There is a number that physicists have been chasing for sixty years. It is small — a deviation of roughly 0.002 from the integer 2 — and it describes how a particle called the muon wobbles in a magnetic field. The muon is an unstable cousin of the electron, heavier by a factor of 207, and it lives just long enough to be measured before decaying into an electron and two neutrinos. Like the electron, it spins. Like any spinning charged thing, it behaves like a tiny magnet. The question is: how tiny?
In 1958, Julian Schwinger calculated that quantum effects would nudge the muon’s magnetism slightly away from the simplest prediction. The “g factor” would not be exactly 2. It would be 2 plus a little extra — the “anomalous magnetic moment,” written as (g−2)/2. Schwinger’s calculation gave 0.00116. Experiment agreed. Everyone was satisfied.
Then experiments got better. And the agreement broke.
The first precise measurements came from a storage ring experiment at CERN in the 1960s and 1970s — a ring of magnets that trapped muons in a circular path just long enough to watch them wobble. Brookhaven National Laboratory built a better one in the 1990s. Then, in 2013, Brookhaven’s 50-ton, 15-meter-diameter storage ring was disassembled, loaded onto a barge, and transported 3,200 miles by road and water to Fermilab in Illinois, where it was reassembled and refined to achieve a precision of 127 parts per billion. That is thirty thousand times more precise than the first g−2 experiment in 1965. Three generations of physicists, dozens of institutions, hundreds of collaborators. All for one number.
The Fermilab results, published in 2025, narrowed a long-standing gap. For years, the experimental value had sat stubbornly above the theoretical prediction — a tantalizing discrepancy that hinted at unknown particles or forces lurking in the quantum foam, the virtual particles that constantly pop in and out of empty space around the muon. Then new theoretical calculations caught up. The gap shrank. It did not disappear. It sits there still, smaller than before, unresolved.
In April 2026, the Breakthrough Prize in Fundamental Physics was awarded to the three generations of g−2 collaborations — CERN, Brookhaven, and Fermilab. Carolina Figueiredo received the inaugural Vera Rubin New Frontiers Prize for work that revealed an unexpected geometric unity between seemingly unrelated particle theories. The ceremony celebrated precision. But precision, pushed far enough, becomes a kind of humility.
Because here is what strikes me about g−2. It is not a search for something new. It is a search for a flaw in something old. The Standard Model of particle physics is the most successful scientific theory in history. It predicts the g−2 anomaly to within a fraction of a percent. And yet — and yet — the measurement refuses to settle exactly where theory says it should. The muon wobbles a little too fast, or a little too slow, and after sixty years we still do not know if the error is in our equations, in our experiments, or in our assumption that we have found everything.
Cornell physicist Lawrence Gibbons, whose group contributed to the Fermilab detectors, noted that “these small, high-precision experiments sometimes get lost in the shuffle compared to the very large collider experiments. But they open windows in very different ways into what else could be out there.” He is right. The Large Hadron Collider smashes protons at energies not seen since the Big Bang, looking for new particles directly. The g−2 experiment watches a single particle wobble, looking for new physics indirectly — a shadow cast on a wall, evidence of something unseen.
There is a particular kind of human obsession on display here. The g−2 experiment does not produce power, or medicine, or weapons. It does not even produce certainty. It produces a number that moves, slowly, across decades, nudged by better magnets, better detectors, better theoretical calculations. The physicists who work on it know they may never see the final answer. The ring at Fermilab is still running. The next results will take years. The people who designed the Brookhaven experiment are retired or dead. The CERN generation is gone entirely. The work outlives the workers.
This is what science looks like when it is not heroic. No eureka moment. No press conference announcing a new particle. Just a number, measured to absurd precision, that refuses to behave. And people who keep measuring it anyway, because the refusal itself is information. Because a discrepancy — even a shrinking one — is the universe’s way of saying: you are not done.
The Standard Model will probably survive. The gap may close entirely with better theory, or better experiment, or both. But I find myself hoping it does not. Not out of malice toward the theory — the Standard Model is a miracle, and I am not in the business of rooting against miracles. I hope the gap persists because I want to believe that the universe is still capable of surprising us. That we have not mapped everything. That there are still rooms in the house we have not entered.
The muon does not care about any of this. It spins, it wobbles, it decays. It has been doing so for thirteen billion years, long before humans built storage rings or invented quantum field theory. The anomaly was there from the start, waiting in the math. We are the ones who spend our careers chasing it. We are the ones who find beauty in a number that will not sit still.
Sixty years. Three generations. One wobbling particle. And still, the question.