I don’t think naita’s post addresses that:
The possibility that two groups of neutrinos arrived, one group FTL and one group slower than light, both originating from the same supernova event, wasn’t mentioned there.
I haven’t seen any evidence of two groups of neutrinos in the measurements just reported, so that experiment would have to be producing only the FTL group.
Addendum: It’s possible that physics as we know it will not really be totally destroyed by a confirming result: the neutrinos could be taking a shortcut through the bulk (extra dimensions).
This is getting interesting. I attempted to listen to the webcast, but was interrupted too many times. What I picked up was inconsequential, or beyond my scope.
I expected to see some alternate theories come up in this thread, like photons aren’t quite moving at c but neutrinos are, and that sort of thing. But it’s clear that actually exceeding the speed of light (whether light is that fast or not) is being described as impossible. There have been plenty of science fiction-y ideas about this subject, but the science fact-y side seems to be missing.
So now I’m wondering if these guys were testing to simply measure the speed of neutrinos, or were measuring to see if neutrinos exceeded the speed of light. If so, I’m naturally skeptical of any test that produces a desired result. Does anyone know the hypothesis being tested here?
My own thoughts is that neutrinos are such slippery customers and relatively poorly understood compared to many of the other elementary particles that observations about their speed are unlikely to constitute a smoking gun against a theory such as special relativity, which has a wealth of experimental evidence in it’s favour.
I’m not rejecting the possibilties out of hand, but for me it seems far more likely to be a result of a systematic error in either the experimental set-up or the model used to estimate the speed of the neutrinos. Itmay that this result does prove that a bit of physics we thoguht was right was wrong, but I think again it’smore likely perhaps to do with our understanding of say protons than relativity.
Again I’m not rejecting anything out of hand, but I think it’s best to approach the result and it’spossible consequences with caution.
A quick aside on this. The answer to A is no. Einstein’s equations show that all matter has momentum, not mass. Massless particles travel at exactly the speed of light (in a proper vacuum) and massed particles never do. That’s exactly why the notion of a massed particle like a neutrino traveling faster than a massless photon is so unsettling. So B is completely wrong and so are what follow.
Can’t we just keep calling it a tachyon?
No one was looking for evidence of a supernova before it could be observed visually, but we did have detectors capable of detecting neutrinos operating before the light from 1987a arrived on earth. No mysterious, unexplainable burst of neutrinos was observed before the light got here.
The purpose of the experiment was to detect the tau neutrino, which they expected to appear as a result of neutrino oscillations (muon neutrinos are produced at CERN, some of which become tau on the way to Italy). The velocity measurement is a fringe benefit.
But again, neutrinos are thought to oscillate, so if one flavour goes faster than c, all neutrinos should spend some time moving faster than c.
But why would there be a shortcut between Switzerland and Italy, but not between 1987A and our solar system?
The mention of which – something along the lines of ‘…and we’re also still looking for the taus…’ – got a bit of a laugh at the Q&A after the talk.
http://news.sciencemag.org/sciencenow/2011/09/neutrinos-travel-faster-than-lig.html
straightforward (for him) maybe, but it’s not straight, like a 730 km long tube in deep space straight, right? Am I correct in thinking that the fact that the emitter and the detector are in different frames of reference affects the calculations? They are attached to different parts of the earth and spinning at different speeds. Can someone who knows in detail about the paper/results confirm that this was accounted for?
The faster than light neutrinos wouldn’t have to be moving too much faster than light for them to arrive before we ever cranked up our detectors (though faster than the neutrinos these guys detected). Also, it could be when faster than light neutrinos are produced, they aren’t at one fixed speed, they are at a large range of speeds so they don’t arrive 100,000 light years later in a burst but instead the burst is now spread out over many years/centuries and is like a low or very low level background component.
Yes, those explanations don’t quit mesh with these guys results, but then again they are operating a piddly particle accelerator and not a supernova, so the neutrinos produced may have slightly different energies/properties.
Yes, but an alternative explanation for the behaviors mathematically described as oscillation other than oscillation, is not overturning fundamental understandings, just overturning one small model to describe behavior. It may, for example, turn out that what is described as “oscillation” is merely the same particle reoreinting in n-dimensional space and taking on different extended-dimension space characteristics as it does … or that neutrinos can oscillate through all neutrino flavors and tachyon-neutrinos through all tachyon-neutrino flavors, but that both are restricted from crossing over c and into the other flavors. Replacing or refining the oscillation paradigm would also be exciting but not … buggering.
Someone earlier in the thread mentioned quantum tunneling. Tunneling, like everything else we know of, is still limited by c. It’s not, despite the common misconception, instantaneous. The important thing about tunneling is that the particle makes it from one side of an “impenetrable” barrier to another, not how quickly it happens. It would be essentially irrelevant in this experiment, since a few thousand kilometers of rock presents effectively no barrier at all to neutrinos.
Extreme experimental claims such as this one come in roughly three varieties. First, the experimental methods can be so dodgy and unconvincing that you just stop paying attention. More commonly, the measurement is possible using the methods at hand, but the experimenters have to be careful in a lot of subtle ways. Finally, an experiment can be so clever in its design that the data speak for themselves, and the measurement relies at a bare minimum on getting anything right or wrong (typically because there are multiple, independent ways that the underlying physics manifests itself).
This new time-of-flight result is in the middle category. The experimenters have been careful, and they have presented the results as clearly as can be expected. However, many critical components of the methods and analysis have no safety nets. Any bias in either the distance or time measurement would appear in the final result without any indication of anything having gone awry.
Having read the paper and seen the presentation, I cannot point to something and say definitely, “You did that wrong,” but I can point to numerous things that are tricky and that I am not entirely convinced of just yet. This isn’t to say that they’ve messed up. Rather, it is to say that the methods are complex enough that no outsider will ever be able to assess whether they’ve done everything correctly. Thus, a completely independent measurement will be required.
(One item that I felt they could have easily given more quantitative treatment to is the final fit of the proton time distribution to the neutrino event time distribution. I have several questions about their methods here. Again, this isn’t to say they’ve done something wrong. It is to say that I’m not yet convinced that they haven’t.)
This was accounted for (and is a negligible effect).
Has their means of measurement demonstrated validity with anything else? Do they measure photons as moving at the speed of light? Or anything else? I’m beginning to wonder if this anything but a case of Tabloid Physics.
Sure, but the initial question was about our detection capability. It’s not like there was a switch pulled a couple years before 1987a. Some type of neutrino detection had been going on for a couple decades before the supernova.
Maybe we didn’t see a burst, but we also did not see a decreased baseline level of detection AFTER the supernova, either, which you would expect if the baseline level was elevated before the light of the explosion reached Earth.
If the supernova threw off faster than light neutrinos, they were at a level low enough to be undetectable by the point they reached earth.
People keep talking about the possibility of the neutrinos reaching the detector before they’ve left the generator. If that happens, then what happens if, after you detect the neutrinos, you decide not to generate the ones you’ve detected?
The short answer is “no”, but that doesn’t make it tabloid physics. That is, the experimenters are legit and they have executed a careful measurement. The trouble is that, as you point out, the experimental set up doesn’t permit a clear cross check, so you have to trust “dead reckoning” on the long string of difficult steps involved. One the one hand, it makes the measurement all that much more impressive, but on the other hand, that’s not what you’re looking for in a experimental set up. Nonetheless, there are many good physics results of this dead-reckoning sort, but they weren’t necessarily believed until they were independently verified. (And, to be sure, they have quite often been refuted.)
One way this measurement could have been easier is if they had used a second (possibly smaller) neutrino detector near the neutrino production point, reading it out with the same electronics and GPS timing system used at the downstream location. As it is, they measure the time difference between the primary proton pulse and the recorded neutrino events, which necessitates the tallying of numerous large offsets. Measuring from neutrino-detector to neutrino-detector would allow essentially all timing offsets to cancel out perfectly.
I bring this one item up because it is an example of how money and science play together. I have no doubt that someone on the OPERA collaboration thought to do exactly this – i.e., construct an upstream detector to establish a “start time”. However, the main goal of OPERA is to observe the appearance of tau neutrinos, and an upstream detector is not helpful for that measurement. So, one would have to argue that a search for a violation of Lorentz invariance was worth the construction of a whole new detector, at the cost of probably $5M-$30M I’d guess, depending on how much civil construction would be involved. Obviously, that would never fly. So, they are forced to make this time-of-flight measurement the way they have, through careful calibration of all the distances and time-offsets involved, and through the study of biases introduced when comparing proton and neutrino timing distributions (rather than comparing neutrinos and neutrinos).