Malcolm Gladwell

Malcolm Gladwell is a famous author and grifter.

Here's a couple of takes on two of Gladwell's books.

The Bomber Mafia

From Brian Castner's review^[1] of one of Gladwell's books, The Bomber Mafia:

My first encounter with him was Outliers, which in classic Gladwell fashion promises to explain sociological events with a surprising counter-intuitive twist. Why are rich New York corporate take-over lawyers Jewish? Why are 40% of professional hockey players born in January? (They’re not.) The book stuck with me because I had a young son obsessed with hockey; should he just “give up” because he wasn’t born in the right month?

Gladwell calls Outliers a how-to guide, but always dissatisfyingly so. I can’t change my son’s birthday. And even if you accept his case for why Jewish people from the Garment District born in the 1930s were destined to become highly successful attorneys, he never explains how the individuals themselves did it. Why one poor boy in the tenement and not his friend? Why one hockey player born in January and not another? One gets the sense that the answer may undermine Gladwell’s thesis and so is left out, or, more conspiratorially, is revealing of other Big Ideas that Gladwell has less interest in exposing, such as the false meritocracy.

I am not a sociologist or a sports psychologist, so I can’t tell you the failures in Gladwell’s arguments in Outliers. But as a former Air Force officer, I know a fair bit about the service’s history and culture, and so I was curious what would happen when he took on a subject I knew.

My conclusion is this: Gladwell is right about Air Force pilots being obsessives, but completely wrong about the object of their desire. Which is surprising, because if anyone should be able to understand amoral perfectionists, it’s a wanna-be Tech Bro like Gladwell.

Outliers

This is the If Books Could Kill^[2] description of the podcast episode named Malcolm Gladwell's "Outliers":

In "Outliers", Malcolm Gladwell posited that it takes 10,000 hours of practice to become an expert in something. Mike and Peter prove him wrong by mastering his dumb book over the span of about 50 minutes.

This audio snippet should start playing 25 minutes and 55 seconds in, just as the podcast hosts dig into the seventh chapter of the book. This chapter is not only wrong about the Korean language, but also basically condemns all Asians. Actually, the eight chapter is perhaps even more messed up from a humanistic, scientific, and racistic perspective... As noted, the following transcript quote is about the seventh chapter:

Peter: And I'm sort of like, where's he going? Right? Where's all the momentum here
going? And then you flip the page and you hit chapter seven. The ethnic theory of
plane crashes. Oh, no. Oh, no.

Michael: Oh, is this the fucking South Korean thing?

Peter: Oh, yeah.

Michael: You would literally crash a plane into a mountain because you didn't want to talk back to your superior or something.

Peter: That is correct. Okay. No, I want to say, before I got to this chapter, I wasn't sure
about doing Outliers for the show, I was like, '10,000 hour rule. Yeah, it's pop science
bullshit, but is there enough here?' And then I got to this and I was like, oh my God,
this episode is on, you know, we're fucking doing it.

The chapter tells the story of Korean Air, an airline that over a several decades span had significantly more plane crashes than other airlines. And as the chapter title suggests, glad. Well, thanks that this is because they are Korean specifically. He says that there are certain Korean cultural norms of communication that make crashes more likely, including more indirect manners of speaking and deference towards people in authority. Okay. Essentially, a well-functioning plane requires consistent and clear communication among the crew and with air traffic control. And according to Gladwell, these cultural norms get in the way. [...]

In an interview, Gladwell once said, The single most important variable in determining whether a plane crashes is not the plane, it's not the maintenance, it's not the weather. It's the culture the pilot comes from. Oh, yeah. He doesn't back that up with data.

Michael: I think it's also the weather and shit. I can't get a lot of other stuff.

Peter: But the thesis is so widespread that in 2013, when Asiana flight [crashed] at SFO, several prominent news organizations covered the question if [...] culture was a factor.

The Washington Post runs a piece with a headline, 'Lack of cockpit communication recalls 1999 Korean Airlines crash near London'. NBC ran a story titled 'Korean culture may offer clues in Asiana crash'. A reporter asked the head of the National Transportation Safety Board in a press conference whether culture might have been the cause of the crash. So you might notice that this is essentially an unfalsifiable thesis, right?

As Gladwell himself admits, airplane crashes are super rare in large part because they require the confluence of events. It's not like it happens in movies where like one engine engine blows up. You need bad equipment, bad weather, tired or inexperienced pilots, poor communication, all of that. So the idea that you can sort of yank a specific conclusion about the role of culture out of all this: on its face: suspicious. But it gets worse as you dig in.

One of the first Korean air crashes Gladwell cites is Flight 801 in 1978. That flight was shot down by a Russian fighter jet. Now, Gladwell even states this. He like acknowledges that very quickly. But then he still folds it into this broader thesis about Korean flight crews and like continues to use, to like, count it as one of the plane crashes that he's talking about when he when he cites the number of plane crashes.

Michael: Oh my God.

Peter: He then goes on to list several other Korean Air crashes throughout the 1980s trying to make the case that something weird is happening with Korean air. He does not give details about these flights. So I looked them all up. The very first one is a 1983 Flight Seven. That was also shot down by the Soviets. This time Gladwell doesn't even mention that, which is notable, especially because this was a very big event in the Cold War. [...] The plane had taken off in New York and stopped in Anchorage. So many Americans died. Reagan called it a crime against humanity; [...] huge deal. But Gladwell just lists it off as if it were another example of Korean flight incompetence.

Next on his list is what he describes only as a Boeing 707 that went down over the Andaman Sea in 1987. Do you want to guess how that plane went down?

Michael: Was it shot down by India?

Peter: No, you got it way wrong. North Korean terrorists planted a bomb on it. So all in all Gladwell identifies seven significant Korean air crashes between 1978 and 1997. And that is sort of the basis of his thesis here. But three of them were the results of terrorists or military attacks. Yeah. I mean, frankly, this should be enough to just write off this fucking chapter entirely. But I want to mention one more thing.

Gladwell uses the National Transportation Safety Board's crash report for Korean Air Flight 801 throughout the chapter and misquotes it several times. And, as Korean speakers have pointed out, grossly mistranslates the Korean. The upshot is that Gladwell wants you to believe that the first officer sensed some danger due to the weather, but was so deferential to the captain that he didn't properly communicate it. In reality, Korean speakers who analyzed it said he very plainly told the captain that the weather was bad. I don't want to get into every little detail here, but a Korean-American blogger put it like this: 'This so-called interpretation of the pilot's true intentions is pure garbage. It is so ludicrously wrong that I cannot think of enough superlatives to describe how wrong this is. Gladwell's exposition on Korean language is completely, definitely, utterly, entirely 120% laughable to anyone who has spoken Korean in a professional setting.'

Michael: Wow. So what's the actual explanation for how the plane went down? It's just like straightforward pilot error. He had the information and he crashed anyway.

Peter: I think that the bottom line with that one, and I look into the other ones a bit, and it seems like if you look at the reports on the crashes, it was a combination of events, but circling around bad weather...and a couple of poor decisions. Right. And that's basically it. It's not much more complicated than that.

Michael: It's also a bigger problem, I think, existentially, too, because if it actually was the case that the co-pilot knew this thing and didn't say anything because of deferentialness, even that is not necessarily a cultural explanation, that could also just be that individual happens to be very deferential.

Peter: Absolutely. Like, let's put behind us the fact that this is unbelievably dishonest and
that he full-on lies about half of these crashes, or at least gets them grossly wrong.
He said 'the single most important variable in determining whether a plane crashes is
not the plane, it's not the maintenance, it's not the weather, it's the culture the pilot
comes from'. And he uses Korean culture as his case study, but all he has as evidence is a single flight transcript and a handful of plane crashes, again, half of which were either shot down or blown up. There's no analysis of data from other Korean airlines, no data about other airlines with higher crash rates than average, nor from what I can tell that he consult with a single Korean person. And finally, as Gladwell himself points out, reforms to internal training at Korean Air were able to completely remediate the Korean Air crash rate, which really undermines the idea that these ingrained cultural norms are playing a big role.

Michael: The interesting thing about this too is that it's in this one, it's not like he's necessarily taking a body of scientific work, there's a meta review of every plane crash or something. He's actually doing some sort of quasi-academic study himself. He himself is going through the flights, but what's weird is if it's a pop science book, there's dozens of scholars who research this, people have logs of every single plane that's crashed, all the reasons. There's probably actually some interesting stuff there that we don't know about, and you could actually do some sort of comparative study of, yeah, there's some cultures that are more collectivistic, there's others that are more individualistic, you could list those as a kind of, where are they on that spectrum? And then say, Germany is also very hierarchical. Like maybe Germany and South Korea do have like 0.2% more plane crashes than other countries, and that might actually be partly a little bit due to their more hierarchical cultures or something. But you could actually do this in a scientific way, and someone probably has, honestly, because it's actually kind of an interesting question, like does culture play into plane crashes? But instead of doing that or like engaging with the literature at all, he's just being like, well, I found six crashes.

Peter: You said maybe there's 0.2% more plane crashes. The problem with plane crash data is that planes almost never crash. Yeah, exactly, yeah. We're talking about crash rates that are something like four out of every million flights for Korean Air.

Michael: and that's extraordinarily high.

Peter: The sample sizes we're working with here are such that science almost can't be done. I also just realized that this book is called Outliers: the story of success. [This] chapter is just about three people being bad at flying planes. That's a better chapter title though. So, look, that was chapter seven. The ethnic theory of plane crashes. I figured we had hit rock bottom, and then I turned the page, chapter eight, rice patties and math tests.

Michael: Oh, we're back in Asia, aren't we?

Peter: We are spending an extraordinary amount of time in Asia.

Michael: It's only half the population of the planet, Peter. I don't see why so much generalizations wouldn't be justified.

Peter: Now, the thesis of chapter eight is that you can track the purported
success of Asians in mathematics to rice farming. He opens up talking about how
important rice is in China, and then he moves on to a very different topic, Asian
numbering systems.

Now, before we go on, I want to note that the fluidity with which Gladwell switches between talking about China specifically and southern China specifically, and Asia generally in this chapter is unreal. Now, it's very simple and somewhat compelling thesis here predicated, of course, on actual researchers' work, which is that Asian numbering systems use simpler rules and simpler words than their Western counterparts, which seems to lead to Asian children developing their counting skills much faster than Western children, giving them a huge developmental head start in mathematics. Of course, I'm saying Asian numbering systems, we're really talking about Chinese numbering systems. And then for reasons I genuinely cannot comprehend, he basically says, and also I think it might have something to do with rice farming. All of a sudden he's talking about how rice farming is very difficult and requires a lot of diligence and persistence, and those are all important in math, and rice farming was an integral part of society in southern China, and that helps explain why Asians are better at math.

Michael: I want to accuse you of oversimplifying this, because that sounds so unconvincing.

Peter: I mean, the weird thing is that he mentions the numbering system. I thought he was just going to go in a completely different direction, and that the rice patties thing was going to be some cute fucking anecdote that helps highlight why the numbering system works or something. But no, the only reason he mentions the numbering system, from what I can tell, is because his actual thesis is so off base, that he has to put in what seems like a much more likely explanation, so that if anyone ever says, what the fuck are you talking about with this rice patty bullshit, he can be like, well, it could also be the numbering system.

I want to point out that there is absolutely no research or data presented to try to specifically tie rice farmin to success in math.

One simple thing to do would be to look at people within Asia who farm rice, and people who farm something else, and prepare their math scores. He does not do that. In fact, and here's the weird thing, he barely mentions math scores at all. The fact that Asians are good at math is supposed to be something that you already believe. He mentions a single piece of research, the trends in nternational mathematics and science study. He notes that students from Japan, South Korea, Hong Kong, Singapore, and Taiwan all score high in math in that study. What you may notice is that China is not on that list, despite the fact that his entire spiel about rice farming is centered entirely around southern China.

Let me pull out a footnote here where he explains this. Mainland China isn't on the list because China doesn't yet take part in the Tim's study. But the fact that Taiwan and Hong Kong ranks so highly suggests that the mainland probably would would probably also do really well. What? So, okay. One quick thing to note is that both Hong Kong and Taiwan are tiny little rich islands. Another thing you can pull from this is that Singapore, which is one of those top performing countries in the study, is not a rice farming country at all.

On the other end, Thailand is a country with large rice production, and they were actually one of the poorest scoring countries in the study. Indonesia, the third largest producer of rice in the world, performed even worse. Gladwell addresses none of this, nor does he address how common is rice farming in Japan and South Korea. So, this chapter is basically a terrible educated guess about why Asians are so good at math, and it doesn't actually show in any meaningful way that Asians are so good at math. It's mind-boggling. This is the worst chapter in the book in terms of presenting evidence, and we're following the plane crash chapter. Asians can't fly planes, chapter, yeah. He doesn't explain why Korean math skills did not translate into the cockpit.

Michael: That's what I would have loved to hear him weave these together.

Peter: You might as well at this point, right? Just make it all one fucking thesis. Who cares?

Michael: Wait, what is his actual argument that it's the rice patties, though? It's like there's something specific about growing rice that you have to count the plants or something?

Peter: It's not counting per se. It's basically that it's very hard and requires a lot of persistence, and he sets some studies that show that some Asian students basically showed higher levels of persistence than Western students in math, and that a lot of being good at math is actually not that you're calculating more. It's that you're willing to spend longer on it, aka you're more persistent. That part seemed to be true, like that sort of little bit of research or whatever for what I could tell. But that's it. There really is nothing specific about rice farming. It's an argument you can make about just about any sort of peasant culture, right? Now, you know, rice farming is still happening, of course, but like 100 years ago, all of our ancestors were struggling a lot more, right? And if you felt like building a narrative around the struggles of like your immigrant ancestors in New York City in the 1880s or whatever, and saying that that helped sort of instill a work ethic into you or whatever, you could do it, right? You could do it with any fucking culture. Everyone has a history of sort of working hard and persevering of whatever. That's a large chunk of why we're all alive. So I truly don't understand why he latched onto rice patties. I really feel like he was just like, 'and here's the Asian section of my book. I'm going to explain what they're up to.'

Michael: What's the what's the first thing I know about Asia? Math and rice. I'm just going to Google those and figure it out. Yeah, because again, I can imagine this being like kind of an interesting theory if you actually did it academically. Like there probably are, yes, some crops that are much easier, some crops that take more perseverance. But surely, rice is not like the most difficult crop in the whole world. So then you would look to other places that like, you know, like maybe the far north or something where you basically can't grow anything. Life is really hard there. So like they should be super good at math, right? Right. But it's like he's not really doing any of the like basic comparative work that you would need to do to prove a cultural explanation.

Peter: Yeah, I mean, look, he himself says, well, like wheat farming, like rice farming is a southern China thing. Wheat farming is much more common in the north, but he's just like, yeah, but we don't have that data. And it's like, well, that's you're going to need that data, bro.
[...]

Michael: Come back to me when you have the data.

Peter: Right. And then I would like to see like maybe some controls for income, et cetera, you know, like you're right. You have really not done even the slightest amount of the legwork necessary to establish this.

Michael: Although he probably tells it in a really compelling way. Oh, yeah. This is also the Gladwell thing, that he's a he is like a really gifted storyteller.

Peter: No, he is. I mean, I was kind of struck here because sometimes he weaves the research and the story together in a way where your initial instinct is that makes sense. And then you think for a second and you're sometimes like maybe not, right? Right. But with this, it was really just two separate things. One, Asians good at math, two, rice farming difficult. Right. That's it. I like couldn't fucking believe this chapter. And if you're going to make a chapter built entirely around one of the mostcommon stereotypes in what you that is promulgated in Western society, maybe, maybe be careful. Right. I couldn't believe it. Even after the ethnic theory of plane crashes thing, I was like jolted by this chapter.

Michael: This is when I write a pop science book, one of the chapters is going to be called 'women be shopping.' Sorry, it's a science. I also think thematically, because I've read a million of these fucking pop science books for various projects. And one thing that I've found, and especially with this tranche of journalists that are really into counter-intuitive explanations, right, it's always about like, you think it's this, but it's actually this, right? The problem with those explanations is that they elevate cultural explanations for things that are really straightforwardly just policy outcomes. Yeah. Like I think if you want to understand why Taiwan and Hong Kong and Singapore do really good on math in these international math scores, it's probably related to their educational policies. It's probably related to something that is like really fucking obvious, like they're per capita GDP, and they're spending on education and like these specific structures of their specific education systems.