Matt Strassler 10/17/11
There is clearly some confusion regarding the degree of scepticism that physicists are manifesting toward the OPERA experiment’s claim of faster-than-light neutrinos. I’ve seen various statements, even by physicists, that suggest this level of doubt has to do with “protection of the establishment”, perhaps similar to a political oligarchy with a vested interest in defending its turf. I disagree with this point of view. While there certainly is a level of politics in any human endeavor, including science, I don’t think that’s what’s going on here at all.
For one thing, everyone should keep in mind that the best thing that could possibly happen to high-energy physics would be for OPERA to be correct! The possibilities for new discoveries, both experimental and theoretical, and for technological advances currently unforeseeable, might be enormous — as they were when classical Newtonian mechanics had to be improved and extended by the introduction of quantum mechanics. Revolutions are wonderful things — and I certainly have a vested interest in having one happen on my watch.
However, false revolutions are another matter. And of course they are much more common than real ones.
The reasons for the intense scepticism about OPERA are both general and specific. The general reasons stem from the track record of experiments on the frontiers of science, which is pretty dismal. This is not because experimentalists are careless or foolhardy (well, occasionally this happens) but because doing first-of-a-kind experiments, using new and clever methods and the latest technology, is extremely difficult, and prone to unforeseen problems. And statistical flukes can always happen, too. Everyone who has worked in high-energy physics for a while knows that the vast majority of exciting results, even from the best experimentalists, simply don’t hold up over time. I made an informal list over the weekend of false alarms that have occurred during the nearly 30 years that I’ve been following or actually doing high-energy physics, and came up with nearly two dozen separate incidents — and I keep thinking of new ones. [I may do some writing later this week about how some of these “discoveries” went awry.] Meanwhile I can think of only three actual discoveries that survived, one of which (the top quark) was expected, one of which (neutrino oscillations) was pretty exciting but not unexpected, and only one of which really violated the prejudices of my field. The last — the only real shocker to occur during my career — won this year’s Nobel Prize: the discovery that the universe’s expansion is accelerating instead of decelerating.
The specific reasons for distrusting OPERA are, I think, of three subtypes. And they don’t have much to do with Einstein and his theories being somehow a sacred cow. First, there is concern about OPERA as an experiment. Remember OPERA’s main purpose was to study neutrino oscillations; it was not designed for measuring neutrino speeds. Consequently there are things that make OPERA’s measurement harder than it would be for an experiment whose design was optimized for tests of Einstein’s speed limit. Second, there are some issues surrounding the experimental technique used by OPERA, including those I discussed after the original OPERA presentation: the complexity of the measurements of times and distances, the employment of rather long pulses of neutrinos to detect very short time shifts, and the use of an aggressive statistical technique in a context where, in my view, they ought to have done something much more conservative. And finally, there are the theoretical arguments (I’ve written about two of them here and here) that show that it is very difficult to accommodate OPERA’s result within any minor adjustment of Einstein’s relativity, such as those that have been often discussed in the past (see Professor Alan Kostelecky’s webpage for some information on the many previous suggestions), and also difficult to make major adjustments to relativity for neutrinos without running afoul of constraints on such adjustments from prior experiments on electrons and photons. This is not blind scepticism; it is reasoned scepticism. Nor is it close-mindedness; it still leaves the door open.
It’s interesting to compare this situation with what happened when the accelerating universe was discovered. Scepticism also ran very high, but the arguments against the claim were weaker than those against OPERA’s. First, the observers had the measurement of the deceleration or acceleration of the universe as a primary goal, and their methods were optimized for that purpose. Second, it was not easy to find fault with their observational techniques — though there were still plenty of proposals for ways they might have been fooled by astronomical effects, or even by the effect of a new particle. Third, although there was a strong theoretical prejudice in the particle physics community against an accelerating universe, there was no calculation anyone could do to argue the result was in strong conflict with previous experiments — no obstructions from prior data to match the ones that have been raised for OPERA. Fourth, there had even been arguments by astrophysics theorists, as far back as 1995, arguing that data from other sources suggested the universe was accelerating. And fifth, the claim came from two independent groups of scientists, who had come to the same conclusion. So the scepticism, while high in some quarters (including mine), never rose to the levels that we see now with OPERA.
Of course, the clincher for the accelerating universe had nothing to do with any of this. Confirmation came from additional, complementary data — from other experiments, such as those observing the cosmic microwave background radiation (the dim glow leftover from the hot Big Bang). In the end, scientific evidence spoke louder than even the most vocal scientists.
Perhaps OPERA’s result will be confirmed, perhaps not. But no one should be complaining about the current high level of scientific scepticism. Experimentalists and theorists who are pointing out potential problems with OPERA are just doing their job, forcing this experimental result, like any other, to run a gauntlet of obstacles to prove its worth. No radical claim enters into the great books of scientific knowledge without passing a series of draconian tests, of a sort that recalls the ordeals suffered in childhood fairy-tales. For this is how science, a process beset with human frailty, protects itself from harm.