1 00:00:10,850 --> 00:00:23,060 So it seems, as we said before, that all of our reasoning about the world, all of the reasoning by which we reach any matter of fact at all. 2 00:00:23,060 --> 00:00:31,160 Beyond those we immediately perceive or remember. Is based on an assumption of uniformity. 3 00:00:31,160 --> 00:00:37,250 An assumption that what we have experienced is a reliable guide to what we haven't experienced. 4 00:00:37,250 --> 00:00:44,390 And if Hume is right, that assumption is based on nothing at all other than animal instinct. 5 00:00:44,390 --> 00:00:49,500 There is no rational basis for it. Whatever. 6 00:00:49,500 --> 00:00:55,530 And that's the famous humour, scepticism about induction. 7 00:00:55,530 --> 00:01:00,630 It's been an inspiration to huge numbers of philosophers of science. 8 00:01:00,630 --> 00:01:12,270 It was seen by many as a complete crisis, by some as an opportunity, the kind of philosophy of science that you find in Popper. 9 00:01:12,270 --> 00:01:21,300 To some extent in Kyun and many, many other philosophers of science takes its start from Hume's results. 10 00:01:21,300 --> 00:01:24,750 Popper, for example, tried to give an account of science, 11 00:01:24,750 --> 00:01:34,860 which in no way depended on induction because he thought that Hume had completely undermined that basis. 12 00:01:34,860 --> 00:01:46,170 As I've said, it seems to imply that our human reason that we tend to be so proud of is actually different from animal reasoning only in degree. 13 00:01:46,170 --> 00:01:50,090 It's not fundamentally different in kind. 14 00:01:50,090 --> 00:01:57,080 The sort of supposed perception of probable connexions that Locke had thought was the basis of human reasoning. 15 00:01:57,080 --> 00:02:01,310 Turns out to be wishful thinking. There is no such perception. 16 00:02:01,310 --> 00:02:08,480 When we when we think that we have insight into the way physical things behave at bottom, 17 00:02:08,480 --> 00:02:14,300 there is this assumption of uniformity which is based on no insight, whatever. 18 00:02:14,300 --> 00:02:21,660 And our understanding of causation. Is not really based on intelligibility. 19 00:02:21,660 --> 00:02:35,500 It's based on observation of uniformity. Now, I'm not going to say very much about Hume's particular view of this, but just very briefly, 20 00:02:35,500 --> 00:02:41,530 does it imply a complete irrational point of view if you go with Hume on this? 21 00:02:41,530 --> 00:02:46,260 Does it follow that anything goes, 22 00:02:46,260 --> 00:02:52,620 that there is no difference between the scientist and the superstitious enthusiast who 23 00:02:52,620 --> 00:02:58,270 base his predictions about the future on the shape of the tea leaves in his teeth, 24 00:02:58,270 --> 00:03:08,990 in his tea cup? Or on tarot cards or astrology or any other superstition you care to mention. 25 00:03:08,990 --> 00:03:15,430 Well, Hume didn't think so. He did deny this inductive inference is founded on rational insight. 26 00:03:15,430 --> 00:03:25,530 But he didn't want to say that, therefore, anything goes. But that raises a major problem of demarcation. 27 00:03:25,530 --> 00:03:29,610 And again, this is a problem that is echoed down the centuries since. 28 00:03:29,610 --> 00:03:37,080 What right do we have for preferring scientific reasoning to superstitious reasoning? 29 00:03:37,080 --> 00:03:41,970 If the ultimate ground of scientific reasoning is just an animal instinct. 30 00:03:41,970 --> 00:03:48,870 So we have animal instincts to be scared of certain things or to have certain superstitions. 31 00:03:48,870 --> 00:03:57,940 Why should we give any less respect to those than we give to our animal instinct that underlies science? 32 00:03:57,940 --> 00:04:07,460 Well, Hume's answer is basically to favour consistency. Hume wants to say that everything we do in life is based on the assumption that 33 00:04:07,460 --> 00:04:12,450 we can learn from experience and that the future will conform to the past. 34 00:04:12,450 --> 00:04:15,840 We can't even wake up in the morning wash. Go out the door. 35 00:04:15,840 --> 00:04:24,300 Eat our breakfast without making assumptions that the behaviour of things in the past is a guide to their behaviour in the future. 36 00:04:24,300 --> 00:04:34,710 Even the entirely superstitious person has to rely on almost everything that they do on that assumption of uniformity and systematic behaviour. 37 00:04:34,710 --> 00:04:43,410 So Hume's answer is to say, well, in that case, the rational thing to do is to accept that, accept that we are part of nature. 38 00:04:43,410 --> 00:04:48,120 Accept that this assumption is one that we simply cannot live without. 39 00:04:48,120 --> 00:04:53,040 And now follow through the consequences. And if you follow through the consequences, 40 00:04:53,040 --> 00:05:02,040 if you remain faithful to that assumption that the basic laws by which nature works are consistent over time, 41 00:05:02,040 --> 00:05:04,320 since you have to assume it in your daily life. 42 00:05:04,320 --> 00:05:12,510 Why not make that the model and then try and systematise what you discover about the world in conformity with that? 43 00:05:12,510 --> 00:05:23,190 That is where science comes from. And according to Hume, it gives a reliable basis for preferring science to superstition. 44 00:05:23,190 --> 00:05:30,990 But it does mean that our attempts to understand the world are reduced to the kind of thing that Newton did in the case of gravitation, 45 00:05:30,990 --> 00:05:36,840 the utmost effort of human reason is to reduce the principal's productive of natural phenomena 46 00:05:36,840 --> 00:05:44,970 to a greater simplicity and to resolve the many particular effects into a few general causes. 47 00:05:44,970 --> 00:05:52,310 But as to the causes of these general causes, we in vain attempt their discovery. 48 00:05:52,310 --> 00:05:58,820 So we have to make do with the science in which the ultimate principles are ones that we just have to accept. 49 00:05:58,820 --> 00:06:05,030 And as I've suggested in the case of quantum mechanics, that is, in fact where we are. 50 00:06:05,030 --> 00:06:09,560 If you now look at what Hume has to say about science, a lot of it will look like common sense. 51 00:06:09,560 --> 00:06:12,130 At the time, it was very far from common sense. 52 00:06:12,130 --> 00:06:24,340 And it's a mark of how far we've come that we now accept that the ultimate principles of science are ones that we cannot hope to base on pure reason. 53 00:06:24,340 --> 00:06:29,510 OK. The last few slides. I'm just going to go summarise very quickly. 54 00:06:29,510 --> 00:06:33,470 Many people have attempted to answer Hume in all sorts of different ways. 55 00:06:33,470 --> 00:06:39,830 All I'm trying to give here is a little roadmap so that you can see where some of these attempts fit in. 56 00:06:39,830 --> 00:06:45,950 To discuss them with any sort of depth would require at least another lecture or two. 57 00:06:45,950 --> 00:06:57,530 So I shall not attempt to do that. One way of trying to answer Hume is to show that actually induction can be justified by pure reason, 58 00:06:57,530 --> 00:07:05,570 but by appeal to probability rather than demonstration. Here are some of the famous names of those who've tried to do this, 59 00:07:05,570 --> 00:07:10,910 including a Simon Blackburn and John Mackey, both Oxford philosophers in their time. 60 00:07:10,910 --> 00:07:20,110 Roy Harod. Other attempts to to answer Hume include the so-called analytic justification of induction. 61 00:07:20,110 --> 00:07:27,680 This is associated particularly with Peter Straughan. The claim is that induction is rational by definition. 62 00:07:27,680 --> 00:07:30,980 When we think of what is a rational way to behave, 63 00:07:30,980 --> 00:07:41,810 basing your assumptions about the future on the past just is rational to assume that what's happened in the past is a reliable guide to the future. 64 00:07:41,810 --> 00:07:48,660 That is what just what the rational person does. How can we make sense of rational behaviour which didn't do that? 65 00:07:48,660 --> 00:07:59,060 And so the claim is that induction, no sceptical problem can be raised about induction in the way that Hume was thought to do. 66 00:07:59,060 --> 00:08:03,820 Some philosophers have argued that induction can be justified by its past success. 67 00:08:03,820 --> 00:08:09,070 Inductive reasoning has always worked very well for us. It's worked in the past, should work in the future. 68 00:08:09,070 --> 00:08:10,700 Shouldn't it? 69 00:08:10,700 --> 00:08:20,840 And the pragmatic justification of induction attempts have been made to show that even if we cannot justify relying on induction by pure reason, 70 00:08:20,840 --> 00:08:35,390 we have pragmatic reasons for doing so. Practical reason. Briefly, I don't think any of these answers really hit him very strongly. 71 00:08:35,390 --> 00:08:41,180 He would agree that we describe induction as a rational, rational way of behaving. 72 00:08:41,180 --> 00:08:51,290 He himself advocates that we rely on it. And inductive justification does just seem circular. 73 00:08:51,290 --> 00:08:58,380 It seems pretty obviously circular. I think ultimately it pretty clearly is circular. 74 00:08:58,380 --> 00:09:04,260 The pragmatic justification doesn't touch Hume's position because Hume himself, after all, 75 00:09:04,260 --> 00:09:09,240 says that we irresistibly, inevitably assume that the future will resemble the past. 76 00:09:09,240 --> 00:09:13,860 Anyone who comes along to me with a pragmatic justification of induction, 77 00:09:13,860 --> 00:09:19,830 who tries to preach to me and tell me why I should rely on it, is wasting his time. 78 00:09:19,830 --> 00:09:26,350 I'm already bound to rely on it. It's the way I am. 79 00:09:26,350 --> 00:09:40,690 The only question is whether I will rely on it consistently. And finally, I end with a couple of slides on Hugh Mellor and on Nelson Goodman. 80 00:09:40,690 --> 00:09:45,340 I'm not going to attempt to bring those into the discussion here. 81 00:09:45,340 --> 00:09:50,500 I've put them there because they are amongst the reading that you will get on this topic. 82 00:09:50,500 --> 00:10:01,578 And I hope that what I've written there will help you to assimilate those into the general framework that I'd given you.