On Driving a Train Off the Rails in Japan


It was just after the morning rush on an April day in Amagasaki, a city of about half a million people close to Osaka. A JR commuter train was taking a bend on the track when it derailed, crashing into a 9-story apartment building.

There were 106 fatalities on the train, including the driver, and more than 500 injuries, not counting the psychological stress on residents of the building. A JR investigation determined the cause of the crash to be the driver’s poor handling of the turn, having taken it too fast to keep control of the train. But there was something else.

Lots of Japanese people dream of being train drivers. The esteem that trains hold in the popular culture here is unrivaled: You can get train-shaped lunchboxes, Kit-Kats in Shinkansen boxes, cutlery and posters about train stations. Books of trains are sold in the art section of bookstores. If you spend any time in Japan you’ll encounter literal trainspotters waiting on platforms with cameras to take pictures of specific cars and engines.

Trains run precisely, and drivers manage to line the doors of the train up with the spaces on the platform where people are told to stand in rows. People form up to 8 lines in pairs, as if waiting for a store to open, rather than milling about the station and then charging for the doors when they open (a familiar, unpleasant and completely senseless aspect of people’s behavior on the Boston subway system).

The precision of the trains and the shock of disruptions is no small matter. Here on Kyushu, we recently experienced a couple of simultaneous system-wide failures. You could tell because the bus stop was full of people, traffic was backed up, and the schools had to adjust their schedules.

You would think that kind attention to detail would make accidents on the JR lines almost impossible. Surely, on a system that runs so precisely, there must have been checks and balances on the driver that could prevent such a catastrophic human error?

A Bad Day
The driver had been late to his stops – only a matter of seconds – and then simply forgot to stop at a station. He eventually overshot the platform by 72 meters before stopping, and was forced into an unusual reversal maneuver. The driver failed at this, too, overshooting the platform again in the reverse. This process wasn’t catastrophic – only, now, a matter of minutes – but it was highly unusual and extremely embarrassing for the driver of a train to forget his stop.

En route to the next station, the driver radioed the conductor on board – the employee charged with tracking the time of trains – asking him to fudge the facts when he reported the delay. This was slightly out of line, because it was asking the conductor to lie to the conductor’s supervisor.

Before the conductor could reply, he was interrupted by an angry passenger demanding that the driver apologize for the delays. The conductor issued the apology himself to the passengers over the intercom, but in the process had ignored the driver’s request.

The conductor survived the crash, and later described to investigators that he was in an awkward situation: He had been asked to fudge a report, but didn’t answer the request, and was worried the driver thought he was angry about the request. He wasn’t – though it was an awkward request, it was also understandable. He hadn’t meant to ignore the driver, he had just been distracted by an angry passenger. In fact, the conductor had reported a slightly reduced delay time to his supervisor over the radio.

Meanwhile, frustrated about the loss of time and desperate to save face, the driver escalated the speed to 125 kph despite traveling in a 70 kph zone. Turns out he had been listening to the conductor’s radio traffic, and had turned to make a note of it so he could make sure the lie matched his own report. But while he was doing this at 125 mph, it seems he missed the indication that a turn was coming. He hit the bend, which the train couldn’t handle.

The idea of “human error” in this situation put the blame squarely on the driver, while ignoring the context that contributed to those errors. But it’s clear that the JR System – the checks and balances – contributed to an anxiety in the driver that interfered with his ability to make good decisions. Speeding up to 125 kph wasn’t the problem, not alone: He also had to make sure his paperwork was in order.

The final report concluded that,

The reason for the delay of using the brake was that the attention of the train driver was distracted in the following manner: he paid attention to the radio communication between the conductor and the transportation commander because the conductor had hung up the telephone while giving a false report to the transportation commander.

When the crash happened, the driver had been on the job for about a year, but had spent the previous four years in lesser roles with JR railways. His dream, like many JR employees, was to rise to the rank of Shinkansen driver. This means establishing a solid record of reliability and safety, which our driver hadn’t done. In his first year alone, he had been marked down several times for demerits in JR’s point-based system, running at a 10 for delays (the standard is 7.3). After a demerit, drivers would attend a training program to learn how to fix their mistakes.

The conductor was an older employee, and knew what would happen when that train got to the end of the line and the younger driver met with his supervisors. So, he agreed to minimize the incident because he wanted to minimize the driver’s humiliation. This, investigators learned, was common practice among non-management-level JR employees. They all dreaded the “training procedures” at JR.

One JR employee described his “training program” as follows:

“I always had to write essays . . . The writing started with a chronological description of my mistake and proceeded to a description of the cause of the mistake. After finishing one essay, a part of the first essay was chosen as the topic of the next essay. For example, if I wrote, ‘Human beings are creatures that tend to take the easiest route . . . however, we need to prioritize safety in every way if we take our duty seriously’ in the first essay, then the assigned topic for the second essay was, ‘Why are human beings creatures that tend to take the easiest route?’ Such revolving topics in essays continued endlessly. In my area of Amagasaki, we were assigned to fill up 80 per cent of a sheet of A4 report pad with such essay writing every hour.”

Lest this appear to be somewhat educational in nature, some employees described supervisors telling them they couldn’t use the bathroom when they were doing this. In other strange “education” procedures, drivers would be instructed to spend weeks of their shifts simply standing at the platform watching trains arrive and depart. This was not only tedious and boring, it was also humiliating, as the employee would be seen by other employees and even passengers. In some cases, employees were “trained” by pulling weeds on the station grounds or cleaning toilets in the offices where they worked. One employee was forced to do these tasks for five months after arriving to work three minutes late. What’s more, workers are often docked pay for these “education” sessions under the claim that they are being charged for special training.

With Unions Like These, Who Needs Managers
The JR Railways Union found that 37 percent of JR West employees had experienced some form of batshit crazy “re-education” after screwing up at work. The Union, however, is more of a JR West lobbyist – a typical arrangement in Japan – and so nothing was done. Perhaps unsurprisingly, about four JR West employees each year hang themselves or throw themselves in front of a train. But JR West is not alone in its educational practice. Other companies have been berated in the press for their “re-education” programs for handling employees who don’t toe the company line.

Avoiding training sessions seemed to become the top priority of JR West employees. The driver of the derailed train had already been in three such education sessions before his mistakes at the station; the crash, certainly, resulted from his panic at repeating them.

If you go to a Japanese high school, you might see kids in shorts running around a field in the wintertime. Likewise, there are “sports festivals” held in the peak of the summer heat, in which children compete to form human pyramids with tremendous risk of heat stroke and injury. It is teaching self-discipline and conviction of purpose. For many disciplinarian types in Japan, all problems stem from a lack of “gaman,” the Japanese concept of perseverance.

If you fail a test, it’s because you were too weak-willed to endure your studies. If you are bullied, it’s because you aren’t addressing the problem the bullies have found in you. In all complaints, someone can find a way to make the individual responsible. In many ways this is accurate, and there’s nothing wrong with self-determination.

But Japanese society is organized in such a manner that make it prone to abuses by well-meaning authoritarians. When you get the idea in your head that you must teach someone how to endure the unendurable, you actually start subjecting them to unendurable situations. Couple this with rigid hierarchies and lifetime employment that makes “take this job and shove it” responses more or less impossible – and you have an employment system that, gone bad, can seem more like the Stanford Prison Experiment than a rewarding career path.

I haven’t seen studies of the managers in the room for these sessions, but we do know that they typically involved several people surrounding the “student” employee. These people are of a higher rank than the employee being disciplined, and the “curriculum” is arbitrary – assigned by the whims of whichever manager is in charge of “training,” The arbitrary nature of the training presented, too, creates a fear that the situation could be more severe.

These managers would have viewed mistakes of the driver as crimes of inattention, and their employees as criminals in need of “education” on their failure to focus on details. The negative reinforcement, in the eyes of management, is the newspaper on the nose of a dog that pees inside the house. And so we have one manager writing in his own defense of the practice:

“A critic once mentioned, ‘Discipline in contemporary Japanese society does not work at all. Nobody penalizes children in families and in schools. Such a social tendency must lead Japan to corruption’. I agree with this view . . . In the end, moral education or discipline must be carried out within business enterprises. Unless brainwashed in terms of the discretion between what is good and bad and of strictly obeying rules in the world of railways, employees tend to violate rules.”

I wrote a post a few months ago about bullying, and how Japanese schools seem to set kids up for punishment when they fall out of line. It’s clear, to me, that something about this kind of “wrath of god” approach is distinctively Japanese. It erupts in children and manifests itself in adulthood, in classrooms and offices.

Clearly, there is some good to be shown from Japan’s rigid hierarchies. The country is respectful; people listen and respect authority – whether that authority derives from power, such as policemen, or from knowledge, as is the case of scientists or legal experts. That’s pretty accurately a meritocracy, where the most knowledgeable people on any given subject are the ones who decide what to do about problems in that particular sphere. Few people in Japan, for example, oppose climate change measures. The Japanese presume the people who study the issue know what they are talking about. In America, at least, this issue is still up to debate, regardless of what the world’s scientists have to say, because it doesn’t “feel right” or whatever. Americans are quick to believe that people studying climate change for 40 years are wrong because it’s really cold outside in May, so global warming must be a myth. 

But there’s a downside to Japan’s blind trust in authorities, of course. Part of it manifested in the massive breakdown of oversight at Fukushima Daiichi nuclear power plant, where three layers of bureaucracy were still not enough to force Tokyo Electric Power Co. to prepare for a disaster its own engineers had predicted. 

Likewise, the reliance on unquestioning authority creates power relationships in which the managers often have almost complete control of their underlings. If you are ordered to dance like a clown in a local parade, you obey. And if your boss doesn’t like you, you’ll be ordered to dance every year.

I mentioned the Stanford Prison Experiment earlier in this post: Where a handful of kids were randomly assigned to the role of prisoner or guard, and told to act these roles out over the course of a few days. Of course, the guards got vicious. But weirder is that the prisoners assumed that they really were worthless and powerless over their fates.

This seems to reflect quite accurately on the entire Japanese system. While many managers take the role seriously and with the responsibility required to keep their workers happy and motivated, others, just like anywhere else, are total assholes. The main difference in Japan is that the assholes can have complete control – and if you fail to put up with it, you’re letting everybody down.

You can follow This Japanese Life right into the depths of Facebook. And check out my book

This entry was posted in Thinking, Uncategorized and tagged , . Bookmark the permalink.

7 Responses to On Driving a Train Off the Rails in Japan

  1. Catspaw says:

    Desperate to save face, 106 dead, 500 injured, seems as if the Japanese would anticipate this failure from fear and plan to avoid such conditions. They don’t? Precision comes at a very high price. Is it worth the cost in lives lost?

  2. kamo says:

    “…people listen and respect authority – whether that authority derives from power, such as policemen, or from knowledge, as is the case of scientists or legal experts. That’s pretty accurately a meritocracy…”

    No. No it isn’t.

    I love your writing Eryk, and as ever so much of what you say here it bang on the money, but this statement is insidiously and dangerously wrong. Authority does not automatically equal merit. You know as well as I do the ‘the appeal to authority’ is one of the most lazily effective rhetorical fallacies around, and it’s especially bad in Japan where authority is disproportionately correlated with chronological seniority. Figures in authority tend to be those who have managed to hang around the longest without pissing people off, and I find very little meretricious about that. Japan is decidedly not a meritocracy, it’s a gerontocracy.

    • owwls says:

      The very next paragraph, and the entire point of the post, is in agreement with you. I meant to say only that Japan envisions itself, ideally, as a meritocracy, but that abuses like the ones the post is about prove that it isn’t. Should have been more careful about my wording!

      • kamo says:

        Yeah, sorry. Pet peeve tickled right there. As I said, you’re on the money both before and after, which I guess is why it jumped out all the harder.

        See also: senpai/kohai. Then maybe have a little cry.


    • Mari says:

      I just wanted to drop in to note that you have an inaccurate understanding of the ‘appeal to authority’ fallacy. An appeal to authority is invalid if the authority itself is irrelevant, or if the appeal is made on the premises that the authority itself is *necessarily* infallible. For example, an authority on the subject of organic reactions cannot be appealed to as an authority with respect to 19th century British literature. Cite: http://www.princeton.edu/~achaney/tmve/wiki100k/docs/Appeal_to_authority.html

      • kamo says:

        “…if the appeal is made on the premises that the authority itself is *necessarily* infallible.”

        This. This is exactly what I’m talking about. You could quibble by saying that in the Japanese context ‘infallible’ should really be replaced by ‘unquestionable’, but the latter is just the practical application of the former. I would dispute your claim that my understanding is ‘inaccurate’.

        BTW, I especially enjoy the irony of the Princeton link, sorry, citation. Far more authoritative than Wikipedia, don’t you find? ;)

  3. Laurel says:

    This is so interesting. I just moved to Tokyo, and this article sums up some general statements I’ve heard from various people I’ve met. Using the train systems as a case study paints a pretty fascinating cultural study.

Leave a Reply.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s