E=mc^2 Verified In Quantum Chromodynamic Calculation

chirishnique and other readers sent in a story in AFP about a heroic supercomputer computation that has verified Einstein's most famous equation at the level of subatomic particles for the first time. "A brainpower consortium led by Laurent Lellouch of France's Centre for Theoretical Physics, using some of the world's mightiest supercomputers, have set down the calculations for estimating the mass of protons and neutrons, the particles at the nucleus of atoms. ... [T]he mass of gluons is zero and the mass of quarks is only five per cent. Where, therefore, is the missing 95 per cent? The answer, according to the study published in the US journal Science on Thursday, comes from the energy from the movements and interactions of quarks and gluons. ... [E]nergy and mass are equivalent, as Einstein proposed in his Special Theory of Relativity in 1905." Update: 11/21 15:50 GMT by KD : New Scientist has a slightly more technical look at the accomplishment.

Comments

completely bogus interpretation

[bcrowell]

The article at theage.com gives a completely bogus interpretation, which is repeated in the slashdot article. The New Scientist article is much better.

It's taken more than a century, but Einstein's celebrated formula e=mc2 has finally been corroborated,

This is just total scientific illiteracy. E=mc2 has been verified over and over again. We see it, for example, in processes like alpha decay, where the sum of the masses of the product nuclei doesn't equal the mass of the original nucleus. Mass is converted into energy in that process, and that's been experimentally established since probably the 1920's. Likewise energy can be converted into mass, as when cosmic rays hit the atmosphere and create electron-antielectron pairs. The theoretical foundations of E=mc2 are also extremely firm; it's deeply linked to the basic logical structure of relativity, and relativity has been abundantly experimentally verified.

Saying that this calculation verified E=mc2 is just stupid. The calculation assumes (1) special relativity, (2) quantum mechanics, (3) some technical stuff about how to make special relativity and quantum mechanics work together (generic ideas about quantum field theory), and (4) a bunch of very specific technical approximations needed in order to get an answer out of this particular flavor of quantum field theory (lattice QCD). The calculation has a bunch of adjustable parameters (quark masses, coupling constants). You play with the adjustable parameters and get a bunch of numbers out (neutron and proton masses, etc). If the number of adjustable parameters that goes in is m, and the number of experimentally testable numbers that pop out is n, then n-m is the number of degrees of freedom that verify whether the calculation is right. (For n=m, it would just be a complicated exercise in fitting the data, like putting two points on a graph and saying "look, it's a line!") I assume they calculated more than just the mass of the proton and neutron, because otherwise n=2 would be less than m. I assume the n-m degrees of freedom checked out fairly well, because they're calling it a success.

To see why this calculation can't really be interpreted as a test of E=mc2, you have to imagine what would have happened if it had turned out wrong. If it had disagreed with experiment, then we would conclude that some of the assumptions built into it were wrong. Let's look back at the assumptions 1-4 above. Well, 1 (special relativity) has been verified a zillion different ways since 1905 (or even as far back as the 19th century, the Michelson-Morley experiment, with hindsight). 2 (quantum mechanics) has likewise been verified a zillion different ways since the 1920's. 3, the general framework of quantum field theory, has some ugly spots, but it's been used to verify things like the magnetic moment of the electron to a dozen decimal places, so it's still on fairly firm ground. 4 is extremely shaky; it's only very recently that anyone has claimed to be able to calculate anything at all useful and realistic with QCD. So if it had failed, no physicist in the world would have interpreted it as evidence that assumption 1 (relativity) was wrong. They would have interpreted it as evidence that assumption 4 was wrong: the lattice QCD approximations weren't good enough, probably for very boring, technical reasons that would only be of interest to a specialist in lattice QCD.