Aryabhata gives pi: the first explicit acknowledgement of an irrational constant

# Aryabhata gives pi: the first explicit acknowledgement of an irrational constant

In 499 CE, a young mathematician in Kusumapura — modern Patna — wrote down a verse that, on its surface, looks like an arithmetic recipe.

> 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. > > — *Aryabhatiya II.10*, trans. W. E. Clark (1930)

Run the arithmetic: (100 + 4) × 8 + 62,000 = 62,832. Divide by 20,000. You get **3.1416**.

That's pi. Four decimal places of accuracy. Off from the modern value of 3.14159265... by about 7 parts per million.

But the number isn't the most interesting thing about this verse. What's interesting is the word "approximately."

Why one word in Sanskrit matters

The verse Aryabhata wrote in Sanskrit uses the term *āsanna* — "near," "approaching but not reaching." Clark renders it in English as "approximately." Modern translators (Kim Plofker, George Joseph) read this word as something more specific: an explicit acknowledgement that the circle's circumference cannot be written down exactly as a finite ratio.

Aryabhata didn't have the modern concept of irrational numbers. The proof that π is irrational wouldn't come until **1761**, when the German-Swiss mathematician Johann Lambert published it — 1,262 years after Aryabhata.

But Aryabhata knew, or at least suspected, that the value he was giving was *as close as you can come*, not *exactly right*. That's an intellectual move no other contemporaneous mathematician had explicitly made. Greek geometers (Archimedes, Ptolemy) gave bounds — π lies between 3 + 10/71 and 3 + 1/7 — but didn't flag the underlying impossibility of finitely representing it. Chinese mathematician Zu Chongzhi (5th c. CE), Aryabhata's near-contemporary, gave the rational 355/113 = 3.14159292... — slightly more accurate than Aryabhata's — but again without the "we are approaching, not arriving" framing.

The word *āsanna* puts Aryabhata in his own category: not just *closer to pi*, but *honest about being only close*.

The mathematical landscape Aryabhata was working in

Aryabhata wrote the *Aryabhatiya* at the age of 23, in 499 CE. It's short — 121 verses across four chapters — and dense. The second chapter, *Ganitapada* ("calculation chapter"), gives 33 verses on arithmetic, algebra, geometry, and the very first systematic sine table in any tradition.

By 499 CE, Indian mathematicians had already inherited:

- A working decimal place-value system (documented from at least the 5th c. CE, but likely older; the **Bakhshali manuscript**, carbon- dated by Oxford in 2017 to 224-383 CE, shows the dot-symbol for zero already in use). - The arithmetic of fractions and proportions. - A geometric tradition stretching back to the *Sulbasutras* (~800 BCE), which gave constructions for altars requiring precise area ratios.

What Aryabhata added to this inheritance: an algorithmic approach. Instead of "this circle has the property that...", he wrote: "to compute this, do X." The pi verse is a procedure, not a description.

The math, step by step

> Add 4 to 100 → 104 > > Multiply by 8 → 832 > > Add 62,000 → 62,832

Aryabhata then says: this is the circumference when the diameter is 20,000. Therefore π = 62,832 / 20,000 = 3.1416.

The procedure structure matters. It's a recipe you can hand to someone who has never seen pi computed before, and they can produce the answer in three steps. The modern decimal notation we use (3.1416) hides the arithmetic; Aryabhata's version exposes it.

Why these particular numbers? 62,832 / 20,000 reduces to 3927/1250. Modern numerical analysis can prove this is the best rational approximation to π with denominator ≤ 1500. Aryabhata didn't have continued fractions or rational-approximation theory. He arrived at the value, plausibly, through the same geometric inscribed-polygon technique Archimedes used 700 years earlier — but with finer subdivisions and Indian decimal arithmetic doing the heavy lifting.

What happened to this verse

The *Aryabhatiya* was studied continuously in India for the next millennium, and absorbed into the Arab mathematical tradition via the 9th-century translations into Arabic at the House of Wisdom in Baghdad. The Persian polymath **al-Khwarizmi**, who gave us the word "algorithm" (from his name) and "algebra" (from a word in his book's title), drew on Indian mathematical sources directly. The Hindu-Arabic numeral system reached Europe through Fibonacci's *Liber Abaci* (1202).

In a sense, every time anyone calculates with a decimal number today, they're using the system Aryabhata was already using comfortably in 499 CE.

The Kerala school of mathematics, ~1400 CE, took Aryabhata's approximation further: **Madhava of Sangamagrama** derived the infinite series π/4 = 1 − 1/3 + 1/5 − 1/7 + ..., a result rediscovered independently by James Gregory and Gottfried Leibniz in the late 1600s. The series is now sometimes called the **Madhava-Leibniz series** in recognition.

The honest comparison

Aryabhata's value isn't the most accurate pi approximation of antiquity — Zu Chongzhi's 355/113 is closer. Greek geometers gave methodologically more rigorous proofs of upper and lower bounds.

What Aryabhata did, that no one else did before 1761, was acknowledge the limit. He gave the best rational approximation he could, and he called it *āsanna* — "approaching but not arriving."

That's the move. Not the number; the framing. A 6th-century Sanskrit text saying: *this is as close as we can get, and we can never get exactly there.*

The rigorous proof of irrationality came thirteen centuries later.

---

Sources

- [Aryabhatiya, W. E. Clark trans., 1930](https://archive.org/details/The_Aryabhatiya_of_Aryabhata_Clark_1930) — verse II.10 cited above. - Plofker, K. (2009). *Mathematics in India*. Princeton University Press. — secondary scholarly synthesis; reads *āsanna* as explicit irrationality-acknowledgement. - Joseph, G. G. (2011). *The Crest of the Peacock: Non-European Roots of Mathematics* (3rd ed.). Princeton University Press. — same reading, with detailed comparison to Greek and Chinese contemporaries. - Lambert, J. H. (1761). "Mémoire sur quelques propriétés remarquables des quantités transcendentes circulaires et logarithmiques." Mémoires de l'Académie royale des sciences de Berlin. — the formal proof of π's irrationality, 1,262 years after Aryabhata.

Related claims

(More claims will be linked here as the Mathematics pillar fills out.)