The width of a horse: how the railway gauge was chosen

About three-fifths of the world's railways are built to a gauge of four feet eight and a half inches — 1,435 millimetres, an oddly precise figure for something nobody sat down and decided. It was not calculated, and it was not the best available option; a wider gauge was demonstrably smoother and faster, and it lost anyway. The number came out of the colliery wagonways of north-east England, where carts had been pulled along wooden rails by horses for a century before anyone thought to put an engine on them, and it survived because the first successful railway engineer happened to have grown up beside one. Parliament made it official in 1846, by which point it had already quietly conquered Britain.

The horse in the story

There is a popular version of this history in which the gauge descends from Roman chariots, whose axles were set to the width of two horses, and whose ruts supposedly shaped every cart track in Europe thereafter. It is a lovely chain of causation and it does not survive contact with the evidence: wagonway gauges in the coalfields varied by a foot or more from valley to valley, which is not what an inherited Roman standard looks like.

The horse is still in the story, though, in a plainer way. On a colliery wagonway the horse walked between the rails, hauling a chaldron wagon behind it, so the track had to be wide enough for a working animal and its stride, and narrow enough that the wagons stayed cheap and the timber went further. Four feet and eight inches, near enough, is what that compromise looked like in County Durham. George Stephenson used it at Killingworth because it was what was already lying on the ground there, carried it to the Stockton and Darlington in 1825, and then, on the Liverpool and Manchester, added half an inch of slack so the wheels would run more freely through curves. That half-inch is the entire creative contribution of the modern era to the world's most-used measurement.

The better gauge, and what happened to it

Isambard Kingdom Brunel did what a good engineer is supposed to do. Building the Great Western Railway in the late 1830s, he reasoned from the physics rather than from the coalfields and chose a little over seven feet, which gave a lower centre of gravity, a steadier carriage, and more room beneath the floor for the machinery. By most accounts the broad gauge rode beautifully. Passengers noticed.

What Brunel could not out-engineer was arithmetic. Wherever the two systems met — most notoriously at Gloucester — every passenger changed trains and every ton of freight was manhandled from one wagon into another across the platform, which was slow, expensive, and visibly absurd. A Royal Commission looked into the matter in 1845, ran trials, and reported the obvious: by then, narrow-gauge track outnumbered broad by something like seven to one. The Gauge Act of 1846 required new passenger lines in Britain to be built to Stephenson's figure. It did not choose a standard so much as notarise one. The Great Western ran its broad gauge for another forty-six years and then converted the last stretch west of Exeter over a single weekend in May 1892, with thousands of men working down the line with crowbars, sleeping beside the track.

Ratified accidents, everywhere

Nearly every exception is another accident wearing a uniform. Ireland's five feet three inches was settled by officials averaging the incompatible gauges already in use — a committee arriving at a number by long division. Russia, Spain, and India each ended up elsewhere for reasons of their own, and every one of those decisions is now cast in a hundred thousand miles of ballast and quite impossible to revisit.

This is how most standards actually arrive: not chosen, but inherited, copied by people in a hurry for reasons that made local sense, and formally approved much later by a committee describing the world as it already was. The rails your train runs on tonight are the width of a Durham cart track, and there is no deeper reason, and nothing whatever to be done about it. If that is the last thought you have today, it is a good one to lie still with — the world is held together, largely and adequately, by things nobody decided. It is the same quality of thought we build our stories around, and the reason low-arousal audio works on a mind that won't stop looking for problems to solve: here is a real one, two thousand miles wide, already settled, not yours.