The Rhythm of Seoul: A Reflection on Speed and Slowness — Epoche B2
The loop, and the speed it was built to deliver Seoul Subway Line 2 is a closed green loop of about 48.8 kilometres and forty-three stations, running round the city and back to where it started, and at rush hour a train arrives every two or three minutes. Living in Seoul, you stop noticing how fast everything is until something moves slowly. The internet is fast; a parcel ordered at eleven at night arrives before breakfast; and on Line 2 missing a train is not a thing that happens to you. For years I read all of this as evidence of something — of competence, of a country that decided to be efficient and then was. Palli-palli , hurry hurry, is said half as a joke and entirely as a description. The least romantic account of where that came from, and the most convincing, is the sociologist Kyung-Sup Chang's. His term is compressed modernity : a condition in which economic, political and social transformations that elsewhere unfolded across generations are packed into a few decades, so that institutions and habits belonging to quite different historical stages end up operating side by side in the same society. Korea did in roughly forty years what the countries it was measured against took a century and a half over. Line 2, built between 1980 and 1984, is that programme in physical form — a ring designed so that the distance between wanting to be somewhere and being there is as close to nothing as engineering can make it. Why a two-minute gap is a different kind of thing from a twenty-minute one The claim that frequency abolishes waiting is not obviously true, and it is worth doing properly, because the rest of this essay is about what people do with the time a system gives back to them. If trains ran at exactly regular intervals of length $h$, and passengers arrived at the platform without consulting a timetable — which is what people do when the service is frequent — then arrival times are spread evenly through the gap, and the average wait is simply $h/2$. Real services are not exactly regular, and the correction matters. Writing $E[h]$ for the mean headway and $\mathrm{Var}(h)$ for its variance, the expected wait for a randomly arriving passenger is $$E[W] = \frac{E[h]}{2}\left(1 + \frac{\mathrm{Var}(h)}{E[h]^{2}}\right).$$ The formula, standard in transport analysis since Osuna and Newell's work in 1972, says something initially odd: irregularity always makes the average wait longer, never shorter, and it does so because a randomly arriving passenger is more likely to land in a long gap than a short one — long gaps occupy more of the clock. A service alternating one-minute and five-minute gaps has a mean headway of three minutes but an average wait of about two and a half, not one and a half. Put Line 2's rush-hour numbers in. With $E[h]$ of about two and a half minutes and gaps held tightly enough that the variance term is small, the expected wait is a little over a minute and a quarter. That is why the psychology is what it is: the penalty for missing a train is smaller than the effort of noticing that you have missed it. The system has made the decision to hurry irrational, and it has done so by controlling variance as much as by running trains often. Ninety seconds over the Han Then, one Tuesday afternoon, I was on the loop between Hapjeong and Dangsan, where the line comes up out of the ground and crosses the Han. I have described this bridge as the old railway bridge and that is wrong, so let me correct it here rather than leave it standing: the original structure of 1983 was found to be defective, closed in the mid-1990s and rebuilt, and the crossing in use today dates from 1999. It is one of the newer things on the line, not one of the older ones. The daylight lasts about a minute and a half, which is a figure worth checking rather than asserting. The crossing with its approaches is of the order of 1.3 kilometres, and trains hold roughly 50 km/h over it, so the time above water is about $1.3/50$ of an hour, which is close to ninety seconds. Then the line goes under again. Everyone in the carriage was on a phone. At the far end, in the priority seats, four older passengers were not. One of them had turned in her seat to watch the river, which is such an ordinary thing to do that I could not remember the last time I had seen anyone do it. They boarded slowly. They found their seats without hurrying, and when the doors opened at Dangsan none of them stood up early to be nearer the exit, which is a manoeuvre I perform without thinking several times a day. The pass, and the law behind it What struck me was not that they were relaxed and I was not. It was the arithmetic, and the arithmetic starts with a legal provision most Seoul residents could not name. Passengers aged sixty-five and over ride the Seoul subway free. The arrangement dates from 1984 and its basis is the Welfare of Senior Citizens Act, which provides for the state and local authorities to grant persons of sixty-five and over free or reduced use of designated public facilities, with the urban railway among those specified in the subordinate legislation. The city argues about it every few years, and the argument has a fixed shape: the operator books every exempted journey as forgone revenue and points at its deficit; the central government replies that the exemption is a welfare measure adopted locally and that compensating it is not its affair; the eligibility age is proposed for raising and is not raised. Both sides quote the city's own passenger statistics, which are published openly, and neither side disputes the numbers — only what should follow from them. Nobody argues about it in the other direction. What the pass actually distributes is not only money. It is a reason to be unhurried. The arithmetic, done properly I have made that last claim loosely in the past and it needs tightening, because as usually stated it is false, and the false version obscures the interesting part. Transport economists price a journey n