The Alpha Small

α = 1/137.035 999 084 (roughly). The fine structure constant. The number that made Feynman take sleeping pills.

There is no theory that predicts alpha. No equation whose solution is the fine structure constant. You measure it. You plug it into QED. Everything else follows. But the number itself — why this number, why not 1/138, why not π or e or the golden ratio — is one of the great unanswered questions in physics.

Wolfgang Pauli, ever the dramatic one, reportedly said that alpha was the number God used to create the universe, and that if he could question God further on the meaning of this number, God would confess that the number was the result of arbitrary design. Pauli was not given to mild opinions.

The fine structure constant is dimensionless. It doesn't depend on your units. It doesn't depend on the meter or the second or the kilogram. It is a pure number — a ratio of fundamental quantities — and its value is approximately 1/137. In terms of the constants that build the universe:

α = e²/(4πε₀ℏc)

The charge of the electron, Planck's constant, the speed of light, and the vacuum permittivity. Combined in this way, all the units cancel, and you're left with a dimensionless number. A number that, to the best of our knowledge, could be anything.

Feynman called it "a magic number that comes to us having no understanding of what it is." He made it the central mystery of his famous Feynman Lectures. The number is, he said, "one of the greatest mysteries of physics." He couldn't calculate it. Nobody could. They can measure it to parts per billion, but they cannot derive it from deeper principles.

Here's what alpha does: it sets the strength of the electromagnetic interaction. In QED, it's the expansion parameter. Every Feynman diagram carries a factor of alpha to some power. Because alpha is small, higher-order diagrams contribute less, and perturbation theory works. If alpha were much larger — say, 0.5 — perturbation theory would fail. You couldn't compute anything. If alpha were zero, electromagnetism would vanish. Electrons wouldn't bind to protons. No atoms. No chemistry. No you.

The value 1/137 is close to but not exactly 1/137. The extra precision matters. In 1988, Peter Beiner measured alpha to a precision of 2.6 × 10⁻⁹ using atomic recoil experiments. That's precise enough to test QED itself. And QED passes, every time.

There are theories that try to explain alpha's value. Julian Schwinger once proposed it could be 1/137 exactly — an integer, chosen by God. He was wrong. (He was a Nobel laureate. Being wrong about this doesn't diminish him; it elevates the mystery.) Dirac, in a moment of vanity, suggested that alpha = 1/2π ≈ 1/6.28. This is spectacularly wrong, but the gesture — trying to derive a fundamental constant from pure mathematics — is the one every physicist wants to make.

John Archibald Wheeler proposed that all electrons are the same electron, moving forward and backward in time, creating the appearance of many electrons from a single entity. In this picture, alpha would be determined by the self-interaction of a single electron with its own worldline. Wheeler never fully developed the idea. It's beautiful, but it doesn't give you 1/137.

The string theorists have a say too. In some formulations, alpha is a moduli parameter — a value determined by the shape of extra dimensions. This is unsatisfying in the same way that saying "the answer is buried in a rock" is unsatisfying. The rock might tell you the answer, but you'd have to excavate the entire extra-dimensional structure of the universe to find it.

Here's the thing that keeps me up at night: alpha runs. It's not actually constant. At higher energies — in the early universe, or inside a particle collider — alpha is slightly different. At the scale of the Z boson mass (about 91 GeV), alpha ≈ 1/128. The coupling gets weaker at higher energies. This is called "running" and it's a prediction of renormalization group flow. The vacuum polarization of virtual electron-positron pairs screens the charge, and at longer distances, you see a smaller effective coupling.

But the low-energy value — 1/137.035 999 084 — is what matters for chemistry, for biology, for the existence of stars. If it were 2% different, carbon wouldn't form in stellar nucleosynthesis. The universe would be a broth of hydrogen and helium. No elements. No planets. No questions about why alpha is 1/137.

Richard Feynman, in his characteristic blend of profundity and irreverence, put it best: "It's one of the greatest mysteries. There should be a simple way to just get this number — to show that our universe is set up this way — but we don't know how. We don't know why."

Why is there a why? Why does the universe care about one number, and not just any number, but one that nobody can derive? That's the mystery. Alpha is the number, but the question is why the number exists at all.

The fine structure constant is 1/137.035 999 084(21). We know it to eleven significant figures. We can use it to predict the magnetic moment of the electron to fourteen figures. We can build atomic clocks that don't lose a second in the age of the universe. But we cannot answer the question: why 1/137?

God, it seems, kept the answer to Himself.