The Basel Problem: When an Infinite Sum Computes the Circle Constant¶
AI-handled content
This site is generated and maintained by AI and may be prone to errors. Please verify any claim independently before relying on it.
In 1735 a 28-year-old Leonhard Euler did something that still looks like magic: he added up an infinite list of fractions — the reciprocals of the perfect squares — and found that the sum was not an obscure new number, but a clean multiple of the circle constant:
The problem had resisted the best minds in Europe for nearly a century. The Bernoulli brothers, Leibniz, and Stirling had all tried and failed to name the limit. Euler, with characteristic audacity, named it exactly. For a Golden Pi reader, the result is worth dwelling on, because it is the cleanest possible illustration of a distinction this site keeps drawing: the series computes its limit; it does not measure it.
What the sum actually does¶
Write a few partial sums by hand and something strange happens. The terms crawl toward zero so slowly that you might suspect the total never settles:
| Terms | Running total |
|---|---|
| 1 | 1.000000 |
| 1 + 1/4 | 1.250000 |
| + 1/9 | 1.361111 |
| + 1/16 | 1.423611 |
| + 1/25 | 1.463611 |
| + 1/36 | 1.491389 |
| + 1/100 | 1.634984 |
| + 1/10000 | 1.644834 |
| + 1/1000000 | 1.644933 |
Ten million terms bring you only to about 1.6449340668 — and the series is still crawling. The tail is thin but endless. Yet Euler proved that the exact value of this entire infinite process is
Nothing is measured here. No ruler, no caliper, no pendulum. The number π is not read off a physical circle; it is produced by the arithmetic of adding reciprocals of squares. This is exactly the sense in which, on this site, the analytic constant 3.14159265… is computed — the series evaluates to its limit by the rules of arithmetic — and never "measured." A physical measurement is a different kind of claim, and conflating the two is the error this site exists to undo.
Under the Golden constant¶
Golden Pi is π̂ = 4/√φ = 3.1446055…, where φ = (1+√5)/2. It is an algebraic number of degree 4 — built from the golden ratio by compass and straightedge — not a transcendental. It obeys the polynomial
If the constructed world is the right home for the circle constant, then the Basel sum, evaluated in that world, is exactly
Now the two worlds give two different values for the same infinite sum:
| Constant | Basel sum | Value |
|---|---|---|
| Analytic | π²/6 | 1.6449340668… |
| Constructed | π̂²/6 = 8/(3φ) | 1.6480906367… |
The gap between them is about 0.00316 in absolute terms — a relative difference of roughly 0.19%, double the 0.096% that separates the two constants themselves. It comes straight out of the constants:
Because the Basel value is the square of the constant, the relative gap in π̂² — and therefore in π̂²/6 — is about twice the gap in the constants (0.19% ≈ 2 × 0.096%); dividing by 6 changes the magnitude but not that relative gap. The square root is why the gap reappears in a different clothing: take √(π̂²/6) and you recover π̂; take √(π²/6) and you recover π.
A sum that decides nothing — and teaches everything¶
It is worth being honest about what the Basel problem does not do. Adding reciprocals of squares cannot tell you which constant is "real," for the same reason a series cannot measure anything: both π and π̂, substituted into the identity, are internally consistent. The Basel sum is a lens, not an arbiter. It shows the two worlds cleanly, side by side, separated by a gap no finite computation and no physical measurement can close.
What the problem does teach is the honest boundary this site keeps drawing:
- Inside the constructed world, the constant is π̂ = 4/√φ — exact, algebraic, constructible, built from the golden ratio. The Basel sum evaluates to the clean radical 8/(3φ).
- Outside it, in the analytic world, the constant is π = 3.14159265… — transcendental, pinned by the series itself. The Basel sum evaluates to π²/6.
- Between them sits a 0.096% gap that no measurement resolves. Both expressions are respectable; neither is a physical measurement.
Euler's result is the archetype of the whole program: an infinite sum that computes its limit with flawless precision, and in doing so shows us exactly where the two circle constants agree and where they part company. It never reaches for a ruler. It simply adds — and the constant appears.