Bold claims require bold evidence


The first use of a hydrogen bubble chamber to detect neutrinos, on November 13, 1970

Scientists have reportedly clocked a subatomic particle going faster than the speed of light in a vacuum and that’s something that, according to observation and theory, should not be possible. The report originated at CERN:

Scientists at the world’s largest physics lab said Thursday they have clocked neutrinos traveling faster than light. That’s something that according to Einstein’s 1905 special theory of relativity – the famous E (equals) mc2 equation – just doesn’t happen. “The feeling that most people have is this can’t be right, this can’t be real,” said James Gillies, a spokesman for the European Organization for Nuclear Research …

That’s a healthy, skeptical attitude and my amateur guess, for what it’s worth, is this will turn out to be a false reading. Neutrinos are exceedingly difficult to detect. They’re teeny-tiny, like electrons, but have no charge and rarely interact with ordinary matter. There are scads of neutrinos streaming through your body right now and you’re none the worse for it.

Given how famously difficult neutrinos are to detect, it seems more likely that the error is not in the Theory of Relativity — the body of knowledge underlying all of macro space-time — but on the method[s] used to measure the transit time of the ghostly particles. Still, scientists will try to replicate the results and determine if the effect is legitimate or an artifact of observation.

Comments

  1. allytude says

    Isn’t this a case of

    “There was a young lady called Bright
    Who could travel much faster than light.
    She set out one day
    In a relative way
    And returned on the previous night.
    Arthur Buller in Punch, 19 Dec. 1923” replace lady with particle

  2. Robert B. says

    There really weren’t many technical details in the article, but I can’t think of any way to do this experiment that wouldn’t be fundamentally statistical. I don’t know of any neutrino source that you can aim in a single direction, you can’t see a neutrino until it rams into something, and any given detector (where by “detector” we mean “glacier” or “building-sized jar of cleaning fluid”) may or may not see any given neutrino – and probably won’t. Plus the sun is always shooting neutrinos through the planet. So basically they must have turned a big neutrino source on, and watched for a spike in neutrino detection nearby.

    They gave a margin of error, which is nice, but I’d also like to see a calculated probability that natural sources could have produced a false positive at just the wrong time. I would think, since the whole experiment is statistical, that there’s a non-zero chance that an unlikely string of “lucky” hits on solar neutrinos could have happened right after the beam turned on. In fact… a longer time would be easier to measure, so I can’t think of a reason to use a detector in Italy, instead of the US or Japan or Antarctica, unless you were worried about the distance spreading your neutrinos out too much to see a clear spike. That means they must be operating pretty close to the edge of their sensitivity.

    For that matter, if you look at historical measurements of the speed of light itself, it took them quite a long time to get the right answer inside their error bars. “Margin of error” means “what we’re pretty sure is the margin of all the errors we’ve thought of so far.” And… +/- 10 ns on a 2 ms measurement would be, um, a very impressively small margin of error for an experiment that takes place over 700 km. I don’t say it couldn’t be done, but wow. On those scales you might have to check the relativistic effects of the earth’s motion between different points on the surface.

  3. tbrandt says

    I’d bet a lot of money that this is wrong. A few articles have already pointed out the Supernova 1987A problem–those neutrinos were detected hours before the first photons (the time it took for the shock to propagate through the star). At the distance to SN1987A, and with the neutrino velocity claimed in this experiment, the delay would have been years.

    Of course, there are also very compelling theoretical reasons to disbelieve the results, and the experimentalists aren’t claiming anything. They just can’t find their mistake and are looking for help.

  4. says

    Particle physicists don’t make announcements like this after a single reading. So It is highly unlikely to be a “false reading”

    In particle physics its not a real reading until you have viewed exactly the same event thousands of times and the readings become so predictable you can be confident down to 5 sigma. Thats 99.9999426697% confident. I read that this neutrino event was observed 16,000 times.To be wrong about that many readings (a “false” reading) requires an epic level of delusion.

    Although please don’t get me wrong, their interpretation of the data could be faulty. but the data itself is probably sound. Neutrinos definitely arrived 60 nanosecond earlier than they should have.

    Also, these aren’t just any neutrino, they were generated and measured in the collisions happening at CERN and then detected again in a dedicated facility 730 kilometers away. Which means that when they left CERN the neutrino direction and speed was KNOWN with unbelievable accuracy. Only neutrinos matching the exact same direction and speed would be counted as a “hit” at the other end. Thats how you know that you have the same neutrino.

  5. says

    As I thought might happen, good scientists would never announce such a bold claim unless they were really sure about what they saw. SIX sigma sure it turns out. The bold evidence that this blog post title demands is starting to appear with every extra detail released.

    http://www.newscientist.com/article/dn20961-fasterthanlight-neutrino-claim-bolstered.html

    The scientists responsible have already begun to turn some of the skeptics (New Scientist Mag was one).

    Again. This does not prove faster than light travel yet. Its just data. Now we wait first for someone to repeat the data, then for someone to interpret. If the experiment is repeatable, shit is is gonna get interesting.

    This is going to be good :-)

Leave a Reply