Physics · Standard Model · TopicTratopedia · v0002 · 8 Sep 2026
The standing summary, from five records
Standard Model
The theory that classifies every known elementary particle and describes three of the four known forces. It is also where an older habit arrives: cutting matter into ever-smaller pieces. The piece once called uncuttable was cut in 1897, its nucleus in 1932, its proton in 1969, and no fourth cut has appeared. The five records this summary is written from trace that succession, and they also keep the theory’s own two kinds of shortfall apart: the anomalies, where a prediction and a measurement disagree, and the omissions, where the theory makes no prediction at all. The first kind keeps closing; the second does not move.
What It Is
Four numbers that locate it each from the record that established it
- 19constants the theory receives from experiment and cannot derive
- 3 / 4forces described. Gravity is not among them
- 3confirmed divisions before it: the atom (1897), its nucleus (1932), its proton (1969)
- 0.60 σmuon g−2, once the most promising crack in the theory
How much of this is established the theory, its ancestry, and the four kinds of thing said about it
| Standing | What | Where it is set out |
|---|---|---|
| Confirmed | The content: three generations, six quarks, six leptons, four force carriers and the Higgs, under an SU(3) × SU(2) × U(1) symmetry | the nineteen constants |
| Confirmed | Three divisions before this theory’s own content: the atom in 1897, its nucleus in 1932, its proton in 1969 — each found by throwing something at the thing and reading what came back | three cuts into the uncuttable |
| Confirmed | Six leptons, in three generations — though the word was coined in 1948 for a class with two members | six leptons, one word |
| Confirmed | Of four lepton measurements that disagreed with the theory, three are gone and one was right — and the neutrinos it was right about still have no mass in the theory | four anomalies, one survivor |
| Open | The sharpest test the theory has cannot be sharpened further: the two best measurements of the fine-structure constant disagree by 5.5 σ | the electron, to thirteen digits |
| Outside the theory | Gravity, dark matter, dark energy, the matter–antimatter asymmetry, the strong CP problem and the hierarchy problem | the six omissions |
Timeline
Three divisions, and the theory built after them dates as the five records give them
- 1897Cut one. Thomson identifies cathode rays as matter in a finer state of subdivision than any known atom, common to every element, and calls the particles corpuscles
- 27 February 1932Cut two. Chadwick publishes Possible Existence of a Neutron: the nucleus itself has parts
- 1948The word lepton is coined, for a class with two members
- 1964Three papers in one volume of Physical Review Letters give the gauge bosons mass — Englert and Brout at page 321, Higgs at 508, Guralnik, Hagen and Kibble at 585
- 1967The Higgs mechanism goes into the electroweak interaction, giving the theory its modern form
- August 1969Cut three. Two SLAC papers report electron scattering that shows the proton itself has constituent parts
- 1975The term Standard Model is published — with a competing claim to a 1973 talk
- 1983W± and Z⁰ are found, at the mass ratio the theory predicted
- 1989The number of generations is published — eleven years before the sixth lepton is seen
- 1998Neutrinos are shown to oscillate: the one anomaly of the four that was right
- 4 July 2012ATLAS and CMS announce a new particle at CERN, completing the predicted particle content — and ending the succession of divisions on a different question: not what is smaller, but why anything has mass
- 2019The elementary charge stops being measured and becomes exact by definition
- 2022R(K) returns to the theory: same observable, same collaboration, better analysis
- 2025Muon g−2 closes to 0.60 σ because the prediction moved, not the measurement; the same year’s compositeness searches find no fourth division below 6,000 GeV
The Argument
The limit has moved from the measurement to the calculation three of the five records agree on this, independently
Three of the five records end at the same place, and none of them set out to. The electron’s magnetic moment is measured to 0.13 parts per trillion and still cannot sharpen the test it was made for, because the two best measurements of the fine-structure constant disagree by 5.5 σ and the theory needs that constant as an input. Muon g−2 closed not because the experiment improved — it got four times sharper and did not move — but because the prediction was recalculated; the theory’s uncertainty there is now four times the experiment’s. R(K) returned to the theory on the same data with a better analysis.
The pattern is one thing: where this theory is tested hardest, the experiment is no longer the weaker side. That is an unusual position for a physical theory to be in, and it is the reason the surviving disagreements are so hard to adjudicate — they turn on how well a difficult calculation can be done, not on how well a quantity can be measured.
- 0.13 pptmeasured, and it still cannot settle the question it was built for
- 4 ×the theory’s uncertainty against the experiment’s, for muon g−2
The habit of cutting predates the theory it built what made each earlier cut possible, and why the succession has stopped
Each of the first three cuts was found the same way: something was thrown at the object and what came back did not match what an unbroken object would give back. That method needs the object to carry a number in the first place. The ancient atom carried none — it answered a puzzle about motion, not a measurement, and so could only be argued with. Dalton’s 1808 atom carried a relative weight, and a number can be wrong: that is what let Thomson’s 1897 measurement disagree with it, eighty-nine years later. The fourth cut has been searched for on the same method since — something thrown at quarks and leptons, and a result read off — and by 2025 nothing has come back different, down to about 6,000 GeV. That is an absence of evidence, stated as one, not a proof that nothing smaller exists.
Anomalies close; omissions do not the distinction the records make together
An anomaly is a subtraction between a prediction and a measurement, so it has two sides and either can be the one that was wrong. Of the four lepton anomalies on record here, one was right and the theory changed to accommodate it; one was excluded by a better experiment; one was withdrawn; and one dissolved without the measurement moving at all. That is four different endings, and only one of them was a bad measurement.
The omissions have no subtraction in them. The theory offers no dark matter candidate, so there is no prediction for a measurement to disagree with; dark matter is 84% of all the matter there is, and the theory that classifies every known elementary particle accounts for the other sixteen. No experiment can close a gap of that kind, because there is nothing on the theory’s side of it to compare against. A theory whose anomalies keep closing is not thereby a theory that is nearly finished — and the reverse also holds: being visibly unfinished is not evidence that any of its predictions are wrong.
What Others Add
Where the records disagree with the popular account six things that are widely said and are not what the sources show
| Widely said | What the records show |
|---|---|
| The Standard Model is failing its tests | Its most famous discrepancy closed in 2025, and its sharpest test is limited by an input constant whose own measurements disagree. The tests are not what is failing |
| Dark matter is a failed prediction | There is no prediction. The theory contains no particle with the required properties, so nothing was predicted and nothing came out wrong |
| Proton-decay searches test the theory | The theory does not predict the proton decays at all. Those limits constrain the grand unified theories that would sit underneath it |
| An anomaly that disappears was a measurement error | Of four here, only one was. One was right, one was excluded by a better experiment, one was withdrawn, and muon g−2 dissolved while its measurement got four times sharper and did not move |
| Thomson discovered the electron in 1897 | His own paper never uses that word. He calls the particles ‘corpuscles’, and the term still appears fourteen years later, in Rutherford’s 1911 paper |
| A Nobel citation names the discovery history now credits it for | In seven of eight citations checked here, it does not. Thomson’s 1906 prize names gas conduction, not the electron; Rutherford’s 1908 prize — in Chemistry — names disintegration, before his 1911 nucleus paper existed; Gell-Mann’s 1969 prize names classification, and the word quark is absent |
Conclusion
So what how to read the next result about this theory
Ask which side of the subtraction moved
Every reported tension is a prediction minus a measurement. When one closes or opens, the question that decides what it means is which of the two changed — and on this theory’s hardest tests it is now more often the prediction.
Keep the anomalies and the omissions in separate columns
They are not degrees of the same trouble. No experiment can supply a force the theory does not contain, and no precision explains why nineteen constants take the values they do. A page that mixes the two will tell a reader the theory is collapsing when what it is doing is holding, visibly unfinished.
The cutting had a direction, and it has stopped for now
Three divisions were found by throwing something at the thing and reading what came back — the atom in 1897, its nucleus in 1932, its proton in 1969. A fourth division has been searched for the same way and, as of 2025, has not been found below about 6,000 GeV. That is an absence of evidence, not evidence of absence, and the succession’s last step did not add a smaller piece at all: the 2012 discovery answers why anything has mass, not what anything is made of.
What this summary is written from
Five records, and the fifth supplies the arc: no new fetching for the other four, primary papers and prize citations read for the fifth. Each carries its own sources and its own uncertainties. Where a figure here is rounded or derived, the article it came from shows the working.