 Coming up next, we have Susan Hack. She's going to talk about credulity and its consequences. Now, she's at the University of Miami. She's a distinguished professor in the humanities. She's a Cooper senior scholar in arts and sciences. She's a professor of philosophy, excuse me, and a professor of law. And as a professor of law, she teaches about scientific testimony, which she's going to talk about a bunch right now. She's, check this out, she's one of the Sunday independence 10 most important women philosophers of all time. That's a cool t-shirt. Here's her haiku. Most of what I know, I learned from Monty Python's philosopher's song. Please welcome Susan Hack. Thank you. I have to say, that thing in the Sunday independent is something of a burden, but never mind. All right, now, let's see if I can work this machine. OK, I ought to admit, I am very risk-averse. The very idea of gambling makes me itch. So if I scratch a little, forgive me. I never thought I would find myself here, especially not with people I can relate to. When I thought about what to talk to you about, I began recalling several words that my students last year didn't understand. Poor kids, where have they been all their lives? There's a pattern to these words. First one they didn't understand was pejorative. They thought it meant demeaning. Oh my God, suppose it did. Then I couldn't say to them, look, the idea in your paper isn't bad, but the writing's terrible, we're going to have to fix it. That would be demeaning. This is terrible. Tendentious, they just look at me funny. And creduity, to my astonishment, drew a complete and utter blank. They have no idea what the word means. I'm by now more or less bilingual in both British and American English, though my American accent's terrible. But I always go to both dictionaries in case there are interesting differences, which there sometimes are. The OED says, credulous means too ready to believe. Marion Webster says it means ready to believe, especially on slight or uncertain evidence. I noticed that the first of those, the Oxford one, makes the word inherently pejorative, whereas the American one doesn't. I shall actually play on those two senses in the course of the talk. Credunity in the neutral sense, the Marion Webster sense, is, I think, part of human nature. Indeed, if children didn't mostly believe what they were told, they wouldn't learn as fast as they do. Richard Dawkins, I believe it, is once called children, information caterpillars. I thought that was one of his better lines. But as we grow up, most of us learn by experience to curb this innate credunity to some degree. I think that's why Alexander Bain, who was a remarkable psychologist, philosopher, character, described our entire intellectual life as innate credulity tempered by checks. That's to say, initially you just believe whatever you're told and then you find, oh, that can't be right. I take it credulity is a matter of degree and that some people are more credulous than others. And most people, at least most of the people I know, are more credulous on certain topics or with respect to certain sources than they are on other topics or with respect to other sources. So it's not a matter of are you credulous or aren't you? It's a matter of with respect to whom or what. From in what follows, I'm going to use the word this way. I'll distinguish being too ready to believe, I'll call that credulity. Being too ready to dismiss ideas, I'll call that the doubting Thomas attitude. And proportioning your belief to the strength of the evidence, I'll call that the rational attitude, just for shorthand, there's nothing deep in the names. You'll notice with this three-fold distinction, essentially what I'm doing is adding one more line to Aristotle's doctrine of the mean. Aristotle had the idea that the virtues were at least often a mean, an intermediate between two extremes. So that courage would be the mean between cowardice on the one hand and rashness on the other. And if I can go back for a minute, I've simply added another line to Aristotle's. Some people are credulous about political discourse. Most people of that tendency, I think, are more credulous about political discourse of one persuasion rather than of another. Some people are credulous about religious or mystical claims, or about religious or mystical claims of one kind or another. Some people are very credulous about gossip or certain kinds of gossip. At least I think they must be because the stuff I read in People's Magazine and the hairdresser, actually, it's enough to curl my hair by itself. It doesn't need any. Some people are credulous, especially, about medical treatments or dietary advice. Some people are about investments or, if they won't throw me out of the hotel, lottery odds. A lot of people, I regret to say, are very credulous about universities and other institutions' publicity materials and such. Maybe I'm a doubting Thomas on this subject, but every time the university administrator uses the word excellence, I just dive under my desk. And some people are credulous with respect to scientific claims. For example, to claims made by perfectly respectable scientists, which are, however, at a certain point in time, pure speculation. They may play out, they may not, and people too often go for them, not knowing whether or not they will play out. And credulous, some people are also credulous about claims made by well-known scientists about fields which are outside their competence. A scientist is not a general practitioner. He or she is usually quite a specialized person, and his opinion about the likely success of some speculation in another area is probably not worth a great deal more than mine. Credunity about scientific claims is by my lights a form of scientism, which in my mouth is a pejorative word meaning being too deferential to the sciences. And I think scientism is as undesirable as its opposite, let's just say, undue cynicism about science. That was the point of this book. Well, any point of this book, I don't know if it's big enough for you to read, but I can just about see the subtitle between scientism and cynicism. Don't buy this book if you don't have your own teeth because you can't pronounce the subtitle. This doesn't mean you shouldn't believe any scientific claim you can't check for yourself. If it meant that, you shouldn't believe many scientific claims at all. It does mean you need to be aware that within the sciences, there is a continuum from the really, really well-established to fairly well-established to, they're quite well-established to the quite speculative, to the purely speculative. You should also be aware that you need to figure out whom to trust on topics beyond your competence. That's not easy. That's actually very hard because by definition, if it's beyond your competence, there are things that they understand that you don't. You also need to be aware of what can go wrong with scientific inquiry and of how evidence can be misleading, how a scientist may be trapped in a pocket of misleading evidence through no fault of their own. This doesn't require you have a sophisticated theory of evidence. It does require, I think, that you have some intuitive grip on how complicated evidence can be. One thing that I feel kind of passionately about is that you need to understand evidence doesn't just have to support the claim that you're asking about. It also has to be reasonably secure itself, supporting the claim however well won't do it if the evidence that's doing the support of it is supporting is itself floppy. And there also has to be enough of it because if there's something really important that you're missing, all the evidence you have, however strongly it points in this direction, may be misleading. I might have been the only person in the country who didn't laugh derisively when Donald Rumsfeld made that remark about the unknown unknowns. I actually think, whatever you think about in the Secretary of Defense, I won't go there. He might have a future as an epistemologist because relevant evidence that you don't have and which moreover you don't realize is relevant really can screw up your belief system, something alarming. You also have to understand that a person's reliability is a source of information. Doesn't just depend on whether he's honest. It doesn't just depend on whether he's a good guy. It also depends on whether he's competent in the matter. And these are two quite different things. And it requires that you be willing to recognize when you're just not in a position to have a belief on this or that question. One thing I notice is, I do this, I seem to do this a lot more than most people. Sometimes my husband says, do you think this will happen? And I will say, I have no idea doing this. And he will say yes, but do you really think it will happen or not? And I said, no, look, when I say I have no idea, I mean I have no idea. Being willing to say I don't know is a very important part of avoiding the bad kind of credulity. All right, now my title promised credulity and its consequences. What are the consequences of credulity? Well, first of all, a credulous individual is a gullible individual. By the way, when I was playing with the dictionary, I discovered the origin of the word gull. I didn't know this before. Apparently it's a metaphor for you. The way gulls cram food down their beak when you can cram false information down the throat of a gullible individual. Because a credulous individual is gullible, he's susceptible to conmen, to crooks, to fakers. Moreover, credulity is infectious. People like to conform. If everybody else is believing this, they feel kind of awkward about saying, I don't see it. Moreover, okay, I think this is probably the best line ever written on the subject of credulity. By Winincand and Clifford, from whom we might have heard a lot more good stuff had he lived longer, as you see, he died very young. But not before he'd said, the credulous man is father to the liar and the cheat. What a great line. What he means, I take it, is that a credulous population creates the market for conmen, crooks, fakers, et cetera. And for every kind of deceptive and misleading claim. You know, those Nigerians trying to send you wild investment schemes, presumably wouldn't find it worthwhile if there were credulous people out there who jump at the chance. And so, the credulous man is a danger not only to himself, but also to others and to society at large. It's kind of like a virus. Ah, okay. I see this all the time in my double professional life. I thought about talking about how I see it in philosophy and then I thought, no, look, I will talk about philosophy in the morning. Let's find some really fine examples from the law in the afternoon, so that's what I'm doing. First example. Many jurors seem far too willing to believe eyewitnesses. Whom, however, we know are thoroughly unreliable. They are not as many jury eligible people will ask you, will tell you if you ask them, like human video cameras. On the contrary, they are filling in their memory. Only a minority of jurors seem to understand that eyewitnesses degree of confidence that this is the guy he saw is only very, very weakly correlated with the likelihood that the witness is right. This can have really tragic consequences. As you see, a very high proportion of those cases in which people who'd been convicted of a crime were subsequently exonerated when DNA analysis became possible were convicted in large part on the testimony of eyewitnesses. All of them, no doubt sincere, but as usual, many of them mistaken about who they saw. Credulity is also relevant in encouraging expert witnesses who in our system are prepared and presented by the parties to a case, either to trim or very often to exaggerate how sure they are to serve their side. I have a few perfectly horrible examples. Dr. James Griggson, I don't know if he's still with us, but he was known in Texas as Dr. Death. I should explain. Texas death penalty law is, I believe, the only state statute that requires, at the sentencing phase, that the jury satisfy itself whether this person, if allowed to live, would be dangerous in future. And Dr. Griggson testified in, I believe, more than 90 such cases. In all but one of which, the jury sentenced the person to death. In 1995, Dr. Griggson was expelled from the American Psychiatric Association for irresponsible testimony in Texas court, after which they took him off the death penalty cases and putting on the nursing home cases where he continued to do his terrible thing. Oh, I also think of the Supreme Court ruling in Barefoot in 1983, where Justice White wrote, admitting Dr. Death's testimony, that Barefoot would be dangerous in future, didn't violate his constitutional rights. It was perfectly okay. This was for the jury to sort out. Even though the Supreme Court had in front of it an amicus brief from the American Psychiatric Association, which said, we can't do this. Our predictions about who will be dangerous in future are wrong two times out of three. You'd be better off flipping a coin. Oh, okay, now if you'll forgive me, I'm getting low-called because every now and then I get involved in Florida cases and get fascinated by the expert testimony at issue. Joseph Ramirez was convicted, not once, but three times, of a stabbing murder, that's to say, convicted once, appealed, new trial convicted again, appealed, new trial. Actually, he's just been convicted a fourth time. I'll tell you about that in a minute. He was convicted in the first three instances, almost entirely on the testimony of a knife mark examiner who said, all that we got left was a bit of cartilage from the neck with a wound in it, half an inch long. And he said, I can identify the knife that made that wound with 100% certainty to the exclusion of all other knives in the world. He even said, I wouldn't care if there were a million knives of identical design. I could still identify this one with 100% certainty. I guess in this room, I can say what I really want to say. Bullshit, I've now done serious looking into this and I'm convinced that this is the story. A knife mark examiner can tell what kind of a knife. And if there's some very distinctive feature of the knife, like it's been worn in such a way that there's a distinctive chip, then but only then can he identify a particular one. Of course, it isn't only forensic witnesses whom Michael Sacks once described as serial exaggerators. I like that. They're always much too sure, which reminds me that David Hume famously said, we distrust a witness when he is too vehement. Why have we forgotten this? If he's banging his fist and saying, I'm absolutely certain you should wonder why is he doing this? Is it because he's not really so sure? Okay. The exaggerations of knife mark examiners, fingerprint examiners, et cetera, are encouraged by credulity on the part of jurors, on the part of attorneys and on the part of judges. So I don't mean to pick on jurors, although I do think one of the reasons it's important to try to teach ordinary people what credulity is, why it's a bad thing, how to avoid it, is ordinary people will serve on jurors. And if jurors are credulous, we are all in trouble. We want jurors to have in the right sense a kind of skepticism about what they hear, to be aware, you may be being lied to, you may be being told something which is a gross exaggeration of what the person concerned is really in a position to say, or he may believe it himself, but he believes it himself because he's a member of a guild. Oh Lord, may I tell you my knife mark examiners story? I was trying to help the public defender working with Mr. Ramirez and was really very shocked by this terrible testimony and began looking into it and I guess it got known I was looking into it because one day I received a bizarre telephone call from California from a knife mark examiner and he said, I want to explain to you the problem is it's East Coast knife mark examiners. We on the West Coast do it properly. Okay, the guys on the East Coast are incompetent, but we do it right. And I said, well, that's very interesting. Would you mind explaining to me what it is that you do and that the East Coast guys don't do? Well, he said, they just take the knife and they make stabbings in dental gel which apparently has roughly the same texture as cartilage and then they eyeball the wounds that they've got left in real cartilage and the thing that they've got in the dental gel to see whether they're the same. Me, yes, I'm aware of that. And what do you do? Well, we take photographs of both and then we eyeball them to see if they're the same. Me, okay, thank you, that was very helpful. I think I need to be in class now. So actually he called me at home which was my husband said, what were you doing for an hour on the phone? Trying to understand if there was anything in these claims and in the end I think the answer was no. This does however teaches me something important I think. The reason he thought what he did was better was that it used more modern equipment. Used fancy photography to get two images that they could compare. And it's another important thing to remember that using scientific technology doesn't necessarily mean that you're getting good upshot. Of course, it's not only jurors who are credulous, it's attorneys, it's some judges. I think many judges eventually get, as one federal judge once said to me, I don't know much about science, but I do know when I'm being snowed. And I think perhaps she was right. I think a judge with a lot of experience in cases involving scientific testimony will get some sort of sense for when they might be being snowed. But I think it's also very clear that some judges are quite credulous about science. I want to talk about the credulity about the peer review process. I need to give you some context. The Supreme Court had never uttered any ruling about scientific testimony until 1993. It had been in existence for 204 years, but not until 1993 did it feel the need to say anything about scientific testimony and when it should be admissible. This was in a case called Daubert versus Merle Dau Thalmasuticals, which was about, initially it was about a drug called Bendectin prescribed for mourning sickness in pregnancy and believed by some to cause birth defects in the children born to the women who took it. By the way, in the last two months, this drug which was withdrawn from the US market in 1984 was approved for sale in the US again, now made by a different manufacturer, a Canadian manufacturer and tall called Dioclesis. I still don't know whether or not it caused this birth defects. I must have read 100 Bendectin cases and it's impossible to tell. Anyway, in the course of this case, the Supreme Court finds itself obliged to say, look, first of all, we think judges should screen scientific testimony, first for relevance, okay, nothing new there and for reliability, that was new. And because most judges don't know anything about science, they then felt, well, we ought to say something about how they're to do this. Well, the first move was to show that it's reliable and you have to show that it's genuine science. No, this was a big mistake. Reliable and scientific are two quite different concepts. Not all scientific evidence is reliable and not all reliable evidence is scientific. Okay, put that aside. And what constitutes genuine science? Oh, dear, now they wheel in Karl Popper. I'm rolling my eyes, but never mind. Then they try to give a list of what should you look for if you're screening scientific testimony, proffered scientific testimony for reliability. And on this flexible list of four factors is peer review and publication. And as I've watched this play out and as I've talked to law students about what they understand by this, as it's called, Dalbert Factor, I've discovered something really interesting. They think that what goes on in the peer review process is you submit a paper to a journal and then somehow or other somebody re-does your experiment or re-does your study and checks everything and only after that's happened is it published. Why, where did they get this romantic idea? I think because American law reviews are not peer reviewed. And they think, well, the law reviews are full of rubbish, but that must be because they're not peer reviewed. And some judges really do seem to behave as if they thought if the paper was peer reviewed, that means it must be right. And I take it everybody in this room knows that it means nothing of the kind. It means perhaps it's quite interesting. It's not obviously mad. It doesn't contain big, enormous, obvious mistakes. That's about all it means. Think of the study of Vioxx, the approved study that got Vioxx FDA approval published in the New England Journal. You don't get any more prestigious than that. If their peer reviewers don't notice something, nobody's peer reviewers will notice it. And guess what? They failed to notice that Merck study kept track of gastrointestinal effects of this drug which they assumed would be favorable for longer than it kept track of cardiovascular effects which they feared might not be favorable. Horrible design flaw. Judges are also remarkably naive, at least some of them I shouldn't be too general. Some of them are remarkably naive about the meaning of statistical significance. They appear to think that if this study is statistically significant, that means it's significant. Well, no, that's not what it means. And this of course makes it very easy for the parties to the case to be impressed a judge with misleading statistics. And now I have exactly zero time to keep talking. And it was quite hard to pack this into half an hour. Phew.