 732 01:26:13,150 --> 01:26:17,830 at different times of the world with different observers, et cetera. 733 01:26:17,830 --> 01:26:20,860 So you picked up a very good point. 734 01:26:20,860 --> 01:26:31,090 And it's it's interesting whether it's answered properly by saying that, because do you see that there's nothing but correlations there? 735 01:26:31,090 --> 01:26:40,920 It's just the correlations have got to work more generally. You don't need any mechanism. 736 01:26:40,920 --> 01:26:46,770 Once again, you're going back to a human idea that there is no more to causation than regularity, 737 01:26:46,770 --> 01:26:59,510 but the regularity must be, if you like, scientifically respectable. So science has got to establish it happens more often than just in this room. 738 01:26:59,510 --> 01:27:03,870 That's every male in this room is his second son. Could be coincidence. 739 01:27:03,870 --> 01:27:11,460 But if every male who ever goes to a lecture anywhere is a second son, that starts to look really quite interesting. 740 01:27:11,460 --> 01:27:15,750 I mean, you might start thinking of possible explanations for that. 741 01:27:15,750 --> 01:27:21,350 But but if you do find an explanation, it'll be a correlation at a lower level. 742 01:27:21,350 --> 01:27:28,650 You can't speak up in the canonical statement of gravity theory. 743 01:27:28,650 --> 01:27:37,810 What is it? C which occurs in the. 744 01:27:37,810 --> 01:27:49,900 The first billiard ball hitting the other one. And the second one rolling off simultaneously. 745 01:27:49,900 --> 01:27:57,850 Why didn't they do, do they? Well, I mean, there is a question whether there can be I mean, if you miss right, there cannot be simultaneous causation. 746 01:27:57,850 --> 01:28:05,460 And lots of people have asked, you know, why there couldn't be simultaneous causation. 747 01:28:05,460 --> 01:28:15,640 This is this statement, modifiable possibility of simultaneous causation, because I can't see if if. 748 01:28:15,640 --> 01:28:23,770 Well, you could just say see below. See is the cause of if no new F C belongs to a set of events, that's non redundantly suffices free. 749 01:28:23,770 --> 01:28:31,780 Why. What you could just take the temple reference out because in such a case are describing the totality of the situation, 750 01:28:31,780 --> 01:28:42,160 in which case it becomes indistinguishable from the effect. Well, you could say that, yeah. 751 01:28:42,160 --> 01:28:49,840 Bob, speak up. Did. Is it thought that Hume had any answers of the questions somebody might have asked him? 752 01:28:49,840 --> 01:28:57,070 What is it that causes the laws of nature to be what they are? 753 01:28:57,070 --> 01:29:01,110 Well, you it would be. 754 01:29:01,110 --> 01:29:04,000 I mean, Huband himself didn't talk about laws of nature. 755 01:29:04,000 --> 01:29:12,880 It was it was M. who introduced laws of nature, an explanation of regularities that are not causal. 756 01:29:12,880 --> 01:29:22,900 So you can't really ask him that question. But you could say, well, what is the explanation of this law of nature? 757 01:29:22,900 --> 01:29:30,550 And you would look further down and all you would find, if humour is right, is yet another correlation. 758 01:29:30,550 --> 01:29:35,330 So you might be able to find an explanation of one. 759 01:29:35,330 --> 01:29:43,100 Correlation in terms of another correlation, and you might be able to find an explanation for that lower correlation in terms of lower correlation. 760 01:29:43,100 --> 01:29:52,870 But what you will never find if you miss right, is is a relation that isn't just a regularity. 761 01:29:52,870 --> 01:30:03,400 So you'll never find a physical relation, for example. So science will never say are this physical relation is the causal relation? 762 01:30:03,400 --> 01:30:16,900 Because Hume thinks that's doesn't make sense. OK, one more, then we'd better finish, you know, cause that. 763 01:30:16,900 --> 01:30:24,380 Well, that that statement make makes. I mean, you can use that in so many different contexts. 764 01:30:24,380 --> 01:30:33,970 And I mean, it could be a moral claim or. But I don't see why he would be saying that. 765 01:30:33,970 --> 01:30:38,110 Well, he is saying that that all our ideas come from impressions. 766 01:30:38,110 --> 01:30:45,070 But that's not the same as saying that everything is relative. Nothing is absolute. 767 01:30:45,070 --> 01:30:55,600 I mean, I don't know why you would want to say that in this context. 768 01:30:55,600 --> 01:31:01,010 Okay, and you're saying that that means that everything's relative, everything's relative to. 769 01:31:01,010 --> 01:31:10,230 Okay. If you want to think of it that way, I would rather not. 770 01:31:10,230 --> 01:31:14,940 But I'm not quite sure how you're understanding it, so I'm not quite sure. 771 01:31:14,940 --> 01:31:19,890 If I if I can if I can say that you're wrong or indeed that you're right. 772 01:31:19,890 --> 01:31:21,730 Okay, let's let's finish there. 773 01:31:21,730 --> 01:31:28,750 Next week, I'll look at the counterfactual theory and you'll find that as we go through looking at the different theories, 774 01:31:28,750 --> 01:31:44,387 we'll give you a better grip on what causation is. Altogether, I mean, I hope you won't go away from these lectures knowing what the causation is.