A very interesting paper appeared on the arXiv last week, from Bob McElrath. Bob is a former theory postdoc at UC Davis and now at CERN. He's spent a good deal of the last several years on this idea, and now it's out there in print, though not yet peer-reviewed. This is more Sean's and Mark's territory, but to me it seems that if he's right a number of sacred cows may be headed for oblivion. One of the great mysteries of modern physics is why gravity is so much weaker than the other forces (strong, electromagnetic, and weak). Many great minds have worked to incorporate gravity into the same sort of relativistic quantum field theory that we use to describe the other three, and have failed more or less utterly for decades. Is there something fundamentally different about gravity? Einstein's general relativity, which links gravity to the warping of spacetime in the presence of matter and energy, is extremely successful in accounting for a wide variety of phenomena from very short (millimeter) to very long (solar system) distance scales. One might argue that GR is not working perfectly well on galactic or larger scales - unless and until we can identify the nature of the dark matter causing galaxies to rotate in a way which apparently violates Einsteinian/Newtonian gravity, and causes the lensing of light from very distant (billions of light years) galaxies. Bob's short paper, presumably a precursor to a much longer and complete description of his work, brings together several different lines of thought from different subfields of physics, including particle physics and condensed matter, to propose a new theory of how gravity arises. In a single sentence, it goes like this: What we know as gravity is actually the result of interactions with relic neutrinos, which satisfy all the conditions necessary to form a superfluid once the universe has expanded sufficiently. Oh, and another sentence, this time from his concluding paragraph:
"...WIMP dark matter scenarios are inconsistent: WIMPs cannot both be decoupled and localized for the age of the universe."
That is to say, we cannot have dark matter particles of mass of the usual magnitude (the 100 GeV scale) and expect them to behave classically for the age of the universe. Bob has given a number of talks on his ideas, and tells me that it's gone well so far - there have been no real show-stoppers raised. The whole picture has a certain compelling nature to it: there must be relic neutrinos (if the Big Bang expansion of the universe is correct), and if so, they must form a superfluid, the condensate of which leads to Goldstone bosons that can be identified with spin-2 gravitons. Now, I hope I am not mangling Bob's ideas here; I am no expert in any of this. But I can immediately think of lots of questions: does this field really approximate Newtonian gravity at large distances, and give us Newton's constant G accurately? Can we develop a self-consistent Big Bang cosmology incorporating this neutrino superfluid? What do we need to explain dark matter (galactic rotation, lensing) if WIMP dark matter is inconsistent? It will be interesting to see this new paradigm grow and spread through the community, if it's right. Or, maybe, someone (Sean? Mark?) will find a fatal flaw.