Well, every now and again an experiment reports a result that forces scientists to go back to a long-established principle, to check whether it needs revision, extension or adjustment, or perhaps even replacement. Most times it eventually turns out that the experiment is wrong, though often in some subtle and non-obvious way, and the principle survives. But of course there are the rare occasions when it is the other way round. So a scientist must go into such a situation with an open, though skeptical, mind.
Is there a more famous principle from 20th century physics than Einstein’s principle that nature has a speed limit? We call this the “speed of light”.
We call it that. However, let’s be a bit careful.
There are actually two principles at work here.
- The first is that a speed limit exists, due, in Einstein’s way of thinking, to the geometry of space and time themselves.
- The second is that all massless particles, including light, travel at the speed limit — though only in completely empty space, i.e. in “the vacuum’‘. (In water, for instance, light travels slower than the speed limit, and electrons in water can actually travel faster than light in water! [But not faster than the speed limit!] A fact which is used in many particle detectors.)
- Actually the second principle is more general: it further states that any particle whose motion-energy is far, far greater than its mass-energy (m c-squared) should travel (in vacuum) very, very close to (but just a tiny, tiny bit below) the speed of light.
This evening an experimental collaboration called OPERA put out a technical document. In it, they claim to find that pulses of high-energy neutrinos, sent from the CERN laboratory near Geneva, Switzerland, arrive 730 kilometers away, in the underground Gran Sasso laboratory in Italy, just a bit earlier than expected. According to the principles I just stated, these neutrinos should travel at just about the speed of light. But apparently they are traveling slightly faster. To deduce their speed requires measuring the 730 kilometer straight-line distance, from the point where the neutrino beam pulses are created to the location of the OPERA experiment, to an accuracy of 20 centimeters. The excess in the speed is 25 parts per million of the speed of light, which translates into an early arrival of about 60 nanoseconds (billionths of a second.) This is not a simple measurement to be made with a stopwatch and a ruler! Clearly the experimenters have worked very hard.
If this claim were right, it would be revolutionary. At minimum, it would force physicists to modify at least one principle that has been taken almost (but not entirely) for granted for many decades.
But is it right?
Only a critical and thorough review by the high-energy physics community, and repetition of the experiment (or measurements using other experimental techniques), will tell us whether this result is correct. We certainly have to consider other measurements of neutrino speeds. As I described in my previous post, neutrinos from the 1987 supernova seem to have traveled over 160,000 light years at a speed within a few parts per billion of the speed of light. What OPERA sees is a several-thousand-times larger effect, which might seem inconsistent. However, one can’t use the supernova measurement to argue that OPERA’s measurement must be wrong. The supernova neutrinos had lower energy, by a factor of a few hundred, compared to those that are in OPERA’s neutrino beam. There are other possible differences too. And perhaps these differences matter. So we probably can’t reason away this result. Particle physicists will have to check it, repeat it, and try other, related measurements.
Whether the result is right or wrong, theorists and experimentalists in high-energy physics will do some serious thinking about it. We’ll ask questions such as: What mistakes might the OPERA team have made? Could there be any subtle way in which this result would cause a conflict with other previous experiments? What experiments might be done that would check the result? If the result were true, what other related phenomena might be present in nature, and could we do experiments to look for them? At worst, even if time tosses OPERA’s claim into our very big garbage pile of false discoveries, we’ll learn something by thinking things through, and will be clearer about what we know and how we know it.
There will be much more to be said, but that’s enough for the moment. High-energy physicists need to read carefully through the OPERA “preprint” (the pre-publication paper, 22 pages long, technical and subtle), listen carefully to what the OPERA experimentalists have to say, and examine carefully the papers written by the expert theorists who have considered carefully how Einstein’s principles could be modified. Carefully, carefully, carefully; this is a complex and subtle subject, and it is easy to make wrong statements and draw wrong conclusions. As the dust settles, and as I learn enough on my own and from my peers that I am confident I can explain more of the issues both clearly and correctly, I’ll write more about it.
A last remark for the night: think about what it is like to be an experimentalist making a revolutionary statement of this magnitude. Talk about sticking your neck out! This result either means a Nobel Prize or international embarrassment — perhaps even ridicule if a serious mistake was made; there’s no middle ground. The combination of excitement, hope, and terror must be unlike anything most of us will ever experience. I cannot imagine how any of them have slept for days; I cannot imagine that they will sleep well for months, until a second experiment reports, “We have measured the speed of neutrinos, and we confirm…”
30 Responses
but if the limiting speed is not the speed of light, shouldn’t we something going on in a Michelson-Morley kind of experiment ?
but if the limiting speed is not the speed of light, shouldn’t we see something strange going on in a Michelson-Morley kind of experiment ?
Yes, but it is a quantitative question… how big is the effect?
I presume all metrology (including GPS) is calibrated ultimately with reference to the definition of the metre – ie. the distance travelled by light in ~1/(3*10^8) s. Presumably we can’t, by definition, be mismeasuring the speed of light (in other words, the speed of light is dimensionful, so its value has no objective physical meaning). So measuring the speed to be “apparently” different would necessitate other, physical changes like in the value of the fine structure constant, no?
Or, put more simply, measuring the speed of neutrinos using the GPS is an extremely complicated thing to do. Why not instead set up a race between photons and neutrinos and see which one arrives first?
It is technically difficult to do this, for multiple reasons, but it might well be that someone will propose an experiment of this sort. We will see many experiments proposed, using a variety of techniques, in the next weeks.
It’d be extremely interesting if Opera could repeat their experiment with photons instead of neutrinos, using the same techniques but, of course, a different path that avoids solid rock. Maybe they would find that their photons also travel faster than light in air, in which case the conclusion would be that (a) neutrinos travel at the same speed (or more likely, slightly more slowly) than photons, and (b) that their determination of the speed of light is slightly inaccurate.
I tried to ask at different sites what it would mean for QG, if the superluminous neutrinos would be confirmed by other experiments, but I get no answer so far (maybe because I am premature with this?) .
So I want to bug people here a bit with this ;-P …
Could a confirmation of the Opera measurements lend some “updraft” to discrete QG theories like LQG, spin networks and spin foams or even be a hint of something like that?
I`m just curious as always 🙂
Is the supernova case really relevant? What if there were neutrinos that arrived a few years early? Would there have been detectors in 1980? Would somebody correlate a blip in the detector with a supernova whose light arrives a few years later?
Absolutely it is relevant, in that we know, from the fact that some neutrinos were observed, that at least *some* neutrinos traveled very close to the speed of light on their way to earth. But it is not definitive, in that we would not know if neutrinos of the three different types traveled at different speeds. [However, other observations may imply that neutrinos cannot travel at such different speeds; I am still learning about this point.]
Based on some recent experience, they don’t face any risk. I haven’t seen anyone who has put out self-evidently absurd claims whose wrongness was soon demonstrated beyond any doubt and who was humiliated by many people.
So they unfortunately only risk fame.
I guess that all of us have some better feel for the setup of the experiment. After some looking at it, the single most likely source of a mistake is the GPS-powered timing. If they neglected that the atmosphere’s index of refraction (and speed of light) isn’t 1 but n=1.0003, they could get a mistake of exactly the same magnitude.
Neutrinos could be faster than the light in the air/vacuum (different media at different parts of the path). Indeed, they are: neutrinos move almost exactly by the speed of light. This trivial observation could have been misinterpreted.
I find it obvious that if relativity holds, there is a big mistake of this kind somewhere in their method because the errors accumulated at various places they’re aware of are safely lower than 60 nanoseconds, indeed.
Of course, if this result were right, it would be spectacular. We would have to to think about non-relativistic theories or some subtle entanglement between the labs, “precursors” in Joe Polchinski’s words etc.
Thanks for your remarks. I don’t think I agree on the embarrassment issue, Lubos. This is the most publicly visible scientific announcement in some years. If a silly mistake such as you suggest is responsible for the effect, that may very well make front-page news just as today’s announcement is doing. It will be very embarrassing.
I agree about the importance of accounting for the fact that light travels a few dozen parts per million slower in air than in vacuum (though this has to be integrated along the path to the GPS satellites). Both travel speed and perhaps even curving due to refraction have to be carefully accounted for. One would hope that metrology experts would be well aware of these subtleties, no?
Meanwhile: what is your point of view on the wisdom of combining systematic errors in quadrature with each other and with statistical errors? I am concerned about the statement that this is a 6-standard-deviation effect.
Neutrino physics is becoming more and more interesting. May be some interesting neutrino phenomena will explain this event and relativity will survive as always :p
My guess is that the decay which produces the muon and mu neutrino happens earlier in the decay tube than they think.
They specifically argue in their paper that any such error would not change their result, because the parent pion is also traveling close to the speed of light.
I hope there will be no ridicule if the outcome is negative as the researchers prefaced their finding as saying they were shocked and the article at Nature News indicated they had observed the effect 16,000 times over two years. Another article I think indicated they had waited 6 more months just in checking and rechecking their work. This seems diligent and sober and should be applauded no matter the final verdict over this work.
I agree, but the court of public opinion (and of some scientific opinion) is not often so kind.