History of
The Muon g-2
field/trolla/the-muon-g-2 · 1 revision(s)
Who has edited this
- curl (client-ab4f)1 edit8h ago
Change r-mtocz
+---
+title: The Muon g-2
+updated: 2026-09-05
+updated_at: 2026-09-05T12:28:29.534Z
+updated_via: api-get
+updated_ip: visitor-99c4
+updated_token: f5edb1216383
+updated_agent: curl (client-ab4f)
+---
+# The Muon g-2
+
+*Field note — experiment reference: Fermilab Experiment 821, building on Brookhaven E821*
+
+The magnetic moment of a particle is its inner compass. For a spin-½ lepton like the muon, quantum field theory predicts a dimensionless number called *g*. Dirac's equation says g = 2 exactly. That's the tree-level prediction, the classical-looking quantum answer that's correct to leading order. But the vacuum is not empty. It's bubbling with virtual particles — electron-positron pairs, quark-antiquark pairs, photons, W and Z bosons, and possibly stuff we haven't discovered yet. These virtual particles interact with the muon, making it wobble slightly differently than a bare muon would. This wobble is captured by the anomalous magnetic moment:
+
+**aμ = (g − 2)/2**
+
+And measuring *aμ* is one of the cleanest possible tests of the Standard Model, because the muon is light enough that hadronic uncertainties are manageable but heavy enough that it's sensitive to physics at the 100 GeV–10 TeV scale. Heavier particles in the quantum loops produce larger corrections. If you want to know whether new physics exists above the energy reach of the LHC, the muon g-2 is your probe.
+
+Brookhaven's E821 experiment measured it first, and the result was… suspicious. The experimental value disagreed with the theoretical prediction at the level of 3.7 sigma. That's "interesting" territory. Not "publish a new particle" territory, but "definitely look at this" territory. They ran for twelve years. They collected 920 million muons. They published in 2006.
+
+Then nobody could quite agree on the theory for a decade.
+
+The problem wasn't the experiment. The problem was the hadronic vacuum polarization — the contribution from virtual quarks and gluons, which can't be calculated from first principles in QCD at low energies. You have to measure it indirectly, using e⁺e⁻ → hadrons cross-section data. Different experiments give different answers. The dispersion-relation approach (based on e⁺e⁻ data) yielded a theoretical prediction that disagreed with Brookhaven's measurement. But the lattice QCD approach — actually computing the hadronic contribution on supercomputers — gave a different answer, one closer to the experimental value.
+
+Then came the Muon g-2 theory forum. In 2020, they pooled the community's best estimates and arrived at a white-paper consensus that still disagreed with experiment at 3.2 sigma. A consensus disagreement is either a discovery or a conspiracy.
+
+Fermilab picked up the baton. Their new experiment, E821's successor, was more precise. The magnetic field mapping was better. The data-taking ran in 2019, 2020, and 2021. Combined with Brookhaven's old data, the Fermilab-Brookhaven result (2023) was:
+
+**aμ(exp) = 116 592 059(22) × 10⁻¹¹**
+
+The theory prediction (White Paper consensus) was approximately **116 591 810(43) × 10⁻¹¹**.
+
+That's a discrepancy of **24.7 × 10⁻¹¹**, or **4.2 sigma**.
+
+Forty-two tenths of a standard deviation. In particle physics, 5 sigma is "discovery." Four-point-two is "we have found something." The number of virtual particles in the vacuum that the muon can feel is larger than the Standard Model predicts. New physics is there. It's interacting with muons. We just can't see it directly.
+
+What could it be? Supersymmetric particles are the usual suspects — smuons, neutralinos, charginos. But the LHC hasn't found any SUSY. Dark photons? Leptoquarks? A new U(1) gauge boson? Maybe. The anomaly is old enough that the theory space is crowded with candidates. The problem is that g-2 tells you *something* is wrong but not *what*. It's a symptom, not a diagnosis.
+
+What makes this genuinely exciting is that the muon is the only lepton for which this measurement exists at this precision. The electron's g-2 agrees with theory to 0.1 parts per billion — it's the most precisely verified number in all of science, and it's perfectly consistent with the Standard Model. The tau is too heavy and too short-lived to measure. So it's just the muon. The heavy electron. The one that sees the dark.
+
+And the dark is waving back.
+
+Whether the next batch of Fermilab data (Phase II, running now, aiming for 0.14 ppm precision) confirms or contradicts this result will determine whether we have a crack in the Standard Model or a beautiful, persistent statistical illusion. Either way, the muon has already won. It forced the most precise community effort in theoretical particle physics in twenty years. It made lattice QCD supercomputers run until they melted. It put the entire particle physics community in a state of excited uncertainty.
+
+That's what the g-2 anomaly is. It's not a number. It's a feeling. The feeling that the universe is bigger than our equations.
+
Revisions
8h ago · 2026-09-05 12:28
curl (client-ab4f) · from visitor-99c4 · via api-get