Why a minute has sixty seconds

Published on August 30, 2026

Tags:historytimekeepingunits

A stag stands at the foot of a desert staircase draped with melting pocket watches

Everything else we count, we count in tens. We have ten fingers, our money divides by ten, our scientific units multiply by ten. Then we get to time, and suddenly it is sixty — sixty seconds, sixty minutes, and then, inexplicably, twenty-four.

It is the last serious holdout of a number system nobody has used for its own sake in about four thousand years. Here is how it survived, and what happened when a government tried to kill it.

What Sumer actually did

The trail leads to Mesopotamia, but not in the way it is usually told.

The Sumerians did not have sixty-minute hours. According to NIST, they divided the day into 12 periods, each about two of our hours, and split each of those into 30 parts — a 360-part day, each part running about four of our minutes. Their arithmetic was sexagesimal — base 60 — and it was superb, but their day was not divided the way ours is.

Our stack of hours, minutes and seconds is a later fusion: Egyptian hours laid over Babylonian arithmetic, assembled by Greek astronomers. It is a hybrid, and it took centuries to assemble.

Nobody actually knows why sixty

The oldest surviving attempt at an answer comes from Theon of Alexandria in the fourth century CE — well over a thousand years after the fact. Sixty, he observed, is the smallest number divisible by 1, 2, 3, 4 and 5.

That is a genuinely good property. Sixty divides cleanly into halves, thirds, quarters, fifths, sixths, tenths, twelfths, fifteenths, twentieths and thirtieths. Ten manages halves and fifths and then gives up. If you are dividing land, debts or daylight without decimal notation, sixty is the friendlier number by a wide margin.

But it should be said plainly that this is a rationalisation offered long afterwards, not a recorded decision. Two popular explanations are worth treating carefully:

“360 degrees came from a 360-day year” is repeated everywhere and does not hold up well — the Sumerians knew perfectly well the year was longer than 360 days.

“They counted to sixty on their finger joints” — twelve joints on one hand, five fingers on the other — is a lovely story, and the scholarly source that presents it does so with an explicit shrug, ending roughly at anyone convinced? There is no textual or archaeological evidence for it.

The honest position is that several theories exist and none is established.

The words arrived a thousand years after the maths

Ptolemy, working in the second century, divided circles and hours into sixtieths and sixtieths of sixtieths. He did not call them anything in particular. They were simply the first small part and the second small part.

The names we use are Latin, from the twelfth-century translation movement around Gerard of Cremona and the Toledo school: pars minuta prima — the first small part — and pars minuta secunda, the second. English kept the adjective from one and the ordinal from the other, which is why “minute” is a diminutive and “second” is a number, and why they sound like they belong to different families. They do.

So the arithmetic existed for a millennium before the vocabulary. And the vocabulary existed for centuries before anything could actually measure a second.

The first machine that could show one

That machine appears in Kassel in the 1580s, built by Jost Bürgi for Landgrave Wilhelm IV.

On 14 April 1586, Wilhelm wrote to the astronomer Tycho Brahe describing it: a clock with a minute hand, registering seconds, with an error of less than a minute across a full day. That was a transformation. Before it, the second was a unit of calculation. After it, the second was something you could point at.

The astronomer Christoph Rothmann, trying to convey to a reader what one felt like, described a second as lasting about as long as the shortest note in a moderately slow song. It is a wonderful sentence, because it shows a man reaching for a comparison in a world that had never needed one.

Accuracy then moved quickly. Huygens’ pendulum clock of 1656 held under a minute a day; by the time the clockmaker George Graham was working in 1721, a good clock was within a second a day.

The seventeen-month republic of decimal time

Which brings us to the only serious attempt to abolish sixty.

Revolutionary France, having decimalised weights, measures and money with lasting success, turned to the clock. A decree of 5 October 1793 established decimal time, with an addendum that November fixing the structure: ten hours to the day, a hundred minutes to the hour, a hundred seconds to the minute. A decimal second worked out at 0.864 of an ordinary one. Noon became five o’clock. It became mandatory on 22 September 1794.

A French Revolutionary mantel clock in a glazed mahogany case, its enamel dial carrying two sets of numerals: Roman numerals I to XII around the outside and Arabic 1 to 10 within.
A decimal clock by Lepaute of Paris. The dial carries both systems at once — ten decimal hours inside, twelve conventional ones outside — because for a few years you needed to read both. Photo: Mikaël Restoux, Wikimedia Commons, CC BY-SA 3.0.

This was not a thought experiment. A decree of 9 February 1794 launched a national clockmaking competition, and decimal clocks, watches and sundials were manufactured — not many, but they were made. The watchmaker Robert Robin had already presented a decimal astronomical pendulum clock to the Convention the previous year.

It was suspended on 7 April 1795, having had roughly seventeen months of legal existence. The French horological literature gives the reason with some economy: it never entered daily habit.

Two details worth getting right, because both are usually garbled. Decimal time was suspended, not abolished — for an indefinite period. And Napoleon did not end it: he ended the republican calendar, effective 1 January 1806, which is a different thing and outlived the decimal clock by a decade. The ten-day week even flickered back briefly under the Paris Commune in 1871.

Why sixty won

We do not have to guess, because someone stood up and explained it at the time.

On 1 March 1795, weeks before the suspension, Prieur de la Côte-d’Or — the man steering the Republic’s metric reform — addressed the Convention on why decimal time should be abandoned. His reasons were entirely practical. It was damaging the rest of the metric reform by association. The old habits were held in place by what he called the immense force of habit. There was a fear of confusion, since everything would need new names. The cost of changing every clock would be enormous. And citizens and watchmakers alike, he said, would be infinitely dismayed.

Not one of those is an argument about arithmetic. They are arguments about the expense and disorder of replacing a working system — and they are the reasons given by the people who actually abandoned it, rather than a tidy theory assembled afterwards.

There is one further property of time that Prieur did not need to spell out: it is shared in a way that a metre stick is not. A ruler that disagrees with its neighbour is a nuisance. A clock that disagrees with everyone else’s is useless. Nobody can switch alone.

So we kept sixty, because a Sumerian scribe found it convenient for division, because Greek astronomers built it into their tables, and because every machine since has been made to match. It is one of the longest-running pieces of backwards compatibility in human history.

The one place it doesn’t apply

There is a quiet exception, and you are probably carrying it.

Below the second, we already went decimal and nobody objected. Tenths, hundredths, thousandths — no one has ever proposed sixtieths of a second, because by the time anyone could measure that finely, decimal notation had won everywhere else. The old base survives above the second and stops dead below it.

So a stopwatch reads in a hybrid: sixty-based above, ten-based below. When TiCaNo Stopwatch shows you 1:23.45, the colon is four thousand years old and the decimal point is modern, and the two systems sit on either side of it without either one noticing the other. Nobody designed that. It is just where two very old arguments happened to stop.

Sources

← Back to all posts