Tag Archives: LUX

Why Scientists Can Be Happy Even When They Find Nothing

Appropriate for General Readership

Last week, the LUX experiment reported its results in its search for the dark matter that (speaking roughly) makes up 25% of the stuff in the universe (see here for the first report and here for some Q&A).  [See this article, specifically the “Dark Matter Underfoot” section, for some nontechnical discussion about how experiments like LUX work.]  Shortly thereafter, a number of articles in the media made a big deal out of the fact that, simultaneously,

  1. the LUX experiment did not find evidence of dark matter
  2. yet scientists at the LUX experiment appeared to be quite happy

as though this was contradictory and mystifying. Actually, if you think about it carefully, this is perfectly normal and typical, and not the slightest bit surprising. But to make sense of it, you do also have to understand the levels of “happiness” that the LUX scientists are expressing.

The point is that whenever scientists do an experiment whose goal is to look for something whose precise details aren’t known, there are two stories running simultaneously:

  1. The scientists are trying to do the best experiment that they can, in order that their search be as thorough and as expansive as it could possibly be with the equipment that they have available.
  2. The scientists are hoping that the thing that they are looking for (or perhaps something else equally or more interesting) will be within reach of their search.

Notice that humans have control over the first story. The wiser they are at designing their experiment, and the more skillful they are in carrying it out, the more effective their search will be. But they have no control over the second story. Whether their prey lies within their reach, or whether it lies far beyond, requiring the technology of the distant future, is up to nature, not humans. In short, story #1 is about skill and talent, but story #2 is about luck. Even a great experiment can’t do the impossible, and even one that doesn’t work quite as well as it was supposed to can be fortunate.

Of course, there is some interplay between the stories. A disaster in story #1 precludes a happy ending in story #2; if the experiment doesn’t work, there won’t be any discoveries! And the better is the outcome in story #1, the more probable is a success in story #2; a more thorough search is more likely to get lucky.

The LUX researchers, in order to make a discovery, have to be lucky in several ways, as I described on Thursday.

  • Dark matter (at least some of it) has to be made from particles which are heavier than protons and have uniform properties;
  • These particles have to be rather smoothly distributed through the Milky Way galaxy, rather than bound up in clumps the way ordinary matter is, so that some of them are likely, just by chance, to be passing through the earth;
  • And they have to interact with ordinary matter at a rate that is not insanely small — no less than a millionth of the interaction rate of high-energy neutrinos with ordinary matter.

None of these things is necessarily true, given what we know about dark matter from our measurements of the heavens. And if any one of them is false, no detector similar to LUX will ever find dark matter; we’ll need other methods, some of which are already under way.

Now, in this context, what’s the worst thing that could happen to a group of scientists who’ve built an experiment? The worst thing that could happen is that after spending several years preparing the experiment, they find it simply doesn’t work. This can happen! These are very difficult experiments requiring very special and remarkable techniques, and every now and then, in the history of such experiments, an unexpected problem arises that can’t be solved without a complete redesign, which is usually too expensive and in any case means years of delay. Or something just explodes and ruins the experiment. Something like this is extremely depressing and often deeply embarrassing.

So if instead the experiment works, the scientists who designed, built and ran it are of course very relieved and reasonably happy. And if, because of a combination of hard work and cleverness, it works better than they expected and as well as they could have hoped, they’re of course enormously pleased, and proud of their work!

Now what could make them happier still — even ecstatic, to the point of staying up late drinking entire bottles of champagne? A discovery, of course. Discovering what they’re looking for, or perhaps something they weren’t even looking for, if it is truly novel and of fundamental importance.  If that happens, then they won’t care as much if their experiment worked better than expected… because, if you’re an experimental scientist, there’s nothing, nothing at all, better than discovering something new about nature.

So with this perspective, I think the LUX scientists’ emotions (as conveyed during his talk by Richard Gaitskell of Brown University, the project’s leader) are actually very easy to understand. They are very happy because their experiment works better than they expected and as well as they hoped… maybe even better than that. For this, they get the high respect and admiration of their colleagues. But make no mistake: they’d certainly be a lot happier — overjoyed and humbled — if they’d discovered dark matter. For that, they’d get a place in the history books, major prizes (perhaps a Nobel, if the Nobel Committee could figure out who to give it to), lasting fame, and the almost unimaginable feeling of having uncovered something about nature that no human previously knew, and that (barring a complete collapse of civilization) will never be forgotten. So yes, they’re happy. But not nearly as happy as can be. They’re frustrated, too, just like the rest of us, that nothing’s shown up yet.

However, they’re also hopeful. Since they’ve built such a good experiment, and since they’ve only run it for such a short time so far, they’ll have another very reasonable shot at finding dark matter when they run it for about a full year, in 2014. Not only will they run it longer, they’ll surely also learn, from their experience so far, to be smarter about how they run it. So expect, at the very least, powerful new limits on dark matter from them in eighteen months or so. And maybe, just maybe, something more.