To start, let me say that there are extremely strong reasons to believe that the OPERA experiment’s measurement of neutrinos travelling faster than light is flawed. We knew that from the moment it came out, because it contradicts General Relativity (GR), which is an extraordinarily well-tested theory. Not only that, but the most obvious ways to modify GR to allow the result to be true give you immediate problems that contradict other measurements. To my knowledge, there’s no complete theoretical framework that makes predictions consistent with existing tests of GR and allows the OPERA result to be right.
But in my view of how experimental physics is done, history has shown us that once in a great while, something is discovered that nobody thought of and nobody can fit into the existing theoretical mold. The measurements that led to the discovery of GR in the first place provide a good example of this. Such shifts are extremely rare, but I don’t like the idea of ignoring a result because it doesn’t fit with the theories we have.
No, we have to address the measurement itself, and satisfy ourselves that there really was a mistake. There are many ideas for what might have gone wrong, and as far as I know, the discussion is ongoing. I’m not an expert on it, but I know enough to disagree with some of the blogosphere discussion lately that has pronounced that the case is closed. There seem to be two categories of claims going around:
- Articles that point out that the OPERA result is inconsistent with other measurements, as in this piece by Tommaso Dorigo (who is, incidentally, my colleague now that I’ve joined CMS). These are of course correct within the context of GR or any straightforward modifications thereof, as I said right at the start of this post. The question is whether there’s some modification that can accomodate the results consistently, and that’s a very hard thing to exclude. (There is some good discussion in the comments of Tommaso’s post about this, in fact.)
- Articles that the OPERA result has been refuted because someone posted an idea on the arXiv server. A current example is this preprint, which asserts that a 60 nanosecond delay might be explained by OPERA having made a relatively trivial mistake in their GPS calculations. Of course, it’s possible that a trivial mistake has been made. But I’m not inclined to consider it definitive, especially because the author has already partially backpedaled upon learning more about how GPS works.
It’s great that people are sending ideas for what might have gone wrong with the result, or how it might be explained. But let’s wait for the discussion to settle down — and, indeed, for OPERA to finalize their paper — before we conclude that the case is closed. I do expect the result to be disproven, but what I want to see is one of these things:
- OPERA finds that there really was a problem with their measurement, revises it, and the “superluminal” effect goes away.
- Another experiment makes the same measurement, and gets a result consistent with GR.
Either way, I’ll consider the case closed, but there’s no reason to get ahead of ourselves. Doing science usually doesn’t mean knowing the answer in time for tomorrow’s news.
Tags: faster than light, neutrinos, OPERA, science
