The slow clock of precession

Nearly everyone learns the trick. Find the seven bright stars called the Saptarshi (the Plough, the Big Dipper). Run your eye off the two at the bowl's lip and you land on the pole star. It works tonight and will work for your great-grandchildren.

It would not have worked five thousand years ago. The star nearest the pole then was Thuban, in the constellation Draco, closest around 2830 BCE. Today most city dwellers cannot find it at all.

Nothing happened to either star. What moved was the earth.

The earth spins, and bulges a little at the equator. The sun and moon pull on that bulge. A spinning object pulled sideways does not topple; it swings its axis round in a slow cone, the way a toy top wobbles as it winds down. So the point the axis aims at walks a big circle in the sky, passing near one bright star after another and staying near none for long.

The rate is well known: about 50.29 arcseconds a year, an arcsecond being one 3,600th of a degree. A full circle holds 1,296,000 arcseconds. Divide, and one lap takes about 25,772 years. Across a human life the pole shifts less than half a degree, narrower than the full moon, which is why nobody notices.

Hipparchus caught it in the second century BCE, and the credit is deserved for an unglamorous reason: he took old records seriously. A man called Timocharis had written down Spica's position 154 years earlier. Hipparchus measured it again and found it had shifted two degrees. Two degrees is 7,200 arcseconds, and 7,200 divided by 154 gives about 46.8 arcseconds a year. That is seven per cent below the modern figure, reached with instruments sighted by eye. What the discovery needed was not a better telescope but somebody writing a number down and somebody else keeping the paper.

Indian astronomy carried the same drift under the name ayanachalana. An old Sanskrit handbook, the Surya Siddhanta, describes it as a swing out and back rather than a full circuit, though scholars argue over how to read those texts.