You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

Why Do Europeans Cause So Many Train Wrecks? Or, the Idiocy of Malcolm Gladwell's Cultural Determinism

Justin Sullivan/Getty Images

In light of two recent, deadly train crashes in Spain and Switzerland, reporters and pundits around the world are all asking the obvious question: why do Europeans wreck so many trains? What is it about their culture that causes these terrible accidents? Is it something deeply ingrained in their European psyche, like monarchism or imperialism, maybe tracing back to the Crusades or the Roman Empire, that produces negligent train conductors?

Actually wait, no, nobody is asking that. But plenty of media outlets were asking the same idiotic questions about Korean culture after the Asiana Airlines crash in San Francisco on July 6. Most cited the opinion of non-Korean pop intellectual Malcolm Gladwell. In a chapter in his 2008 book Outliers called “The Ethnic Theory of Plane Crashes,” he posits that the historically poor safety record of a different carrier, Korean Air Lines, is the product of a uniquely Korean “cockpit culture” in which Confucian values of deference to authority creates communication barriers between pilots, first officers and engineers. According to Gladwell, Koreans are too preoccupied with decorum and respecting social hierarchies to speak directly to one another, even in life-or-death situations like landing a plane, and deadly crashes are the result.

READ: How Malcolm Gladwell went from armchair sociologist to self-help guru

Gladwell’s fuzzy understanding of Korean language and culture, cited uncritically by various news sites, was based on a lot of assumptions and not much evidence. One blog, “Ask a Korean!” tore Gladwell to shreds by showing that he not only ignored very obvious, non-cultural explanations for KAL plane crashes (such as KAL’s rapid and reckless expansion during South Korea’s Yushin era of state-led industrialization, or getting shot down or blown up by the Soviets and North Koreans), but that he also has no clue as to how Confucianism actually works (it is a system of multiple, overlapping hierarchies based on age, education, profession, etc.; in one crash he cited, the first officer outranked the captain. And they were speaking mostly in English anyway).

It may be equally presumptuous to blame such journalistic negligence on outright racism, something most journalists generally understand to be bad. But the Asiana coverage does demonstrate the media’s continued acceptance of cultural determinism as a valid framework for interpreting events. This tendency to look for causes to complex outcomes and chalk it all up to the black box of “culture,” is not merely prejudiced. It’s the mark of lazy analysis: an unwillingness to sift through the empirical evidence and evaluate all possible explanations, or even to understand the varied historical, political, religious and social forces that combine to create this evolving thing we call culture.

Yet cultural determinism continually gets trotted out by media to drum up false debates that shouldn’t exist at all. Another example is “Islamic exceptionalism,” the notion that the reason democracy has had such a hard time taking root in the Middle East is because there is something about Islam that makes Muslims culturally predisposed toward authoritarianism. Echoing earlier theories about East Asian authoritarianism before the fall of one-party regimes in South Korea, Taiwan and the Philippines, Islamic exceptionalism made the rounds most prominently post-9/11 and in the run-up to the Iraq war; it’s been on the wane since the various Arab uprisings but still manages to worm its way into mainstream discourse.

As a hypothesis, it is simplistic and correspondingly simple to disprove. Hesham Sallam, a colleague of mine from Georgetown and currently a fellow at Stanford’s Center on Democracy, Development, and the Rule of Law, ran a statistical test based on a prior study by Columbia’s Alfred Stepan and Graeme Robertson, in which he compiled a database of all countries in the world coded by various measures, including a democracy index and a variable for whether a country is majority Muslim or not. Sallam showed that if you do a simple correlation between Islam and democracy alone, their relation will indeed be negative. But if you start controlling for other factors (e.g. a country’s wealth, oil rents, ethnic fragmentation, colonial history, and region of the world) using multivariate regressions–a statistical tool by which one can measure the effect of multiple causal variables on an outcome–the statistical significance of Islam drops out. Statistically speaking, you can’t conclude the relationship between Islam and authoritarianism is due to anything other than random error. In layman’s terms, it’s a coincidence.

Last year, when I was teaching a seminar on Latin American politics, I told my students that back in the 1970s, some made the same argument about Catholicism: that it was incompatible with democracy because of the hierarchical nature of the church or some other stretch of logic. It was, of course, an easy and stupid way to explain the fact that at that time, most Latin American countries happened to be dictatorships. My students, most of whom were from Latin America and born after their countries’ transitions to democratic rule, couldn’t believe it. Perhaps the most hopeful side effect of the Arab uprisings is no longer being subjected to debates about whether any societies are “culturally predisposed” toward authoritarianism, as if dictatorships are somehow natural occurrences for certain people, rather than something imposed on them by force.

This is not to say that academics are never guilty of shoddy or prejudiced thinking. But it speaks to the difference between what you can get away with in journalism versus social science. If you want to say, as Gladwell did, that “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,” you can have it treated as a valid opinion on a news site. If you submit it to a peer-reviewed journal, your referees will say “OK, show us your evidence.” And if you show them a simple correlation or second-hand impressions of Confucianism, you’ll get laughed out of the room.

Of course not everyone is lucky enough to be culturally essentialized as much as Arabs or Asians. Currently police are still investigating the cause of a train derailment in Quebec that killed 47 people. In the meantime, we’re still waiting for Gladwell to blame it on the religious and linguistic heritage of French Canadians.