If Books Could Kill - Malcolm Gladwell's ＂Blink＂

00:00:00.000 Michael. Peter. What do you know about Blink? All I know is that my first impression of this book

00:00:05.580 was that it was very dumb. And I'm looking forward to hearing about how first impressions are always 0.96

00:00:09.760 true. All right. The book is Blink by Malcolm Gladwell. Heard of it. Heard of him. Big best

00:00:29.300 at least a few million copies sold came out in 2005 this is also the first gladwell book i think

00:00:36.740 just based on my sort of casual perusal to get some very negative early reviews i think there

00:00:43.960 was like a good number of people at this point who were getting wise to the shallowness of the

00:00:49.420 pop science airport book genre yeah this is this is at the time when like four of the top 10 ted

00:00:57.000 talks are just like thoroughly debunked i'm sorry grit is not a thing i guess it's sort of perfect

00:01:02.100 for me because i am ever a contrarian and i guess the twist here is that i kind of like this book

00:01:07.920 i think yeah i think this book is pretty good i think we should talk about your complicated

00:01:11.820 feelings on this because you've been agonizing about this episode for like weeks i have i expected

00:01:16.200 this book to be pseudoscientific trash yeah same there is a lot of pseudoscience in this book and

00:01:23.300 we'll talk about it. But like, I don't think it's quite as bad as some of his other work.

00:01:28.180 Even though he makes a lot of these narrative mistakes, he touches on a bunch of really

00:01:33.000 interesting science. And I'm not an expert. I might have missed something, especially because

00:01:37.840 this book is just a torrent of anecdotes. There are so many anecdotes. It's wild. But I didn't

00:01:45.360 find too many significant cases of him misrepresenting the science.

00:01:49.740 Or your brain is so cooked from hosting this podcast that you're like, only 30% of the

00:01:53.540 anecdotes are fake.

00:01:54.900 Rookie numbers.

00:01:56.000 Before we get into it, we have housekeeping, right?

00:01:58.500 For the first time ever.

00:01:59.780 Yeah. 0.97

00:01:59.960 We've decided to do one episode where we do a little bit of sellout shit because we wanted 0.93

00:02:04.920 to plug our merch. 0.95

00:02:06.200 We have merch.

00:02:06.880 T-shirts, sweaters, mugs, the whole, you know, the usual stuff.

00:02:10.500 Ifbookspod.dashry.com.

00:02:12.300 We'll put a link in the show description. 0.81

00:02:14.340 And then also, because we're already doing sellout shit, subscribe to our Patreon. 0.95

00:02:19.060 Lean in. 0.98

00:02:19.740 We have monthly bonus episodes.

00:02:22.000 Get Peter's newsletter.

00:02:23.060 Oh, yeah.

00:02:23.460 My newsletter, stringinamaze.net.

00:02:25.460 What do you want to sell?

00:02:26.380 Anything?

00:02:27.020 Nothing.

00:02:27.960 I'm against this entire thing, but we're doing it because you want to mention it on the show.

00:02:31.500 This is why you put out teasers of our bonus episodes that are 85% of the episode.

00:02:38.800 I mean, whatever.

00:02:40.340 I feel like there is a huge contingent of listeners who just think that like some episodes cut off with like a little sound.

00:02:45.780 Because nobody reads the descriptions of podcasts.

00:02:47.780 So I feel like some people are just like, well, that was an hour long podcast that ended abruptly and have no idea that we have a Patreon or like anything else.

00:02:54.380 That's because Michael does teasers for these episodes that consist of just an unbelievable amount of the episode.

00:03:02.080 I'm trying to get the word out.

00:03:02.960 You don't do it for every episode, which is the only reason that we're still friends.

00:03:10.320 But I swear this last one we put the last one where you did 60 minutes of an 80 minute episode.

00:03:16.040 I thought that you were trying to pull some kind of prank on me.

00:03:20.200 I thought that this was like a joke that was specifically targeting me where you're like,

00:03:24.420 let's see how much of this episode I can release for free without Peter getting mad.

00:03:27.620 We did not talk about it specifically, but I just said, I'll schedule a teaser.

00:03:30.740 And then you're just like, I'll publish a teaser.

00:03:32.780 And I was like, cool.

00:03:33.500 And then I saw that it was 60 minutes long.

00:03:37.620 All right. 1.00

00:03:38.120 Enough sellout shit. 1.00

00:03:38.920 That's that's it for that's it forever. 0.98

00:03:40.680 We will never be mentioning our Patreon or a merch again.

00:03:43.700 And the teasers will get even longer until morale improves. 0.99

00:03:46.660 People who don't listen to the Blink episode will never know about this shit. 0.98

00:03:52.220 Let's get into the book here. 0.98

00:03:53.920 I'm going to send the opening paragraph to you.

00:03:58.580 Okay.

00:03:59.040 He says, in September of 1983, an art dealer by the name of John Franco Becina approached

00:04:04.560 the J. Paul Getty Museum of California.

00:04:06.840 He had in his possession, he said, a marble statue dating from the 6th century BC.

00:04:11.480 It was what is known as a kuros, a sculpture of a nude male youth standing with his left

00:04:16.000 leg forward and his arms at his sides.

00:04:18.260 There are only around 200 kuroi in existence, and most have been recovered badly damaged

00:04:22.740 or in fragments from grave sites or archaeological digs.

00:04:25.660 But this one was almost perfectly preserved.

00:04:28.040 It stood close to 7 feet tall, it had a kind of light-colored glow that set it apart from

00:04:32.120 other ancient works.

00:04:33.480 It was an extraordinary find, but Sheena's asking price was just under $10 million.

00:04:37.840 We did spoil this one in the Outliers episode.

00:04:40.040 Yeah. And look, this is now a very famous anecdote because of this book.

00:04:44.800 Oh, yeah.

00:04:45.040 So the museum conducts a 14-month investigation into the statue's provenance, and everything

00:04:50.820 looks great. The ownership records are consistent. A geologist says that the rock looks right.

00:04:56.600 The style looks accurate to the period. But Gladwell goes on.

00:05:02.020 The Kuros, however, had a problem. It didn't look right. The first to point this out was

00:05:05.960 an Italian art historian named Federico Zeri. When Zeri was taken down to the museum's restoration

00:05:10.740 studio to see the Kuros in December of 1983, he found himself staring at the sculpture's fingernails.

00:05:16.340 In a way he couldn't immediately articulate, they seemed wrong to him. Evelyn Harris was next. She

00:05:21.020 was one of the world's foremost experts on Greek sculpture, and she was in Los Angeles visiting

00:05:25.080 the Getty just before the museum finalized its deal with Pacina. Arthur Houghton, who was then

00:05:29.240 the curator, took us down to see it, Harrison remembers. He just swished a cloth off the top

00:05:33.640 of it and said, well, it isn't ours yet, but it will be in a couple weeks. And I said, I'm sorry

00:05:37.640 to hear that. What did Harrison see? She didn't know. In that very first moment when Houghton

00:05:42.100 swished off the cloth, all Harrison had was a hunch, an instinctive sense that something was

00:05:46.940 amiss. Her first impression, Peter, her blink was correct. For Freakonomics, I did Freakonomics.

00:05:54.080 And for this one, it's just blink. You got blinked. Every title we just do a stupid voice for

00:06:01.140 to make fun of it. Several other experts have the same immediate reaction. The museum gets 0.99

00:06:06.700 concerned. They convene a symposium on the issue, and then everything starts to fall apart. Not only

00:06:12.880 are the experts skeptical, eventually some of the documentation used to verify the past ownership

00:06:18.080 proves fraudulent. Other experts explain how certain forgery techniques could have fooled

00:06:22.720 the geologist. It turns out the statue very likely a fake, and the initial instinct was proven

00:06:29.780 correct right gladwell says quote in the first two seconds of looking in a single glance they

00:06:35.560 were able to understand more about the essence of the statue than the team at the getty was able to

00:06:39.400 understand after 14 months blink is a book about those first two seconds i mean we talked about

00:06:43.960 something outliers episode that like these are people with specialized expertise this is actually

00:06:48.240 part of gladwell's thesis okay one intuition is a very powerful thing two it's something that is

00:06:53.700 most effective when someone has an expertise in the subject okay and then three there's sort of

00:06:58.220 this ancillary idea that even though their intuitions are telling them something, these

00:07:03.960 experts can't quite like articulate the source of it. For all of the faults of this book, and there

00:07:08.460 are quite a few, all of these basic ideas have a lot of truth to them. Also because my brain is

00:07:13.560 also cooked from hosting the show. So much like discourse now is about like egghead elites and how

00:07:19.140 we should disregard anything that quote unquote experts say. A book that is like, hey, experts are

00:07:23.920 right about something, even if they can't describe it all that well, it's like kind of fine overall.

00:07:28.220 right right like yeah maybe maybe listen to somebody who knows more about greek sculptures

00:07:31.580 than you i actually was kind of surprised by this because i think a lot of the like

00:07:35.120 popular understanding of the book is just like intuition is magic yeah right not at all what

00:07:40.360 gladwell is saying and in fact like there are various parts of the book where he basically

00:07:43.500 says that like untrained intuition is very dangerous and precarious right so the heart

00:07:49.580 of the book is the concept of thin slicing okay gladwell defines thin slicing as the ability of

00:07:56.180 are unconscious to find patterns in situations and behavior based on a very narrow slice of

00:08:01.720 experience. He's not making this up. This is a term coined a few decades back by a couple of

00:08:06.920 researchers, Nalini Ambady and Robert Rosenthal. They conducted a study, which Gladwell talks about,

00:08:13.900 where people were exposed to silent video clips of teachers teaching. The clips were very short,

00:08:20.440 two, five, and ten seconds. And then they were asked to rate the teachers across different

00:08:25.400 metrics based on what they saw. The researchers took those ratings and then compared them with

00:08:29.600 ratings that students gave those same teachers at the end of a whole semester. They found that

00:08:34.700 the ratings based on even the two second clips were strongly correlated with the ratings that

00:08:39.900 the students gave teachers after an entire semester. So people were basically observing

00:08:43.860 these very basic cues, body language, posture, et cetera, and coming away with the same impression

00:08:49.000 that someone did after like actually being taught by these teachers, right?

00:08:51.960 So from this one study, I can make broad sweeping conclusions.

00:08:55.660 Absolutely.

00:08:55.900 I'm thin slicing from the research of thin slicing.

00:08:58.060 I think you can say based on this is that like one, our brains are making determinations very quickly.

00:09:04.260 And then two, those determinations are lasting.

00:09:07.600 So in this particular case, there's a caveat here, which is that teacher ratings are subjective.

00:09:11.720 It's not like the statue where like it's either fake or it's not, right?

00:09:15.120 It's not like the unconscious mind here is identifying some objective truth in a matter of seconds.

00:09:19.960 Right, because you could also make the argument that the kids in those classes were also basing their impression on body language, these kind of superficial cues, but then they just kept those impressions throughout the course of the semester regardless of new information.

00:09:32.260 So that's sort of like the seminal thin slicing study.

00:09:36.820 He tells another story.

00:09:38.560 He says, imagine that I were to ask you to play a very simple gambling game.

00:09:42.440 In front of you, there are four decks of cards, two of them red and the other two blue.

00:09:46.580 Each card in those four decks either wins you a sum of money or costs you some money,

00:09:50.200 and your job is to turn over cards from any of the decks one at a time in such a way that maximizes your winnings.

00:09:55.400 What you don't know at the beginning, however, is that the red decks are a minefield.

00:09:59.040 The rewards are high, but when you lose on the red cards, you lose a lot.

00:10:02.740 Actually, you can win only by taking cards from the blue decks,

00:10:05.660 which offer a nice steady diet of $50 payouts and modest penalties.

00:10:09.580 The question is how long will it take you to figure this out?

00:10:11.940 A group of scientists at the University of Iowa did this experiment a few years ago,

00:10:15.160 and what they found is that after we've turned over about 50 cards,

00:10:18.180 most of us start to develop a hunch about what's going on.

00:10:20.920 We don't know why we prefer the blue decks,

00:10:22.820 but we're pretty sure at that point that they are a better bet.

00:10:25.300 After turning over about 80 cards,

00:10:26.980 most of us have figured out the game

00:10:28.240 and can explain exactly why the first two decks are such a bad idea.

00:10:31.580 Interesting.

00:10:32.080 And the twist here is that researchers then hooked participants up to some machinery

00:10:36.720 to measure their stress responses like sweaty palms

00:10:39.320 and found out that people started to have stress responses to the red decks

00:10:43.200 only 10 cards in.

00:10:45.160 well before they know something's wrong and even farther before they can articulate what's

00:10:52.760 happening. So it's like your body starts to notice and then your instincts start to notice

00:10:56.900 and then finally your brain notices and you're like, fuck those red cards. 1.00

00:11:00.060 Right. This isn't quite thin slicing in the same sense as I like originally understood it, 1.00

00:11:04.820 but it's a cool example of how your subconscious is pretty powerful and ahead of you at times,

00:11:11.240 Right. Before you know what's going on, some part of your brain is like building the case and you're actually acting differently before you know why.

00:11:19.900 Right. Although, I mean, racing through my brain is a million examples of when that is not the case. Right. Right.

00:11:24.300 I always think of those studies that find that conventionally attractive people are considered more honest.

00:11:29.060 That's not us like accurately assessing a situation.

00:11:32.860 That's us believing that we're accurately assessing a situation, but we're actually operating on something else.

00:11:36.580 I love how every you're going to think of like all these different examples of irrationality

00:11:41.640 and stuff. 0.98

00:11:42.240 And I guarantee you Gladwell somewhere in this fucking book has this example because 0.98

00:11:46.600 there are just so many anecdotes. 0.98

00:11:48.060 He has a whole chapter that he calls us the Warren Harding problem because Warren Harding

00:11:52.120 was sort of, you know, tall and handsome and people thought that made him presidential 1.00

00:11:55.480 and it turns out he's sort of a dipshit. 0.99

00:11:57.600 But that is like a good example of where this sort of irrationality bubbles up. 0.97

00:12:02.180 The next big anecdote is about a psychologist, John Gottman, who specializes in analyzing couples.

00:12:09.340 Oh, yeah.

00:12:09.500 He's at the University of Washington.

00:12:11.120 He's, like, kind of famous in Seattle.

00:12:12.400 He does this, like, marriage lab stuff.

00:12:14.020 Yeah.

00:12:14.220 Yeah. 0.55

00:12:14.760 You know about the marriage lab. 0.97

00:12:16.120 All right. 0.99

00:12:16.440 Perfect.

00:12:17.440 Gladwell tells a story where a young couple visits Gottman's lab, and he hooks them up to some equipment to monitor their movements.

00:12:24.120 They have a videotaped counseling session where they talk about little things like whether they want a dog for about 15 minutes.

00:12:31.400 Okay.

00:12:31.560 So this is what Gladwell says.

00:12:33.120 Gottman has developed a coding system that has 20 separate categories corresponding to every conceivable emotion that a married couple might express during a conversation.

00:12:41.120 Disgust, for example, is a one.

00:12:42.860 Contempt is a two.

00:12:43.980 Anger is seven.

00:12:45.140 Defensiveness is 10.

00:12:46.580 Whining is 11.

00:12:48.080 On and on.

00:12:48.960 So Gottman has this coding system.

00:12:50.720 He runs all the interactions through it.

00:12:52.800 And then this gets translated into a row of 1,800 numbers.

00:12:57.440 And then he has a system for analyzing those numbers.

00:13:00.620 And quote, on the basis of those calculations, Gottman has proven something remarkable.

00:13:05.460 If he analyzes an hour of a husband and wife talking, he can predict with 95% accuracy

00:13:10.140 whether that couple will stay married 15 years later. 0.52

00:13:12.620 If he washes a couple for 15 minutes, his success rate is around 90%. 0.64

00:13:17.240 This is very cool.

00:13:18.680 This is very interesting.

00:13:19.780 It's also definitely not thin slicing, right?

00:13:22.840 Thin slicing, like in the opening anecdote, is about these very fast, almost split second

00:13:27.140 reactions we have to certain things.

00:13:29.000 And then here Gottman is using a ton of prior experience and information to create complex heuristics for evaluating marriages in a short period of time.

00:13:38.760 And so I'm reading this and I'm like, well, this is just like a different thing.

00:13:42.600 Right. Because a thin slicing version of this would be like someone who doesn't know anything watches a couple and then they predict it.

00:13:49.180 If an expert watched a couple for 15 seconds and was able to be like, they're going to get divorced with 75% accuracy, that would be thin slicing in my mind.

00:13:59.820 But this is actually quite scientific.

00:14:00.980 You're actually like doing a discourse analysis of the way that they're interacting with each other.

00:14:04.200 Right.

00:14:04.460 And you're testing it over time and you're developing a model that presumably becomes more predictive as you refine it.

00:14:09.700 So I'm going to send you what Gladwell says about this.

00:14:13.000 It's funny.

00:14:13.680 This all comes back to before we were recording, I was telling you about I met a guy online and he gave me his phone number.

00:14:18.880 and I was putting it in my phone and it turned out he was already in my phone as Martin Bad

00:14:22.800 Vibes. Now I feel good about making that determination. That's how I know that your

00:14:27.080 marriage will not work. I was thin slicing. I was like, something's off, man. Using the

00:14:33.720 incredible power of my brain, I figured out that the giant spider tattoo on this man's neck

00:14:37.940 makes him unlikely to be a winning candidate. He says, thin slicing is part of what makes

00:14:46.500 the unconscious so dazzling. But it's also what we find most problematic about rapid cognition.

00:14:51.640 How is it possible to gather the necessary information for a sophisticated judgment in

00:14:55.280 such a short time? The answer is that when our unconscious engages in thin slicing, what we're

00:14:59.740 doing is an automated, accelerated, unconscious version of what Gottman does with his videotapes

00:15:04.800 and equations. Can a marriage really be understood in one sitting? Yes, it can. And so can lots of

00:15:10.000 other seemingly complex situations. What Gottman has done is to show us how. Oh yeah, he's basically

00:15:15.220 saying like isn't it the same thing when you're like martin has bad vibes and also when you spend

00:15:20.940 years refining a model to predict marriage dissolution i guess what gladwell is saying

00:15:25.920 is like maybe your brain can also condense a ton of information down right in ways that we can't

00:15:32.040 really see which i i guess is true but yeah like many gladwell anecdotes it sort of feels like he

00:15:37.920 was like i'm gonna put this anecdote here because i love it right and then later he's like and and

00:15:42.300 here's my reason at the very end yeah i'll add my reasoning afterwards yeah they're both good

00:15:46.820 stories only a few of the examples that he gives in this book are really thin slicing so before we

00:15:52.780 keep going i want to do a little dive into the idea of thin slicing and specifically into like

00:15:59.380 what i think is really the thesis of the book which is that intuition is very powerful especially

00:16:04.100 when combined with expertise oh okay so somewhat ironically some of the clearest work i found on

00:16:10.400 this was written by another airport book author, Daniel Kahneman, who's a psychologist who wrote

00:16:16.460 Thinking Fast and Slow. Future Ibaka episode. He wrote a paper in 2009 alongside another

00:16:23.420 researcher, Gary Klein. They basically conclude that it's true that experts can develop extremely

00:16:29.440 reliable intuition, even when handling very complex problems, but only under certain conditions.

00:16:36.380 First, the environment has to be what they call high validity.

00:16:39.940 In layman's terms, it means that there are reliable cues that an expert can depend on.

00:16:44.880 Right.

00:16:45.020 The statue is fake or the statue is not fake.

00:16:47.300 That's something that you can confirm.

00:16:48.400 And then second, there needs to be an adequate opportunity for the expert to actually learn

00:16:52.800 those cues, like practice, right?

00:16:55.600 You need your 10,000 hours, folks.

00:16:57.640 It's all one book.

00:16:58.460 To give some examples, a lot of the early research on expert intuition involved chess

00:17:03.120 masters.

00:17:04.180 Chess has both of these qualities.

00:17:05.340 it's very high validity. The player can see the entire board. The pieces can only move in very

00:17:10.340 specific ways. And also experts can play many, many games over their lifetimes, allowing them

00:17:15.880 to learn all of the relevant patterns. Right. And there's a clear outcome too, right? You win or

00:17:20.000 you lose. Yeah. That's right. A less obvious one is firefighters. Gladwell touches on this too.

00:17:25.880 And there's a good amount of research on it. There's research showing that experienced

00:17:30.160 Most firefighters make very effective, quick decisions about where fire will spread, whether

00:17:35.660 a structure might collapse, et cetera.

00:17:37.880 And part of that is because the environment is high validity.

00:17:40.260 There are reliable environmental cues for them to use, even though a lot of researchers

00:17:45.100 were sort of baffled about what exactly they were seeing and firefighters were not always

00:17:50.500 able to articulate it.

00:17:51.340 I think a counterexample would be something like, you know, you often hear cops say something

00:17:56.120 like, I know when someone's guilty.

00:17:58.260 But the problem is the way that you measure someone's guilt versus innocence, like you

00:18:02.300 can measure whether they went to jail.

00:18:03.760 That's like a that's like an objective metric.

00:18:05.780 But it might just be you're good at telling the kind of person who will go to jail later,

00:18:09.420 which is not the same as somebody who's guilty.

00:18:11.280 Yeah.

00:18:11.480 Cops don't have blink. 0.92

00:18:12.620 Cops have something else.

00:18:14.200 It's called bonk. 0.98

00:18:15.460 And it's where your brain barely works.

00:18:18.240 And there are actual examples of environments that are not high validity.

00:18:23.920 A good one is politics.

00:18:25.260 There was a researcher who tracked the long term forecasts of political and economics experts and found that they were no more accurate than like casual newspaper readers over a long period.

00:18:36.720 This is probably because politics and economics are low validity environments.

00:18:39.700 There are too many variables, which means there are very few reliable cues that experts can learn from.

00:18:45.700 There was a review from the early 90s showing that expertise was found, quote, in livestock judges, astronomers, test pilots, soil judges, chess masters, physicists, mathematicians, accountants, grain inspectors, photo interpreters, and insurance analysts.

00:19:03.060 In contrast, there is poor performance by experienced professionals in stockbrokers,

00:19:10.320 clinical psychologists, psychiatrists, college admissions officers, court judges,

00:19:15.900 personnel selectors, and intelligence analysts. 0.95

00:19:18.680 It is a little bleak that the blink professions are like livestock inspectors and then the bonk 0.99

00:19:24.560 professions are like who goes to college and who goes to jail. 0.99

00:19:28.020 You could think of more broadly applicable scenarios where you are a quote-unquote expert because you encounter these situations all of the time. 1.00

00:19:36.800 Like gay men on the internet. 1.00

00:19:37.880 I know the vibes. 1.00

00:19:39.320 I know about Martin.

00:19:40.400 To simplify all of this a bit, intuition stem from pattern recognition.

00:19:44.440 If you have the skill to recognize a pattern and the opportunity to practice identifying it, over time you can develop this sort of seemingly magical intuition.

00:19:54.960 You're blinking.

00:19:55.980 Yeah.

00:19:56.100 Bad intuitions can be the result of having inadequate information or inadequate practice,

00:20:02.420 right?

00:20:02.580 Yeah.

00:20:03.020 Which results in your brain drawing on unhelpful or irrelevant information.

00:20:09.380 So like a lot of the biases that we know are real, like anchoring biases, right?

00:20:13.520 The idea that like if I put a certain number into your brain and then ask you how much

00:20:18.520 a car costs, you're slightly more likely to move toward that number, right?

00:20:23.160 That bias is because you're not pulling on accurate information in your brain.

00:20:28.360 And so your brain is just sort of reaching around, right?

00:20:30.640 If I tried to do that same experiment with someone who is an expert in cars, it wouldn't

00:20:35.920 have any effect on them because they have accurate information to draw on, right?

00:20:38.980 This is a really interesting example of like how science is supposed to work, right?

00:20:43.600 Because the TED Talk airport book version of this idea is like, did you know that your

00:20:49.440 first impression is always correct?

00:20:50.980 But then once you get into the actual data, it's like, well, your first impression is correct under very specific circumstances.

00:20:57.940 So first of all, you have to be an expert.

00:20:59.860 Second of all, you have to be an expert in a domain where you can reliably measure over the course of your career whether or not your expertise is correct.

00:21:07.500 It's like there's like this broad sort of fun fact, like cocktail party fact.

00:21:12.080 But then actually the actual rule or the actual science tells us it's far more conditional.

00:21:18.100 Right.

00:21:18.240 It's like, well, sometimes.

00:21:19.060 Another early chapter is about the concept of priming, social priming or behavioral priming in this context.

00:21:26.300 A concept I'm actually kind of skeptical of.

00:21:28.380 Oh?

00:21:28.840 Yeah. 0.97

00:21:29.180 You just blinked the shit out of this right now.

00:21:33.060 Just because like you come across so many of these studies and some of the findings are like pretty implausible.

00:21:37.100 Let's talk about it.

00:21:38.140 All right.

00:21:38.460 So priming is the idea that you can influence behavior by exposing people to an idea or even just a word, right?

00:21:45.500 Gladwell talks about an experiment devised by a psychologist named John Barg.

00:21:50.480 I'm going to send you something.

00:21:51.660 He says, imagine that I'm a professor and I've asked you to come and see me in my office.

00:21:55.780 You walk down a long corridor, come through the doorway and sit down at a table.

00:21:59.500 In front of you is a sheet of paper with a list of five word sets.

00:22:02.900 I want you to make a grammatical four word sentence as quickly as possible out of each set.

00:22:07.600 It's called scrambled sentence test.

00:22:09.640 Ready?

00:22:10.300 I feel like you're going to make me do this right now and I don't want to do it.

00:22:12.400 I'm not going to make you do it.

00:22:13.420 I thought about it, and then I was like, that's too much of a dumbass podcasting gimmick, even for me.

00:22:19.380 I do feel like you'd make me do it, and I want to say that.

00:22:22.060 I would absolutely make you do it.

00:22:24.300 This is why I thought it was coming, because I would make you do it.

00:22:26.400 So I'm just going to read a bunch of, again, these are five-word sets, and the idea is that you lose one word and develop a sentence from the other four, right?

00:22:34.980 Okay.

00:22:35.200 Him was worried, she always, from our Florida orange temperature, ball the throw toss silently.

00:22:43.420 shoes give replace old the he observes occasionally people watches be will sweat lonely day sky the

00:22:50.940 seamless gray is should now withdraw forgetful we us bingo sing play let sunlight makes temperature 0.99

00:22:57.820 wrinkle raisins are you priming me to think of old people holy shit you got it dude are you got it 0.99

00:23:03.180 because it's like wrinkle florida wow you're fucking blinking right now boom got him you can't 1.00

00:23:09.080 blink me, son. All right. I just sent you that. This seems straightforward, right? Actually, 0.99

00:23:13.560 it wasn't. After you finished that test, believe it or not, you would have walked out of my office

00:23:17.540 and back down the hall more slowly than you walked in. With that test, I affected the way you

00:23:22.540 behaved. How? Well, look back at that list. Scattered throughout it are certain words such

00:23:27.380 as worried, Florida, old, lonely, gray, bingo, and wrinkle. You thought that I was just making

00:23:32.820 you take a language test. But in fact, what I was also doing is making the big computer in your

00:23:37.660 brain, your adaptive unconscious, think about the state of being old. It didn't inform the rest of

00:23:43.120 your brain about its sudden obsession, but it took all of this talk of old age so seriously

00:23:47.580 that by the time you finished and walked down the corridor, you acted old. You walked slowly.

00:23:53.000 This is why I'm a little skeptical. Like the concept of old people, like being introduced 1.00

00:23:59.620 that I don't know that that would make me walk slower. He references a bunch of comparable

00:24:04.120 experiments. There's one where priming people to think about aggression makes them more likely to

00:24:08.540 interrupt a conversation, for example. Similar to you, I was blinking my ass off reading this part.

00:24:15.040 Like, it just feels a little too stupid. People walk more slowly after hearing the word Florida. 0.99

00:24:19.940 Yeah, yeah, yeah. And lo and behold, the science here is very shaky. Now, I will say this is not

00:24:26.120 really a Gladwell error. From what I can tell, at the time Blink was published, this was not

00:24:31.760 particularly controversial. In Thinking Fast and Slow, the Daniel Kahneman book, that's 2011.

00:24:38.380 And in that book, he said that the evidence for this was such that, quote, disbelief is not an

00:24:44.240 option. Since then, social priming has gotten caught up in what's often called the replication

00:24:49.900 crisis. The replication crisis is referring to a situation in many scientific fields,

00:24:55.540 but particularly in psychology where there's been a widespread inability to replicate the

00:25:01.840 findings of prior research. In 2011, there was an experiment by a guy named Daryl Bem where he

00:25:07.640 brought some students into a room. There was two doors. Behind one was an erotic picture.

00:25:14.380 And he said that they were able with statistical significance to predict which door it was behind. 1.00

00:25:21.740 Fuck yes. 0.84

00:25:22.600 A lot of people saw this and said there are some results bad enough, weird enough that you should assume that there has been a methods error somewhere. 0.98

00:25:31.940 Yeah.

00:25:32.140 There's also the one about how female hurricanes, hurricanes with female names do more damage. 0.99

00:25:38.080 That was another famous one where people were like, no, what the fuck are we doing here? 1.00

00:25:42.780 I mean, look, when a psychologist is just like, yeah, I can see a pussy through a door. 1.00

00:25:47.160 What are you talking about? 1.00

00:25:48.100 this hurricane's named after my bitch ex-wife that's how you know it's bad you're like whoa 0.99

00:25:53.620 i don't think this is real i i never looked into the hurricane thing i always i always thought 0.98

00:25:59.260 there was some plausibility in that we were maybe just naming the bad ones female names but are they

00:26:05.600 alternating though do they alternate male i don't know okay so then it can't be up through the

00:26:09.420 alphabet by letter the first letter of the name and they alternate got it got it it's totally

00:26:13.180 So then in 2012, some researchers wanted to highlight the potential for errors in priming

00:26:19.340 studies, and they conducted a study intended to lead to a ridiculous result where they

00:26:23.880 found that people who listened to When I'm 64 by the Beatles became literally younger

00:26:29.420 than the control group.

00:26:30.640 What?

00:26:31.300 They basically just looked at the data in different ways, found a statistically significant

00:26:36.060 result by fluke, and they said, look, this approach is common in the field, and we just

00:26:41.120 used it intentionally to come up with a result where you can show like, oh, people who listen

00:26:46.840 to this group are just are going to be younger on average. Right, right. This causes people to

00:26:51.640 perk up. Some researchers try to replicate John Barg's original study, the original one where

00:26:56.140 people walk slower when they're being primed to think about age. They could not, except when the

00:27:01.720 people observing the experiment were told about the expected result, which of course suggests that

00:27:07.400 They are probably, without really realizing it, manipulating the data to produce the result.

00:27:11.600 On top of this, around the same time, a well-known social psychologist,

00:27:16.280 Dierdrich Stapel, was found to be faking data, including in studies involving priming.

00:27:22.920 From there, you had researchers undertaking larger-scale efforts to replicate social priming

00:27:28.420 studies and finding that very few of them actually did replicate.

00:27:32.100 To give you a sense of where the consensus is now, the way that one researcher put it

00:27:36.800 was a good example is that people can be primed with words like diet on a menu to make lighter

00:27:42.800 food choices, but only if those people are trying to eat lighter. It's not like this magic thing.

00:27:48.220 It's just sort of like serving as a reminder, right? The sort of real version of this that

00:27:52.180 I've heard is in the order of polling questions. Yep. Where if you ask people like, what do you

00:27:56.880 think about immigration? Do you think immigration is out of control? How do you feel about immigration?

00:27:59.920 And then you're like, what are your top 10 priorities for the United States? Like people 0.76

00:28:03.300 are more likely to put immigration at the top because you've like reminded them that that is

00:28:06.160 an issue right i keep thinking about walking slower after saying florida why am i i'm just 0.94

00:28:12.720 like i'm wearing like fucking bifocals going 15 miles an hour down the road slow down you're like 0.99

00:28:18.720 all of a sudden doing anal after it's pride month you're like i don't know why i just have the urge 0.99

00:28:22.600 to do this did you know that for two hours after watching fire island a straight man will walk like 0.97

00:28:31.680 a little fairy you become six percent more likely you go yes in your daily life girl 0.69

00:28:38.100 damn i've been primed that's priming baby there is another theme that gladwell taps into that's

00:28:48.980 basically the idea that people who are making these split-second decisions and determinations 0.97

00:28:53.560 are generally unable to articulate why they're making them so i'm gonna send you

00:29:00.400 a little bit. Not long ago, one of the world's top tennis coaches, a man named Vic Braden,

00:29:05.900 began to notice something strange whenever he watched a tennis match. In tennis, players are

00:29:10.480 given two chances to successfully hit a serve, and if they miss on their second chance, they're

00:29:14.380 said to double fault, and what Braden realized was that he always knew when a player was about

00:29:19.080 to double fault. A player would toss the ball up in the air and draw his racket back, and just as

00:29:23.880 he was about to make contact, Braden would blurt out, oh no, double fault, and sure enough, the

00:29:28.880 ball would go wide or long or it would hit the net. One year at the big professional tennis

00:29:32.800 tournament at Indian Wells near Braden's house in Southern California, he decided to keep track

00:29:37.300 and found he correctly predicted 16 out of 17 double faults in the matches he watched. Okay.

00:29:44.240 We don't have to dig into this particular example. This guy's claim has not been rigorously tested.

00:29:48.420 I'm also a little bit confused about why he can catch double faults, but not faults. Yeah. Since

00:29:52.800 they're basically the same thing, but it's not totally impossible. Maybe the something about

00:29:57.180 the player's body. Maybe the player realizes and he can see that. I don't know.

00:30:00.720 Flow confidence, something, yeah, sure.

00:30:01.760 But there's a degree to which we all know that our brain makes decisions without our conscious

00:30:06.780 input, right? If I throw a ball at your head, you're not thinking like, oh no, there's a ball

00:30:11.940 coming towards me. I should get out of the way. You just duck, right? If I asked you why you ducked

00:30:17.100 after I threw the ball, you would probably say I ducked because the ball was going to hit me,

00:30:21.260 right? But there's a real question of whether you know that or whether you're backfilling

00:30:25.980 a plausible explanation for what you did. Right. Right. So back in the seventies, there's a very

00:30:30.240 famous paper published by Richard Nesbitt and Timothy Wilson called telling more than we know

00:30:36.360 verbal reports on mental processes. They do a series of experiments where they ask the subjects

00:30:41.600 questions, but then manipulated some of their responses. So for example, they ran an experiment

00:30:47.560 where they asked consumers to rate several identical products based on quality. People

00:30:52.440 would choose one. And then when asked for their reasoning, the people would provide a reasoning

00:30:57.540 despite the fact that the products were identical. They consistently found that people were unable

00:31:02.500 to reliably identify the sources of their decision making.

00:31:06.020 Yeah, I think that's all the time. People are very bad narrators of like why they do the things that

00:31:10.000 they do.

00:31:10.480 Right. The thing about that experiment where they give the identical products is they found out

00:31:14.780 that people were choosing the right most product frequently when they were looking at it. And then

00:31:19.640 they would ask people are you just choosing that because that's on the right and they were like no

00:31:22.900 no no yeah of course they said well hey they don't know it but they're choosing it because it's the

00:31:26.860 right most and we don't know why but that's what they're doing and then some other folks postulated

00:31:31.160 actually what's probably happening is that they're going through these left to right

00:31:34.320 and because they're identical they're just choosing the most recent one that they looked at

00:31:40.100 yeah right because like they trust themselves more or whatever so it's not that it's to the

00:31:43.740 right it's that that's the it's that it's the most recent one that they looked at do you know

00:31:47.220 the thing about nuclear submarines they do on public polls? I don't think so. There's a thing

00:31:51.220 when you take online polls that pollsters use where people will just fill out the first bubble

00:31:55.660 of every question. And so now they include a question, are you licensed to pilot a nuclear

00:32:00.840 submarine? And the first answer is yes. And that way, if somebody just clicks the first answer for

00:32:05.660 every single one, it'll come up with that. And then they can throw out the rest of the test

00:32:08.420 because they're just clicking the first one. That's interesting. Yeah. So in 2005, there's

00:32:12.560 some good follow-up on this idea by some researchers who use the term choice blindness.

00:32:17.680 They showed people two pictures of women and asked them to choose the one that they felt was more

00:32:23.300 attractive. Because in 2005, even scientific research was doing hot or not. Most of society

00:32:29.260 was just hot or not in various forms. But the twist is they're both behind a door.

00:32:33.460 You can't tell how the women look. So you just have to pick one. So then what they would do is 1.00

00:32:37.160 show them the picture again and ask them to explain their choice, except they would show

00:32:41.200 them the wrong picture they would choose they would show them the one that they didn't choose

00:32:44.720 okay less than a quarter of people caught the error and the ones that did not provided an

00:32:49.780 explanation for the choice that they did not actually make so you show them the less hot one

00:32:53.960 and then you're like why do you think she's the hotter one and they're like well i just love her

00:32:58.140 hair like whatever right that's so this is something that gladwell pretty consistently

00:33:03.420 pokes at and that he's basically right about that there's this disconnect between our subconscious

00:33:09.860 decision-making process and our conscious mind. There's also evidence, which Gladwell talks about

00:33:14.820 a little bit, that actually trying to articulate the reasons for your decisions can make your

00:33:19.860 decision-making worse. This is called verbal overshadowing. Gladwell talks about Jonathan

00:33:25.080 Schooler, who publishes about this in 1990. He showed people video of a robbery, and he found

00:33:31.780 that they were less likely to pick the robber out of a lineup accurately if they were asked to

00:33:36.260 verbally describe him first there's another very famous experiment by schooler that gladwell

00:33:41.780 references i'm going to send you this by the way i know i'm sending you brutally long uh gladwell

00:33:47.120 excerpts and i promise you that each one of these is edited down i feel like you're including

00:33:51.480 excerpts with like as many proper names as possible to catch me in a mispronunciation

00:33:55.620 no this is just what he does he like he loves a character so like yeah he won't just be like

00:34:00.420 there's a study and do a dry rendition of the study he's like it was a windy day in chicago

00:34:05.300 All right, here you go.

00:34:08.480 He says,

00:34:09.240 Consumer Reports put together a panel of food experts

00:34:11.860 and had them rank 44 different brands of strawberry jam

00:34:15.220 from top to bottom according to very specific measures of texture and taste.

00:34:19.420 Wilson and Schuller took the first, 11th, 24th, 32nd, and 44th ranking jams

00:34:24.260 and gave them to a group of college students.

00:34:26.620 Their question was, how close would the students' rankings come to the experts?

00:34:30.320 The answer is pretty close.

00:34:31.820 The students put Knott's Berry Farm second and Alpha Beta first, reversing the order of the first two jams.

00:34:37.600 The experts and the students both agreed that Featherweight was number three.

00:34:40.940 Overall, the students' ratings correlated with the experts' ratings by 0.55, which is quite a high correlation.

00:34:46.260 What this says, in other words, is that our jam reactions are quite good.

00:34:49.720 But what would happen if I were to give you a questionnaire and ask you to enumerate your reasons for preferring one jam to another?

00:34:55.680 Disaster.

00:34:56.120 Wilson and Schooler had another group of students provide a written explanation for their rankings,

00:35:00.840 and they put Knott's Berry Farm, the best jam of all according to the experts, second to last,

00:35:05.820 and Sorrel Ridge, the expert's worst jam, third. The overall correlation was now down to 0.11,

00:35:11.740 which for all intents and purposes means that the students' evaluations had almost nothing at all

00:35:16.200 to do with the experts' evaluations. So basically all of the kids are testing the jam, tasting the

00:35:21.600 jam. But some kids just taste it and immediately are like one, two, three, four, five. And the

00:35:25.400 second group tastes the jam. And then they're like, okay, this one's like a little bit more

00:35:28.900 sweet. This one's like a little bit more sour. And the ones that are describing the jams end

00:35:32.720 up scrambling it compared to the expert assessments. That's right. So what is this

00:35:36.140 explanation of this effect? It's not entirely clear. We don't we don't know exactly why this

00:35:40.760 happens. I'm sort of prone intuitively, if I'm just blanking it to the idea that the verbalization

00:35:47.380 is sort of interfering with your memory yeah we don't really know why we do or feel things and so

00:35:52.960 being asked to describe it it's just like it's fucking you up it gets you thinking about these 0.52

00:35:56.880 different factors that you weren't thinking about before right and then you're trying to map your 0.66

00:36:02.580 instinct on to this framework and it just gets cluttered in your brain right it's like hearing

00:36:07.920 somebody try to describe like why a joke is funny right like ultimately you're talking about an

00:36:12.740 involuntary response you're talking about something totally subjective and by trying to

00:36:17.320 intellectualize it, you can sort of convince yourself of anything or you can talk yourself

00:36:21.700 out of your gut level response. I was a little bit skeptical of this stuff. Probably the most

00:36:27.020 established version of this phenomenon is in that facial recognition context, like the robbery

00:36:32.580 example. And after the replication crisis drama unfolded, they attempted a replication of that

00:36:39.420 study that actually did manage to replicate an effect. I think it was smaller than in the earlier

00:36:44.300 studies, however. So at least in certain circumstances, it does seem like this is a

00:36:48.120 real replicable phenomenon. Gladwell does dedicate a good amount of time to talking about

00:36:53.780 how our intuitions can fail us. I mentioned that, you know, Kahneman had said in situations where

00:37:00.460 you don't have experience, you're not, your intuitions are not necessarily going to be

00:37:05.020 reliable, right? And Gladwell basically arrives at the same place. He does not do this whole like

00:37:10.940 intuition is magic thing. And he talks about a bunch of areas where your intuitions might lead

00:37:16.200 you astray. Some very serious and dark, some not at all. One of his simplest explanations is the

00:37:22.860 Pepsi challenge. I know you remember the Pepsi challenge. Yeah. They blindfold you and they

00:37:27.700 give you Coke and Pepsi and you're like, which one tastes better? And allegedly like more people

00:37:32.320 liked Pepsi. So this is right. So they started doing this as a marketing ploy in the 80s and

00:37:39.200 then it continued into the 90s. Pepsi is getting people on the street, giving them a little sip of

00:37:43.500 Coke, little sip of Pepsi, and they are choosing Pepsi. Coke does internal testing, and they find

00:37:48.660 the same thing. People prefer Pepsi by like 15% margins. That's hella funny. Crazy margins. Coke

00:37:55.660 freaks out, right? Very famously, this leads to new Coke. They change the formula, make it a little

00:38:01.060 sweeter. Everyone gets mad. Grotesque humiliation for the company, right? What happened here? It

00:38:08.080 turns out the advantage that pepsi had only lasted a couple of sips right people's initial reaction

00:38:13.140 was to prefer it but for drinking entire cans or bottles or whatever the advantage dissipates yeah

00:38:18.860 again i thought this was interesting it's also like are we talking about intuition or is this

00:38:22.600 about like sugar also i've noticed it's like every single thing they give you at like larry's market

00:38:27.580 you get like a little thing of smoked salmon like a little tiny dollop of it and then you buy the

00:38:32.320 whole thing of smoked salmon it just isn't as good when you have more of it yeah it's like every 0.89

00:38:35.540 little fucking snack at trader joe's one bite and you're like nice yeah three bites you're like 0.99

00:38:40.140 jesus christ a whole bag of this shit it's like too much yeah i don't think that's blink yeah 1.00

00:38:45.220 that's not yeah okay that's like a cute little story about coke and pepsi uh blink i ate too 1.00

00:38:50.380 many s'mores what are you talking about god damn you gladwell so he points out how intuition can 0.98

00:38:59.180 result in stereotyping. His big example is the murder of Amadou Diallo in 1999, a black immigrant 0.98

00:39:08.180 murdered by NYPD. He was mistaken for a rape suspect or a robber. The officers used both

00:39:15.040 justifications. He's confronted in a hallway. Officers say he didn't comply. They confused his

00:39:20.260 wallet for a gun. He's shot many times. Gladwell basically posits a theory of the case where

00:39:26.100 the officers were not explicitly racist, probably, but made a series of misjudgments that may well

00:39:32.720 have been influenced by Diallo's race. The more problematic portion of this is just a complete

00:39:38.300 curveball. Gladwell starts talking about autism. His basic premise is that most people are able to

00:39:44.440 intuitively read people's faces and gestures very quickly, blink style, while people with autism

00:39:50.460 struggle with that. He calls it mind blindness because they are purportedly unable to read the

00:39:56.340 intentions of other people. Right. I don't think that's true anyway. It's not true. It's a grotesque

00:40:01.800 oversimplification. Is this in the last couple of chapters of the book, Peter? Is this where

00:40:05.120 he's putting it? We're in the last chapter and you know, like I was halfway through the book and I

00:40:10.680 was like envisioning the episode, you know, and I was like, yeah, you know, we'll probably talk

00:40:15.080 about some of the science that's outdated, some of the things that I think he oversimplifies,

00:40:19.480 But a lot of it's just going to be science anecdotes.

00:40:21.340 And then you hit the last chapter and it's Amadou Diallo.

00:40:24.580 Yeah. 0.97

00:40:24.900 And I was like, here we fucking go. 0.97

00:40:28.180 Now, I'm going to send you this. 0.92

00:40:30.080 In this context, he has just told a story about an autistic guy named Peter.

00:40:35.400 Almost changed the name just to avoid you taking a little stab.

00:40:39.140 I'm not doing any stabs.

00:40:40.400 I know.

00:40:41.000 You're too woke. 0.90

00:40:42.080 But our listeners are like an autistic guy named Peter, huh?

00:40:45.140 he says but i can't help wonder if under certain circumstances the rest of us could momentarily

00:40:53.420 think like peter as well what if it were possible for autism for mind blindness to be a temporary

00:40:58.940 condition instead of a chronic one could that explain why sometimes otherwise normal people

00:41:03.760 come to conclusions that are completely and catastrophically wrong well if they would ban

00:41:08.440 the vaccines peter we wouldn't have this problem so he then goes on to theorize loosely referring

00:41:14.280 to some science, that certain very high stress situations like confrontations involving firearms

00:41:20.560 cause us to lose our ability to read people's faces and gestures and movements, a phenomenon

00:41:26.180 that he calls temporary autism.

00:41:29.100 Oh, my God.

00:41:30.380 And then he says that an important element of police training is to, quote, avoid the

00:41:35.040 risk of temporary autism.

00:41:36.940 Dude, he made up a term, mind blindness.

00:41:39.200 Just say temporary mind blindness.

00:41:41.300 He created a fake concept, but then he used a real one in his terms.

00:41:44.280 use the fake one of course he throws in temporary autism at the end of the book he's like he's like

00:41:50.340 final chapter it's malcolm time baby let's go like he's like everyone stop reading it's just

00:41:56.820 you and me it's like he's got like this notepad of his most problematic ideas and he's just like

00:42:02.480 final chapter baby but also is he using this to kind of exonerate the guys that killed amadou

00:42:07.560 diallo that like well this is a normal human thing rather than just like rank racism which

00:42:11.680 i think is probably a much stronger factor i think he's sort of i mean i will say that he sort of

00:42:16.360 seems to be implying that like training can fix this and it's like all right yeah they should be

00:42:20.940 better trained sure but the idea that they were unable to like sense diallo's intentions

00:42:29.080 is built on an assumption that i don't think you can make yeah and we've seen a lot of like body

00:42:35.640 cam footage now where it's not so much that they couldn't read the intentions as they were looking

00:42:42.920 to pull that trigger right now do we know what happened in the diallo case no there's no body

00:42:47.060 cam footage there are no eyewitnesses besides the cops so we don't really know what happened there

00:42:52.800 but like a story that is told based on police testimony where it's just sort of like yeah

00:43:00.180 they were in this really high stress situation and those are really tough to function in and

00:43:05.320 like, yeah, they became autistic suddenly. Yeah. And like, they actually got autism for a little

00:43:11.140 bit. Also, it's like the perfect TED Talk brain thing where it's like, you might think this was

00:43:15.680 racist policing, but actually, there's like a more kind of, I guess, quote, unquote, interesting,

00:43:20.620 like scientific explanation of it. But it's sort of like, I think it was probably just the racism.

00:43:25.500 This is a department of like a long history of racist policing. Oh, you think I'm racist? Actually,

00:43:30.840 you are being ableist because for that one and a half seconds i was autistic i was actually 0.99

00:43:36.180 that's hmm ever think of that i was as bad at reading faces as koreans are at flying planes 0.84

00:43:42.220 malcolm gladwell's like yes we got it ship it oh god final chapter baby that's where i was

00:43:48.620 basically gonna end this episode and then then i encountered something even worse i want to know

00:43:54.440 what your rabbit hole this morning was yeah i i had a little bit about this and then i made a

00:44:01.280 quick discovery at like 9 p.m last night that sent me down a rabbit hole and i was hell yeah my wife

00:44:06.080 was like hey are you coming to bed at like at 12 30 last night and i was like i need a little more

00:44:10.600 time in the diallo chapter in the parts about being able to like read people's expressions and

00:44:16.900 such there's an interview with a guy named paul ekman ekman is a psychologist who claims to be

00:44:23.060 an expert in micro expressions, little expressions that cross our face. And he has mapped out

00:44:29.680 different types of expressions as he believes that they correlate to different emotions and

00:44:35.040 so forth. This feels like some TikTok body language expert shit. I'm going to send you 0.99

00:44:39.460 a very brief excerpt. Ekman recalled the first time he saw Bill Clinton during the 1992 Democratic

00:44:45.300 primaries. I was watching his facial expressions and I said to my wife, this is a guy who wants

00:44:50.260 to be caught with his hand in the cookie jar

00:44:51.980 and have us love him for it anyway.

00:44:54.500 I mean, that's kind of true.

00:44:55.720 Yeah, it is true,

00:44:56.520 except it's a pretty convenient little story to tell

00:44:59.560 after Bill Clinton got caught cheating.

00:45:02.160 Also, he had numerous sex scandals during that campaign.

00:45:05.100 It's not like the Lewinsky thing

00:45:07.500 was like totally impossible to predict.

00:45:09.260 This guy's face looks like he's gonna murder Vince Foster. 0.99

00:45:14.300 And endorse a sex pest in New York for some reason, 0.99

00:45:17.100 like 25 years later.

00:45:18.240 Ekman claims that he can use this approach

00:45:19.700 not just to see someone's emotions but to detect deception to see when someone is lying and he got

00:45:27.540 extra popular after the publication of blink the tv show lie to me revolves around a character that

00:45:35.740 can read facial expressions and so forth and it's based on ekman god he starts to sell training

00:45:41.260 modules primarily to cops and other government agencies god he lands a large contract with the

00:45:48.660 TSA. This ends up being TSA's SPOT program screening of passengers by observation techniques,

00:45:56.100 which launches in 2007. In 2010, the government accountability office looks into SPOT and they

00:46:02.940 put out a report where they conclude with this, quote, TSA deployed SPOT nationwide without first

00:46:08.820 validating the scientific basis for identifying suspicious passengers in an airport environment.

00:46:13.500 Yes. A scientific consensus does not exist on whether behavior detection principles can be

00:46:18.100 reliably used for counterterrorism purposes.

00:46:21.420 This results in congressional hearings.

00:46:23.400 Ekman himself testifies, and he claims that his methods can detect deception with 90%

00:46:29.060 accuracy.

00:46:29.920 Okay.

00:46:30.100 Maria Hartwig, a psychologist who specializes in deception detection, testifies that not

00:46:35.420 only is there no research indicating that you can detect lies through analyzing micro

00:46:39.960 expressions, there's no evidence that people consistently exhibit micro expressions at

00:46:44.160 all.

00:46:44.820 Perfection.

00:46:45.300 Ekman's 90% accuracy claim is not verified because he has never published a study testing the efficacy of his training.

00:46:53.260 When asked about that, he claims that he cannot publish it for national security purposes.

00:46:58.540 Hell yeah.

00:46:59.140 Because now that we're using it, terrorists could try to beat it.

00:47:02.620 That's actually how you know it's good.

00:47:04.300 Yeah.

00:47:04.560 Because there's no evidence for it.

00:47:05.800 In 2015, someone leaked to The Intercept, the criteria the TSA was using.

00:47:11.880 Yes.

00:47:12.280 It is a 92-point checklist where different behaviors get you either one, two, or three

00:47:19.120 points depending on how risky they are.

00:47:22.480 If you get to four points, you get additional screening.

00:47:25.880 Okay.

00:47:26.200 I'm sending it to you now.

00:47:26.840 Yeah.

00:47:27.060 Please, God.

00:47:27.660 Yes.

00:47:27.940 Yes.

00:47:28.380 Yes.

00:47:28.760 The very first one, which gets you one point out of four needed for additional screening,

00:47:33.400 is arriving late for your flight.

00:47:35.860 That doesn't mean you're lying.

00:47:37.520 As an early to the airport guy, I say, go get them, folks.

00:47:41.380 Lock them up.

00:47:42.700 Exaggerated yawning as the individual approaches the screening process.

00:47:47.100 How would you even know?

00:47:48.220 Also, people are like, oftentimes for flights, you're getting up at like three in the morning

00:47:51.320 to make your flight.

00:47:52.540 Excessive fidgeting, clock watching, head turning, shuffling feet, leg shaking.

00:47:58.480 Imagine looking at your watch at the airport.

00:48:01.620 My favorite one, Peter.

00:48:02.740 These are the ones that are one point each.

00:48:04.460 These aren't even like major stress factors.

00:48:06.300 Three points each, deception factors.

00:48:09.260 one of them is appears to be in disguise that's one of my favorite ones because again you need

00:48:16.020 four points for additional screening as long as everything else is smooth yeah if the tsa knows 0.98

00:48:21.260 that you're in a full-ass disguise you still can't get screened i got the big fucking fake 0.96

00:48:27.660 mustache and they're like god we want to screen him but he's not yawning it also has wearing 0.98

00:48:32.220 improper attire for location but if somebody has flown in from hawaii they're probably in like

00:48:37.380 shorts and flip-flops they might be in detroit the location's an airport i'm trying to think

00:48:42.040 of what would be improper location because you could be going anywhere yeah literally there's

00:48:46.700 no such thing as improper attire for an airport it depends on where you're going you're going

00:48:50.260 powerful grip of a bag cold penetrating stare i will i will never get through a security line

00:48:58.640 bag appears to be heavier than expected expected sure you haven't found the most overtly racist

00:49:05.720 one oh face pale from recent shaving of beard recently shaved beard is one of is one of the

00:49:11.580 indicators he's trying to pretend he's not muslim

00:49:15.500 nice try the tsa by the way revised its criteria around this time and then in 2017 the accountability

00:49:25.820 office put out a third report saying that there's no evidence for the revised criteria either

00:49:30.500 are they still using this do you know i think that the program still exists in some form but

00:49:35.640 it's like not in this form anymore that's that's my recollection okay by the way i was so close

00:49:40.380 to being like you know this book is pretty harmless at the end of the day but then it turns out that

00:49:44.560 it indirectly launched a completely fraudulent and possibly racist tsa screening program dude

00:49:50.760 malcolm gladwell is so powerful i know there had to be congressional hearings because in 2004 when

00:49:57.500 malcolm gladwell heard this guy say that he could like read people's faces he wasn't like i don't

00:50:02.220 think so buddy that's got to be gladwell thinking it's like a better story if that if this guy is

00:50:06.480 an expert rather than just like debunking his claims because debunking his claims doesn't 0.97

00:50:10.160 really fit into gladwell's book this is why it's so exhausting doing gladwell shit like i told you 0.59

00:50:16.160 there's like a hundred anecdotes in this book i didn't do a deep or even medium dive into most of 0.53

00:50:21.580 them because it's just so time consuming yeah but it's basically a guarantee that gladwell did not

00:50:28.200 look enough into yeah any of these it does make me question the double fault guy to be honest

00:50:34.120 before we wrap i want to talk about some of the bad reviews oh yeah because i actually thought

00:50:41.380 that a lot of the reviews were sloppy in the same exact way that they accuse gladwell of being

00:50:47.220 sloppy yeah the big one was richard posner in the new republic wrote a uh a review called blinkered 0.98

00:50:56.060 And Richard Posner is a judge, famous conservative libertarian guy, revolutionary and introducing economic analysis into judicial opinions in case you thought the law wasn't stupid enough. 0.98

00:51:11.440 These are all of my deception factors. 0.98

00:51:13.540 These are three points, libertarian, conservative, judge.

00:51:16.640 He passes the threshold.

00:51:17.700 That's more than four points. 1.00

00:51:18.220 If I see a deceptively heavy bag, this guy is fucked. 1.00

00:51:21.320 He's going in the cage. 0.99

00:51:22.640 he writes this scathing review he hits on some well let me let me send you some excerpts hold

00:51:30.480 on Posner says one of Gladwell's themes is that clear thinking can be overwhelmed by irrelevant

00:51:35.320 information but he revels in the irrelevant an anecdote about food tasters begins one bright

00:51:40.900 summer day I had lunch with two women who run a company in New Jersey called sensory spectrum

00:51:45.020 the weather the season and the state are all irrelevant and likewise that hospital chairman

00:51:49.560 and Brendan Riley is a tall man with a runner's slender build, or inside JFcom looks like a very

00:51:55.680 ordinary office building. The business of JFcom, however, is anything but ordinary.

00:52:00.720 These are typical examples of Gladwell's style, which is bland and padded with cliches.

00:52:05.240 Yeah, I mean, to be honest, I kind of, I also find descriptions like this annoying,

00:52:08.640 but they're fairly standard for this kind of nonfiction writing.

00:52:10.520 I'm not saying you have to like this style of writing. I agree that it's a little cliche,

00:52:15.980 But like these are just examples of Gladwell painting a picture for his audience, right?

00:52:20.360 Yeah.

00:52:20.740 Sorry that it's not written like a judicial opinion.

00:52:23.560 Like otherwise the book is just a series of descriptions of studies.

00:52:28.200 Right.

00:52:28.420 Exactly.

00:52:28.860 Yeah.

00:52:29.000 You have to have some of this.

00:52:30.540 And journals constantly do this.

00:52:31.780 They're like, we met in his office where he was wearing a blue polo shirt and his hair

00:52:35.080 in a ponytail.

00:52:35.540 Like this is just like part of nonfiction storytelling.

00:52:38.020 You make a scene.

00:52:38.800 Well, the color of her hair is not relevant to the final conclusion here.

00:52:43.940 You love that voice.

00:52:45.000 That's your Posner voice.

00:52:46.620 Yeah.

00:52:47.580 That was actually a recording of Richard Posner that I played as a trick.

00:52:51.480 Gladwell talks at one point about an experiment by a law professor who sent a different mix

00:52:57.080 of white and black men and women to car dealerships to negotiate deals for cars.

00:53:02.000 And he found that black people, especially black men, ended up being quoted higher prices 0.77

00:53:06.180 even after negotiation.

00:53:07.600 He says, look, this probably isn't conscious racism because why would the salesman purposefully

00:53:13.140 price something inefficiently. That's Gladwell's thesis, right? And so guess what Posner's

00:53:19.040 objection to this is? It's not exonerative enough. He says, it would not occur to Gladwell,

00:53:24.040 a good liberal, that an auto salesman's discriminating on the basis of race or sex

00:53:27.860 might be a rational form of the rapid cognition that he admires. If two groups happen to differ

00:53:33.260 on average, even though there is considerable overlap between the groups, it may be sensible

00:53:37.660 to ascribe the group's average characteristics to each member of the group, even though one knows

00:53:43.120 that many members deviate from the average. So Gladwell is being like low key racist. And he's

00:53:47.460 like, no, no, it should have been high key racist. Gladwell is like, look, these guys are

00:53:51.840 accidentally doing racism, which definitely does not give enough credit to how racism operates in

00:53:58.640 reality. Yeah. And then you get the objection from Posner, which is like, well, maybe black

00:54:02.880 people are just worse at negotiating. Yeah. Like they were racist and that's chill. I love this 0.99

00:54:07.560 mostly because to go through this entire book and pick this out as like a specific thing that

00:54:13.820 irritates you to the point where you dedicate like three paragraphs to it it's the most libertarian

00:54:18.320 judge thing you could possibly do remember when you were talking about um people thinking that

00:54:23.560 like uh tall and handsome people are more honest and i told you that gladwell sort of pokes at that

00:54:28.300 too uh here's richard posner addressing that the average male ceo of a fortune 500 company is

00:54:33.800 significantly taller than the average American male, and Gladwell offers this as yet another

00:54:38.100 example of stereotypical thinking. That is not very plausible. A CEO is selected only after a

00:54:43.300 careful search to determine the candidate's individual characteristics. Gladwell ignores

00:54:47.840 the possibility that tall men are disproportionately selected for leadership positions

00:54:51.580 because of personality characteristics that are correlated with height, notably self-confidence,

00:54:56.600 and a sense of superiority perhaps derived from experiences in childhood when tall boys lord it

00:55:01.420 over short ones.

00:55:02.580 Has he considered that tall people are better?

00:55:04.440 The idea that decision makers at large companies might be subject to biases rather than like

00:55:09.320 perfectly efficient market machines is offensive to Poster.

00:55:12.960 He's like, how dare you question the wisdom of Fortune 500 companies?

00:55:17.060 This is such good like insight into the conservative brain because someone is just like, people

00:55:21.880 have irrational biases.

00:55:23.340 And then he's like, maybe there's a slightly more rational explanation.

00:55:27.000 like maybe maybe society is ordered exactly as it should be do not question it also it's like

00:55:33.600 why even do this like just admit that there's a little bit of bias in favor of tall people like

00:55:38.220 i'm you know me i'm a miniature gentleman i don't particularly care that much but it's like you have

00:55:42.380 to be able to at least admit like yeah discrimination exists also the i poked around at this a little

00:55:47.420 bit it seems pretty well established i have to say i didn't i didn't dig too deep because this 0.98

00:55:51.500 is just fucking richard posner who gives a shit but like there are various contexts in which tall 0.97

00:55:55.340 people seem to have an irrational advantage yeah the idea they're just like cooler yeah 1.00

00:55:59.640 part of the reason i wanted to talk about the posner review is because i hate richard posner

00:56:04.920 i don't care how uh i don't care that he got more liberal in his old age um he wasted his entire

00:56:10.260 life so it doesn't matter to me in a vacuum it's good that you saw this sort of like reaction to 0.91

00:56:17.340 pop science right where people are like hey this stuff is sort of bullshit right but the fact that

00:56:23.220 the new republic is like well why don't we let a federal judge have at it you couldn't find a

00:56:27.660 psychologist right this is so bizarre and like speaks to how like even in our rejections of

00:56:36.240 anti-intellectualism we don't really know what intellectualism is it's also part of this like

00:56:41.400 this era of the new republic and sort of like liberal magazines writ large absolutely you might

00:56:46.020 think you're like liberal values are correct but have you heard from the worst person you've ever

00:56:50.780 met next time you theorize uh why don't you show a little respect for fortune 500 companies

00:56:56.460 i wanted to put this at the end because even if malcolm gladwell accidentally got thousands of

00:57:04.680 muslim presenting people detained i still feel a little bit angrier towards richard posner

00:57:10.440 the enemy of our enemy is not our friend we're a hater podcast that doesn't mean you can get

00:57:16.120 on our good side by hating on a book that we hate. Nice try, Richard. Still hate you, nerd.

00:57:21.720 He's in my phone as Richard Bad Vibes. 0.99

If Books Could Kill - July 17, 2025

Malcolm Gladwell's ＂Blink＂

Episode Stats

Transcript