Time zones

Why does the world keep clocks that deliberately disagree by whole hours?

There is an obvious way to keep time, and for most of history everyone used it: noon is when the sun is highest where you are. It is free, self-correcting, and needs no authority. It also means that every town keeps a slightly different clock, since the sun crosses their meridians at slightly different moments.

We abolished that. In its place we built a system in which most of a continent agrees to be wrong about the sun, in unison. What made the wrong answer better than the right one?

Ask first what a clock is for. Two things, and they are separable. One is describing the day — has the sun risen, is it time to eat. The other is coordinating with other people: meeting, trading, catching something that leaves. Local solar time serves the first perfectly and the second only within earshot.

For most of history that was enough, because coordination happened locally. What changes if you can move faster than news? Consider a railway. A train leaves one town and arrives in another, and a timetable must state both moments. If each town keeps its own solar time, whose clock is the timetable in? In Britain the answer was that Bristol ran about ten minutes behind London, which is small until two trains are approaching one another on a single track and the difference decides whether they are in the same place at the same moment. The railways solved it the only way it can be solved: the Great Western adopted London time in the 1840s, other companies followed, and "railway time" spread outward from the stations to the town clocks.

Notice what kind of good a shared clock is. Its value to you depends entirely on how many other people share it — one person keeping London time in Bristol has simply made themselves wrong. Two people have a working appointment. A whole railway network has a timetable. So the benefit rises with adoption while the cost, being a few minutes off the sun, stays flat and small. That asymmetry decides the whole question, and it means the tipping is fast once it starts: the American railroads imposed four standard zones on a single day in November 1883, before any law required it, and the United States only legislated the arrangement in 1918. Britain's legal standardisation came in 1880, decades after the practice.

So why hours? Why not simply have one clock for the world? Because the first purpose has not gone away — people still want noon to mean something about the sun. An offset in whole hours is the compromise: it keeps every place's clock recognisably solar within about half an hour of error, while making arithmetic between places trivial. Whole hours are a convention chosen for ease of conversion, not a physical fact, which is why the exceptions exist — India runs half an hour off, Nepal three-quarters.

Now push on the geometry. The Earth turns 15 degrees an hour, so a zone genuinely 15 degrees wide keeps everyone within about half an hour of the sun. Are zones that wide? Look at China: one official time, UTC plus eight, across a country spanning roughly sixty degrees of longitude, so that in the far west the sun peaks nearly three hours after the clock says noon and people keep an unofficial local time alongside the official one. The date line is likewise not a line but a series of detours between island states — Kiribati moved a stretch of it in 1995 so its territory would stop being split across two dates.

Why would anyone accept the sun being three hours out? Return to the asymmetry. The cost of a mismatched clock is paid whenever you deal with someone on the other side of a boundary; the cost of a misaligned sun is paid privately and can be absorbed by shifting your schedule. Splitting an administrative unit — a country, a province, a business day — across two zones is expensive in exactly the way the system exists to prevent. So boundaries follow political lines, and where a state prefers unity it simply widens its zone.

Three refinements connect the steps.

First, "solar noon" was never as tidy as the argument assumes. Because the Earth's orbit is elliptical and its axis tilted, true solar noon at a fixed spot drifts by roughly a quarter of an hour either side of the mean across the year — the equation of time. Mean solar time, the thing zones actually approximate, was already an averaged abstraction before anyone standardised anything.

Second, the offsets we live with are not the zone offsets alone. Daylight saving adds an hour on top for part of the year — the same trade made twice, accuracy against the sun given up for coordination between people.

Third, on the causal claim. Railways are the standard account of why standard time arrived when it did, and the timeline supports it, but they were not alone: the telegraph made simultaneity across distance perceptible in the first place, and observatories were already distributing time signals. The 1884 International Meridian Conference then supplied a common reference at Greenwich, which is a different act from creating the zones; countries adopted offsets from it one at a time, over decades, and some are still adjusting.

• Why daylight saving persists despite repeated evidence against its original justifications.
• How computers handle a timekeeping system whose rules governments change at short notice.

Every term the collection defines is gathered in the glossary.