When I heard about the Asiana crash in San Francisco today I had three thoughts:
- As someone who flies more frequently than I like, and not long ago to and from San Francisco, my sympathy goes to the families of people who died in the crash and my very best wishes go to those who were injured.
- We flew from Seoul to Anchorage returning from China probably in the 1990's. I know we've flown that route on Korean and on Asiana, so I'm not sure which airline this was on. On our seats was an English language newspaper, maybe the International Herald Tribune with an article about Asiana's (or Korea's) airline travel partners dropping the airline because of safety issues. Not something that made for a relaxing flight.
- Malcolm Gladwell's Outliers has a chapter on how culture impacts cockpit communications, with a number of pages dedicated to Korean Airlines. That's what I want to focus on now. You can read the chapter online beginning here.
There are three preconditions that lead to plane crashes, he tells us:
- a minor technical problem
- bad weather
- tired pilot
Gladwell then focuses on pilot behavior citing two pyschological/anthropological models:
- mitigated speech
- "Hofstede's Dimensions"
". . . 'mitigated speech,' . . . refers to any attempt to downplay or sugarcoat the meaning of what is being said. We mitigate when we're being polite, or when we're ashamed or embarrassed, or when we're being deferential to authority. If you want your boss to do you a favor, you don't say, "I'll need this by Monday." You mitigate.
You say, "Don't bother, if it's too much trouble, but if you have a chance to look at this over the weekend, that would be wonderful." In a situation like that, mitigation is entirely appropriate. In other situations, however—like a cockpit on a stormy night—it's a problem.
The linguists Ute Fischer and Judith Orasanu once gave the following hypothetical scenario to a group of captains and first officers and asked them how they would respond:
You notice on the weather radar an area of heavy precipitation 25 miles ahead. [The pilot] is maintaining his present course at Mach .73, even though embedded thunder storms have been reported in your area and you encounter moderate turbulence. You want to ensure that your aircraft will not penetrate this area.
Question: what do you say to the pilot?
In Fischer's and Orasanu's minds, there were at least six ways to try to persuade the pilot to change course and avoid the bad weather, each with a different level of mitigation.
Command: "Turn thirty degrees right." That's the most direct and explicit way of making a point imaginable. It's zero mitigation.
Crew Obligation Statement: "I think we need to deviate right about now." Notice the use of "we" and the fact that the request is now much less specific. That's a little softer.
Crew Suggestion: "Let's go around the weather." Implicit in that statement is "we're in this together."
Query: "Which direction would you like to deviate?" That's even softer than a crew suggestion, because the speaker is conceding that he's not in charge.
Preference: "I think it would be wise to turn left or right."
Hint: "That return at twenty-five miles looks mean."
This is the most mitigated statement of all. Fischer and Orasanu found that captains overwhelmingly said they would issue a command in that situation: "Turn thirty degrees right." They were talking to a subordinate. They had no fear of being blunt. The first officers, on the other hand, were talking to their boss, and so they overwhelmingly chose the most mitigated alternative. They hinted.
It's hard to read Fischer and Orasanu's study and not be just a little bit alarmed, because a hint is the hardest kind of request to decode and the easiest to refuse. In the 1982 Air Florida crash outside Washington, DC, the first officer tried three times to tell the captain that the plane had a dangerous amount of ice on its wings. But listen to how he says it. It's all hints:
Look how the ice is just hanging on his, ah, back, back there, see that?
See all those icicles on the back there and everything?
Boy, this is a, this is a losing battle here on trying to de-ice those things, it [gives] you a false feeling of security, that's all that does.
Finally, as they get clearance for takeoff, the first officer upgrades two notches to a crew suggestion:
Let's check those [wing] tops again, since we've been setting here awhile.
I think we get to go here in a minute.
The last thing the first officer says to the captain, just before the plane plunges into the Potomac River, is not a hint, a suggestion, or a command. It's a simple statement of fact—and this time the captain agrees with him.
Larry, we're going down, Larry.
I know it.
Mitigation explains one of the great anomalies of plane crashes. In commercial airlines, captains and first officers split the flying duties equally. But historically, crashes have been far more likely to happen when the captain is in the "flying seat." At first that seems to make no sense, since the captain is almost always the pilot with the most experience. But think about the Air Florida crash. If the first officer had been the captain, would he have hinted three times? No, he would have commanded—and the plane wouldn't have crashed. Planes are safer when the least experienced pilot is flying, because it means the second pilot isn't going to be afraid to speak up." (pp. 60-61)
Hofstede's three dimension across cultures are:
- individualism - collectivism scale
- uncertainty avoidance
- power distance index
"Power distance is concerned with attitudes toward hierarchy, specifically with how much a particular culture values and respects authority. To measure it, Hofstede asked questions like "How frequently, in your experience, does the following problem occur: employees being afraid to express disagreement with their managers?" To what extent do the "less powerful members of organizations and institutions accept and expect that power is distributed unequally?" How much are older people respected and feared? Are power holders entitled to special privileges?
"In low-power distance index countries," Hofstede wrote in his classic text Culture's Consequences:
power is something of which power holders are almost ashamed and they will try to underplay. I once heard a Swedish (low PDI) university official state that in order to exercise power he tried not to look powerful. Leaders may enhance their informal status by renouncing formal symbols. In (low PDI) Austria, Prime Minister Bruno Kreisky was known to sometimes take the streetcar to work. In 1974, I actually saw the Dutch (low PDI) prime minister, Joop den Uyl, on vacation with his motor home at a camping site in Portugal. Such behavior of the powerful would be very unlikely in high-PDI Belgium or France." 25You can imagine the effect that Hofstede's findings had on people in the aviation industry. What was their great battle on mitigated speech and teamwork all about, after all? It was an attempt to reduce power distance in the cockpit. Hofstede's question about power distance."How frequently, in your experience, does the following problem occur: employees being afraid to express disagreement with their managers?".was the very question aviation experts were asking first officers in their dealings with captains. And Hofstede's work suggested something that had not occurred to anyone in the aviation world: that the task of convincing first officers to assert themselves was going to depend an awful lot on their culture's power distance rating.
The power distance in Korea is extremely high. More from Outliers:
The Korean linguist Ho-min Sohn writes:[UPDATE 9/29/13: Ask a Korean adds context to this which challenges Gladwell convincingly. He notes one co-pilot was older than the pilot and both had graduated from a much more prestigious institution, which counterbalances the status Gladwell mentions. He also translates the sections of the flight recorder that were in Korean and that would seem to contradict what Gladwell presents.]
At a dinner table, a lower-ranking person must wait until a higher-ranking person sits down and starts eating, while the reverse does not hold true; one does not smoke in the presence of a social superior; when drinking with a social superior, the subordinate hides his glass and turns away from the superior;… in greeting a social superior (though not an inferior) a Korean must bow; a Korean must rise when an obvious social superior appears on the scene, and he cannot pass in front of an obvious social superior. All social behavior and actions are conducted in the order of seniority or ranking; as the saying goes, chanmul to wi alay ka issta, there is order even to drinking cold water.So, when the first officer says, "Don't you think it rains more? In this area, here?" [from a pre crash cockpit tape] we know what he means by that:
Captain. You have committed us to visual approach, with no backup plan, and the weather outside is terrible. You think that we will break out of the clouds in time to see the runway. But what if we don't? It's pitch-black outside and pouring rain and the glide scope is down.
But he can't say that. He hints, and in his mind he's said as much as he can to a superior. The first officer will not mention the weather again.
It is just after that moment that the plane, briefly, breaks out of the clouds, and off in the distance the pilots see lights.
"Is it Guam?" the flight engineer asks. Then, after a pause, he says, "It's Guam, Guam."
The captain chuckles. "Good!"
But it isn't good. It's an illusion. They've come out of the clouds for a moment. But they are still twenty miles from the airport, and there is an enormous amount of bad weather still ahead of them. The flight engineer knows this, because it is his responsibility to track the weather, so now he decides to speak up.
"Captain, the weather radar has helped us a lot," he says.
The weather radar has helped us a lot? A second hint from the flight deck. What the engineer means is just what the first officer meant. This isn't a night where you can rely on just your eyes to land the plane. Look at what the weather radar is telling us: there's trouble ahead.
To Western ears, it seems strange that the flight engineer would bring up this subject just once. Western communication has what linguists call a "transmitter orientation", that is, it is considered the responsibility of the speaker to communicate ideas clearly and unambiguously. Even in the tragic case of the Air Florida crash, where the first officer never does more than hint about the danger posed by the ice, he still hints four times, phrasing his comments four different ways, in an attempt to make his meaning clear. He may have been constrained by the power distance between himself and the captain, but he was still operating within a Western cultural context, which holds that if there is confusion, it is the fault of the speaker.
But Korea, like many Asian countries, is receiver oriented. It is up to the listener to make sense of what is being said. In the engineer's mind, he has said a lot.
Korean Airlines got help.
In 2000, Korean Air finally acted, bringing in an outsider from Delta Air Lines, David Greenberg, to run their flight operations.
Greenberg's first step was something that would make no sense if you did not understand the true roots of Korean Air's problems. He evaluated the English language skills of all of the airline's flight crews. "Some of them were fine and some of them weren't," he remembers. "So we set up a program to assist and improve the proficiency of aviation English." His second step was to bring in a Western firm—a subsidiary of Boeing called Alteon—to take over the company's training and instruction programs. "Alteon conducted their training in English," Greenberg says. "They didn't speak Korean." Greenberg's rule was simple. The new language of Korean Air was English, and if you wanted to remain a pilot at the company, you had to be fluent in that language. "This was not a purge," he says. "Everyone had the same opportunity, and those who found the language issue challenging were allowed to go out and study on their own nickel. But language was the filter. I can't recall that anyone was fired for flying proficiency shortcomings."
Greenberg's rationale was that English was the language of the aviation world. When the pilots sat in the cockpit and worked their way through the written checklists that flight crews follow on every significant point of procedure, those checklists were in English. When they talked to Air Traffic Control anywhere in the world, those conversations would be in English. . .
Greenberg wanted to give his pilots an alternate identity. Their problem was that they were trapped in roles dictated by the heavy weight of their country's cultural legacy. They needed an opportunity to step outside those roles when they sat in the cockpit, and language was the key to that transformation. In English, they would be free of the sharply denned gradients of Korean hierarchy: formal deference, informal deference, blunt, familiar, intimate, and plain. Instead, the pilots could participate in a culture and language with a very different legacy. (p. 68)
And the training eventually got Korean Airlines back into the top ranks of international air carriers. However, This was a while back and it wasn't at Asiana. I'm sure that anyone who works in Airline safety has read this chapter and is thinking about this. Or, it could be something totally different.
This also reminds us of the importance of anthropology. It helps us understand cross-cultural differences - differences that make people from one culture more useful in some situations and from other cultures in other situations. (A little power deference might be helpful in some parts of the US politics these days.)
You can read Gladwell's chapter "The Ethnic Theory of Plane Crashes" here.
[UPDATE Sept. 29: The Ask A Korean post gives context which refutes Gladwell's narative about the the Korean airline pilots. If you read this, I strongly recommend you read Ask A Korean's critique of Gladwell's chapter as well. He convincingly punctures much of what Gladwell says.
I've included these updates today because many people are reading this post and going on to the original Gladwell chapter, but only a few are following the Ask a Korean link which credibly challenges Gladwell. If you've gotten this far, you owe your world view a chance to see another view of this.]