In 499 CE, a 23-Year-Old Said the Earth Spins

Somewhere in northern India, a young man finished a book and did something almost nobody in the ancient world bothered to do. He told you how old he was.

“When three yugapadas and sixty times sixty years had elapsed,” he wrote, “then twenty-three years of my life had passed.”

Run the arithmetic on his own calendar and the year falls out: 499 CE. The author is Aryabhata. He was twenty-three, and he had just written the book that would carry Indian astronomy for the next thousand years.

It is called the Aryabhatiya, and it is tiny — 121 verses, compressed into a Sanskrit so dense it reads like a zip file. Terse enough to memorise. Terse enough, it turns out, to argue with for centuries.

Because in the very first stanza of astronomical data, Aryabhata lists the revolutions of the heavenly bodies in a great cycle of time. The Sun: 4,320,000. The Moon: 57,753,336. And then, without ceremony, this:

“Of the Earth eastward, 1,582,237,500.”

Not the revolutions of the heavens. The revolutions of the Earth.

The man in the boat

That number is the number of times the Earth turns on its axis in a yuga. Divide the cycle out and it lands on one rotation per day, eastward — which is exactly, precisely right.

He knew what he was claiming, and he knew it sounded absurd. So he gave his readers an image instead of an argument:

“As a man in a boat going forward sees a stationary object moving backward, just so at Lanka a man sees the stationary asterisms moving backward.”

Lanka is his prime meridian on the equator. The asterisms are the stars. And the point is the one your gut still refuses at 30,000 feet: you cannot feel the thing you’re riding on. The riverbank slides past the boat. The stars slide past the Earth. Only one of those is actually moving, and it isn’t the scenery.

Everything the sky does every night — the wheeling, the rising, the setting — Aryabhata attributed to the ground under his feet. Elsewhere he states it flatly, in a line his critics would quote back at him for a thousand years: the Earth moves one minute in a prāṇa — one minute of arc in a breath.

There are 21,600 minutes of arc in a circle. A prāṇa is about four seconds. Multiply it out and you get 86,400 seconds. Twenty-four hours. He didn’t just assert the spin. He clocked it.

What he did not say

Here is where the internet usually oversells him, so let’s be exact.

Aryabhata was not a heliocentrist. He did not put the Sun in the middle. The Aryabhatiya runs a thoroughly geocentric system, with the Sun and Moon carried on epicycles around the Earth and the planets governed by two more — a slow manda epicycle and a fast śīghra one.

Some scholars, notably B. L. van der Waerden, have argued that the arithmetic of those epicycles conceals an older heliocentric model underneath. The historian of astronomy Noel Swerdlow demolished the idea in print, calling it a complete misunderstanding of Indian planetary theory and flatly contradicted by every word Aryabhata wrote. The consensus sits with Swerdlow.

So: no moving Earth around the Sun. A spinning Earth at the centre of everything. That is a smaller claim than Copernicus made — and, in 499, a far lonelier one.

Four decimals, and a word that refuses to sit still

The mathematical half of the book contains a verse that mathematicians still quote:

“Add 4 to 100, multiply by 8, and add 62,000. The result is approximately the circumference of a circle of which the diameter is 20,000.”

That is (100 + 4) × 8 + 62,000 = 62,832, over a diameter of 20,000. Which is π ≈ 3.1416.

The true value is 3.14159265. He is correct to four decimal places, in the fifth century, with no symbol for the ratio and no way to check.

Now the contested part. The Sanskrit word he uses for “approximately” is āsanna — “approached.” Some historians read enormous weight into that choice: he didn’t say this is the circumference, he said the value is approached, as though he understood no exact ratio exists. MacTutor states outright that Aryabhata realised π is irrational.

It’s a seductive reading, and it may be right. But it rests on one word, and Aryabhata never says it. He gives no argument for irrationality — a fact that would not be proved until Lambert, in 1761. Treat it as a tantalising possibility, not a discovery. The four decimals are astonishing enough without it.

(A very Aryabhata footnote: having computed the best π of the ancient world, he then mostly used √10 ≈ 3.162 in practice, because it was easier.)

The demon that wasn’t there

Every culture that watched the sky had a story for eclipses, and India’s was Rahu — the severed head of a demon, swallowing the Sun.

Aryabhata’s verse says something else entirely:

“The Moon obscures the Sun, and the great shadow of the Earth obscures the Moon.”

That’s it. That’s the modern answer, in 499 CE. A solar eclipse is the Moon in front of the Sun; a lunar eclipse is the Earth’s own shadow falling across the Moon. He backs it with geometry — a method for computing the length of the Earth’s shadow and the size of the shadow where the Moon crosses it.

Note what he does not do: he never picks a fight with Rahu. He doesn’t debunk the demon. He simply declines to need it, and computes the shadow instead.

His readers noticed anyway. Brahmagupta — the man who, a century later, would give the world the arithmetic of zero — attacked Aryabhata for exactly this, insisting the eclipse is caused by Rahu. The genius who made nothing into a number kept the demon.

The bowstring in your calculator

Aryabhata also left a table of twenty-four sines, given as differences: 225, 224, 222, 219, 215… down to 7. Add them and you get 3,438 — his radius, measured in minutes of arc, which is just 21,600 divided by 2π.

Reconstruct the table and check it against a modern calculator: across all twenty-four entries, his worst error is about 0.7 minutes of arc out of 3,438 — better than three parts in ten thousand. From a man with no decimals for the values and no trigonometry as we’d recognise it.

The Sanskrit for these half-chords was jya — literally “bowstring,” because a chord and its arc look like a strung bow. That word went travelling. Arabic astronomers took it over phonetically as jiba, written without vowels as jb. Latin translators, meeting jb on the page, read it as jaib — “fold,” “bosom” — and rendered it with the Latin word for a fold of cloth: sinus.

Which is why the button on your calculator says sin. It’s a mistranslation of a transliteration of the Sanskrit for a bowstring. (The chain from jya to sinus is solid; the “bosom” mix-up is the traditional account, and a few linguists grumble that it’s too neat.)

The size of the Earth — and an honest complication

You’ll frequently read that Aryabhata pinned the Earth’s circumference to within a fraction of a percent. It’s the one famous claim about him that doesn’t survive contact with the primary text.

What the Aryabhatiya actually says is that the Earth’s diameter is 1,050 yojanas — and that a yojana is 8,000 times the height of a man.

Everything depends on that man. Make him 1.6 metres and the Earth comes out 13,440 km across; make him a little taller and you get 15,360 km. The true figure is 12,742 km. So his Earth is somewhere between 5% and 20% too big, depending on whose height you use.

The much-repeated “24,835 miles, accurate to 0.3%” comes from a different pair of numbers — 4,967 yojanas, converted at five miles each — that don’t appear in the standard translation of the text and that imply a diameter contradicting the 1,050 the text gives. Somewhere in the retelling, a tidy number got loose and everybody copied it.

He had the Earth spherical, and the right order of magnitude, and a real method. That’s the honest version. It’s still remarkable. It just isn’t a rounding error.

The tradition that edited him

Now the strangest part of the story.

Aryabhata’s rotating Earth did not catch on. It was rejected — not by priests, but by other astronomers.

Brahmagupta’s objection is the one every schoolchild eventually raises: if the Earth is spinning, why don’t tall things fall down? It is, almost word for word, the argument Europe would throw at Copernicus a millennium later, and nobody could answer it properly until Galileo and Newton.

And the rejection went further than argument. Nine hundred years on, the great Kerala astronomer Parameśvara — heir to the tradition that would later produce infinite series two centuries before Newton — sat down with the verse that says of the Earth eastward and read it as of the asterisms. One syllable, changed. He then took the boat simile and reversed its meaning: the stars move west, and only a false impression makes the Earth seem to turn.

Clark, translating the text in 1930, called this what it was — a commentator determined to prove Aryabhata did not teach the rotation of the Earth, and the proof is Brahmagupta, who attacked the rotating-Earth line so specifically that we know exactly what the original said.

Aryabhata’s own successors couldn’t quite believe him. So they fixed him.

Coda: 19 April 1975

On 19 April 1975, a 360-kilogram, twenty-six-sided metal polyhedron went up from a Soviet launch site at Kapustin Yar. It was India’s first satellite — the eleventh nation to reach orbit. Its power system failed after four days.

ISRO named it Aryabhata.

There is something exactly right about that. Not a temple, not a statue: an object put into orbit around a turning Earth, by the country of the man who first said it turns — and who was, for fourteen centuries, mostly not believed.

The stars never did move. He knew that at twenty-three.