Faster-than-light travel discovered? Slow down, folks

By Phil Plait | September 22, 2011 3:27 pm

So the web is buzzing right now over news that scientists have detected some subatomic particles moving faster than the speed of light.

Yeah, well, not so fast. Let’s think about this for a sec.

First, what happened is that they create these particles, called neutrinos, at CERN in Geneva. Neutrinos don’t interact with normal matter well, so they can pass right through the Earth as if it isn’t there. In a fraction of a second, some of them enter a detector called OPERA in Italy where they are recorded (pictured here). If you divide the distance between those two points by the time it takes for the neutrinos to travel, you get their speed.

And when the scientists did that, they find the neutrinos get to Italy about 60 nanoseconds faster than a photon would.

Photons travel at the speed of light — they are light! — so this means the neutrinos, if this is all true, traveled faster than light speed. Holy Kessel Run!

But is it true?

Now first off, if it were this would overturn so much physics that they may as well have discovered that gravity pushes, not pulls. So right away we need to treat this claim with lots and lots of skepticism. I’ll note these are actual particle physicists making this claim, and not some crackpots who will shake their fists at the sky and say how Galileo was laughed at too.

The thing to do is to look at where this claim might have gone awry. First, the timing is interesting. They claim a measuring accuracy of 10 nanoseconds, so 60 ns would be pretty significant. However, my first thought is that light travels about 30 centimeters in 1 ns, so they need to know the distance between the source and the detector to an accuracy of 3 meters. If they are off by 20 meters, then we’re done; that would explain the difference entirely. I suppose this depends on how they measured the distance and the speed of the particles, too. However, they haven’t published a paper on this just yet, so that’ll have to wait.

[UPDATE: The paper is now up on the arxiv preprint server. I took a look, and must say at first glance their reasoning looks solid. They appear to have the baseline distance nailed and the timing as well. However, the devil’s in the details, and this isn’t my field, so I’ll be very curious to see how the pros in this discipline react to the paper.]

Also, as pointed out in a Science Magazine article, knowing the exact moment the neutrinos are created isn’t easy either. Mind you, 60 nanoseconds is 0.00000006 seconds, so they need a pretty good clock here. That page also says they used GPS to determine the distance, which could be off a bit.

There’s another point that actually is quite important here. If neutrinos travel faster than light, then we should’ve detected the neutrinos from Supernova 1987A before we saw the explosion itself. That exploding star was formed when the core of a massive star collapsed, detonating the outer layers. The collapsing core blasted out a furious wave of neutrinos strong enough to be seen here on Earth, over 160,000 light years away.

The distance from the detector in Italy to the source in Geneva is about 730 km. The travel time at the speed of light is about 2.43 milliseconds, and the neutrinos appear to have outraced that speed by 60 nanoseconds. If true, that means they were traveling just a scosh faster than light, by about 1 part in 40,000. The neutrinos from SN1987A traveled so far that had they been moving that much faster than light, they would’ve arrived here almost four years before the light did. However, we saw the light from the supernova at roughly the same time as the neutrinos (actually the light did get here later, but it takes a little while for the explosion to eat its way out of the star’s core to its surface, and that delay completely accounts for the lag seen).

But I wouldn’t use that argument too strongly; perhaps this experiment creates neutrinos in a different way, or the neutrinos from this new experiment have different energies than ones created in the cores of supernovae (a good bet). Still, it’s enough to make me even more skeptical of this FTL claim.

I’ll note that the scientists will be presenting these results tomorrow at a conference at CERN. We’ll learn more then. It’s not clear to me if these results are being published, or have been peer-reviewed, or what. As usual, we’ll need to have other scientists either confirm this result using other equipment, or show where things went wrong. That’s how science works. And the scientists involved are asking for criticism here! That’s just so; incredible results need to be tested incredibly well.

So don’t let your imagination run away with this just yet. This result will, in my opinion, probably turn out to be incorrect for some reasons dealing with measurement. Faster than light travel is still a dream, even though I wouldn’t say it’s impossible… just very, very, very, very unlikely.

Maybe someday we’ll boldly go. But for now, I’m not betting my dilithium on it.

Image credit: OPERA; NASA/ESA/Hubble


Comments (350)

Links to this Post

  1. Faster-than-light travel discovered? Slow down, folks | Bad Astronomy | Today Yoga Blog | September 22, 2011
  2. Warp Speed and Time Travel to Past and Future might now be possible! - Baha'i Forums | September 22, 2011
  3. Neutrino discovery could prove existence of Yngwie Malmsteen | September 22, 2011
  4. Strange Particles May Travel Faster Than Light – Fox News | Tucson AZ News and Weather | September 22, 2011
  5. Neutrinos más rápidos que la luz « Círculo Escéptico Argentino | September 22, 2011
  6. Quora | September 22, 2011
  7. Ask a physicist anything. (6) - Page 98 - Christian Forums | September 22, 2011
  8. Faster Than Light Particles! So, Warp Speed Ahead, Right??? « A Quantum of Knowledge | September 22, 2011
  9. Let’s Go Slow on CERN Faster-Than-Light Claims « The Skeptical Teacher | September 22, 2011
  10. Faster-than-light travel discovered? Slow down, folks | Bad Astronomy | America Society | September 22, 2011
  11. Mundane Neutrino Explanations | Science Comedian | September 22, 2011
  12. Faster Than Light Particles! So, Warp Speed Ahead, Right??? | A Quantum Of Knowledge | September 22, 2011
  13. If it wasn’t for those pesky neutrinos | cartesian product | September 22, 2011
  14. Faster than Light Travel « The Gaming Atheist | September 22, 2011
  15. Vad kan färdas snabbare än ljusets hastighet? « vemvadhurvarfor | September 23, 2011
  16. OPERA och de alltför snabba neutrinerna | Stjärnstoft och kugghjul | September 23, 2011
  17. Those tachy neutrinos « Lost in Science | September 23, 2011
  18. CERN claims faster-than-light particle measured | Kajunman's Blog | September 23, 2011
  19. World’s Strangest | Neutrinos Beat the Theory of Relativity, and Light | September 23, 2011
  20. Faster Than Light Neutrinos? Don’t Bet On It. « News Hub Today | September 23, 2011
  21. FTL Neutrinos? Not so fast…. « Whiskey…Tango…Foxtrot? | September 23, 2011
  22. Please be true: Beating cancer, and the speed of light | bluejay's way | September 23, 2011
  23. Assorted links — Marginal Revolution | September 23, 2011
  24. Faster-Than-Light Neutrinos? | Richer Ramblings | September 23, 2011
  25. More news about the FTL neutrino observations | Entangled States | September 23, 2011
  26. The neutrino velocity thing « Econstudentlog | September 23, 2011
  27. BookBanter Links Roundup For 09/22/11 « The BookBanter Blog | September 23, 2011
  28. Can the Speed of Light Law Be Broken? – About | Alkaon Network | September 23, 2011
  29. SN1987a - Page 3 - Christian Forums | September 23, 2011
  30. So Was Einstein Wrong? « Victorian Skeptics | September 23, 2011
  31. Naughty ‘Faster Than Light’ Neutrinos a Reality? | wpgrabber | September 24, 2011
  32. Faster-than-light neutrinos show science in action » | September 24, 2011
  33. NOT “faster-than-light” but still exciting « WhizDumb | September 24, 2011
  34. Naughty ‘Faster Than Light’ Neutrinos a Reality? | Healthy Living | September 24, 2011
  35. It’s too early to take a stand. « Psi-Pi(Ψ-π) | September 24, 2011
  36. Relatively funny « the.soft.anonymous | September 24, 2011
  37. Why the Faster-than-Light Neutrino Experiment May Be Wrong « adafruit industries blog | September 25, 2011
  38. Science Sunday #15 « Radio Freethinker | September 25, 2011
  39. Faster-than-light travel discovered? Slow down, folks | Bad Astronomy | Discover Magazine | snarkarchive | September 25, 2011
  40. faster than light? - Page 7 - PriusChat Forums | September 25, 2011
  41. - Was Einstein Wrong? U.S. Accelerator Lab To Test CERN's Results | September 26, 2011
  42. Ye Olde Soapbox | September 26, 2011
  43. Riskbloggen » Den ena vetenskapliga sensationen efter den andra? | September 27, 2011
  44. RelativKritisch Schwerpunkt: Neutrinos schneller als Tempolimit c0? | RelativKritisch | September 27, 2011
  45. Funny Neutrino Jokes | Guru: A Magazine for Ipad, Smartphone, e-Reader and more! | September 28, 2011
  46. links for 10/07/2011 « Alan Vonlanthen's blog | October 6, 2011
  47. Why is the Universe is Expanding and Accelerating? Here’s my pet theory… (by a non-physicist) « Doctor Stu's Blog | October 7, 2011
  48. Saltklypa #27 – Om rånete nøytroner og punkterte lunger | Saltklypa | October 7, 2011
  49. I’ll be back…to the future | gee/k/ink | December 8, 2011
  50. Note til journalister: Videnskab er sjældent revolutionær | Hennings blog | February 23, 2012
  51. Science Roundup: Dawkins, Doubt, and Probability « Awaiting A White Robe | February 28, 2012
  52. Can neutrinos travel faster than light? Part Two « Skeptical Kids | April 29, 2012
  53. Supernovae | October 4, 2012
  1. Adam

    I do also like that they stated that they have checked their work several times and now want other scientists to go over it and see what they can find. This seems to be following legitimate scientific processes which lends a lot of credence to the scientists presenting the information. It will be fun to follow.

  2. Jess Tauber

    O ye of little neutrality! I heard about these claims several minutes before they made them!

  3. I’m willing to bet a large amount that this will turn out to be untrue. (Some sort of localized relativistic space contraction or time dilation, perhaps?)

    That said, I’d wish fervently that I lose that bet.

  4. Jeff Johnson

    The #mundaneneutrinoexplanations hashtag has been great entertainment on this.

  5. Rob P.

    My understanding was that this had not been peer reviewed, but rather was considered to be so unusual and interesting that they wanted to get it out into the world to be picked apart as quickly as possible.

  6. Joe

    Just out of curiousity, I wonder what the implications are if this turns out to be true? What did we previously believe to be “impossible” that would then be “possible”

  7. Ibeechu

    The way I read the initial announcement was that it’s not that all of their neutrinos were going faster than light, but that there was a probability involved on whether or not any given neutrino would go faster than light. If that’s the case, then maybe we recorded only a certain number of neutrinos from that supernova, and recorded the rest of them four years earlier.

    But that’s getting into kind of “conspiracy theory” territory, and I might have completely misinterpreted the original articles.

  8. Patrick

    Thanks for the explanation. I saw an article that acknowledged the margin of error, but it pointed out that the measurements were far greater than that margin. However, you make a great point in suggesting other causes of measurement error.

    I’m asking from a position of near-ignorance (I have read a couple of books…), but light can be slowed down by the medium it travels through, right? If neutrinos can pass by matter, is it possible that the photons were being slowed below the speed of light in a vacuum and the neutrinos sped past them? Or am I way off because I don’t really know what I’m talking about?

  9. Kieron

    Well light can be slowed down by matter, right? so maybe even in the vacuum of space it is slowed down slightly, while neutrinos which don’t interact with matter much at all wouldn’t be slowed down at all.

    However I still think this is likely just a bad study.

  10. Douglas Troy

    This story is sooooo 60ns ago.


    In all seriousness, this is great stuff and I look forward to what others find and have to say about it.

  11. Mimi c

    So you mean to say you are skeptical of particle physisicsts (who probably thought of all your concerns already) who SAT on their findings for MONTHS because they too were trying to see if there was something that was incorrect in their work. Your point about the supernova is already incorrect because the scientists say they “fired” those particles, and laboratory accelerated particles are already different than particles which may have been affected by a giant mass next to it.

  12. Pete Jackson

    Very nice analysis Phil!

    One easy way to remember the speed of light for discussing experiments like this is that it’s about “one foot per nanosecond”.

  13. Renee Marie Jones

    My guess is that there’s a 60 nanosecond error somewhere. On the other hand, general relativity does not actually prohibit faster-than-light travel, though it’s hard to believe that Switzerland and Italy are not really, really close to being in the same inertial reference frame.

    On the *other hand* (man, that’s a lot of hands!) we know for a fact that there’s something wrong with general relavitity and/or quantum mechanics as well as our understanding of gravity. So, maybe nature is trying to give us some clues here.

  14. Brian Hurt

    Wait- they used flipping *GPS* to determine the distance?

    Two seconds with google turned up this page: in which Garmin claims GPS is accurate to +/- 15 meters. Say they’re using some super-accurate version of GPS, which is an order of magnitude more accurate that Garmin’s. So each location is only known to +/- 1.5 meters- and the combined error for both sites could then easily be 3 meters. There’s your two meters right there.

  15. OK, not being a particle physicist, and not having read anything but the news reports, I wonder:

    Do we know how much, if at all, the speed of neutrinos is affected by passing through air?

    We know the refractive index of air for visible light is about 1.0003 (according to Wikipedia), and the speed of light in air is less than that in a vacuum.

    Is the index of refraction of the atmosphere the same for photons, which interact electromagnetically, as it is for neutrinos, which do not?

    Surely they accounted for this, and it can’t be that simple, right?

    In regard to comment 8, Relativity doesn’t expressly prohibit faster than light speeds, it prohibits anything with mass traveling at the speed of light, which prevents anything with mass traveling less than the speed of light transitioning through c to achieve superluminal velocity. Various mathematical absurdities, make superluminal velocities of massive particles (tachyons, etc) unlikely and unworkable.

  16. I wonder if they accounted for this:

    The speed of light is different in different transparent mediums, air, water ,glass. The mediums the supernova neutrinos were going through was mostly vacuum. The medium that these neutrinos were going through was air and rock.

    I realize that neutrinos go through every as if its not there, but maybe that is an oversimplification that accounts for 60 ns.

  17. This is pretty entertaining. For what it’s worth, I vividly remember a physics colloquium at Stanford sometime in the late ’90s where the speaker put up a plot of some parameter. The plot was a straight line, and the y-intercept was supposed to be mass^2 for the neutrinos in that experiment. The intercept was clearly negative, though within the error bars. Someone asked about this, and the speaker (can’t remember who this was) basically said that with better stats this would almost certainly go away. Regarding the comment above about GPS, I’m sure that these folks are doing better than simple commercial-grade GPS. You can do differential GPS and get accuracies down to 10 cm.

  18. MadScientist

    Meh; I’ve seen the effects of particles traveling faster than light (not in vacuum). It produces such a beautiful glow.

    I’ll have to see much more information than what’s in the news to even convince me that the observation is anything new.

  19. BJN

    Science in action. Here’s what we think we’ve found and we can’t see where we’re wrong, please help us shoot holes in it.

  20. Brian, GPS can be very accurate if you average over a long time, like a week. Even more accurate if you kncorporate differntial GPS with a transponder located at a known location. centimeter accuracy or better.

  21. Chris

    Yeah, well, not so fast.
    Pun intended?? :-)

    @14 Brian
    I’m sure when they were doing this they used something more accurate than a Garmin GPS! I’m sure you’ve seen the high resolution movement of glaciers and volcanoes where they can see mm of movement, so they can get very high accuracies. Just for commerical uses, a few meters of accuracy is pretty good.

    Maybe the neutrinos aren’t traveling faster than light, there is a wormhole in the earth between Italy and Geneva!

  22. Tensorguy

    I totally agree with your assessment. I remember getting very excited in the 90’s when two experimenters claimed that their electrolytic cell was producing more energy than it should and that the only solution was, well, er, nuclear in nature. I don’t need to go into what a debacle that was when results could not be replicated. This is however different. It would seem that there is a fundamental flaw in the neutrino experiment as a whole or that as of yet unrecorded phenomena have been observed.
    Things going faster than light are not new. Cerenkov radiation is produced when particles move faster than light within a medium(much lower than the speed of light in vacuo, c) . There are recorded quantum tunneling effects in which subatomic particles appear to simultaneously bridge a gap. Similarly there are reasons to believe that space-time itself can expand or contract faster than light and that photons or electrons or bicycles traveling on that bit of spacetime would appear to move faster than c to observers. Finally there are the as yet theoretical tachyons , particles which Only travel faster than light.
    We must be prepared to deal with these phenomena at some point even though we have not seen them yet. The same occurred with planets and black holes and even neutrinos themselves. We theorized they existed and it was years before they were observed. We get too worked up over the great truths. Einstein clearly showed how nothing can travel faster than light and thus clocks and yardsticks dilate and contract to make this statement true and yet this was only under the special case. That’s why it was named the Special Theory of Relativity. Both Quantum Mechanics and General Relativity provide with theoretical exception to this law of physics. Einstein himself fought against the Accepted Scientific Truth that the Luminiferous Eather was all encompassing and had relative motion to the rotation of the Earth. On what side do you think he would be now?
    Again, this might be just an error but it opens our minds and hearts to the possibilities and I like to think that there always are, posibilities…

  23. Sean H.

    Ibeechu: I think if it were the case that some of the neutrinos traveled faster than light and others traveled at the speed of light then there would possibly be others that fell in between those two speeds and there would be a detectable distribution likely peaking with the arrival of the light.

    Upon seeing this article making the rounds today my first thought was “interesting, but I need more data. How are these neutrinos distributed? Is the beam highly directional similar to a laser or more like a light bulb? What is the density of the beam, how many neutrinos per second? How do they determine the departure and arrival time of specific neutrinos? What are the implications if it does hold that under certain conditions a subatomic particle with a non-zero rest-mass can travel faster than a photon which has zero rest-mass?” My second thought was “What will Phil Plait have to say about this?”

  24. Sorter

    Link to the live-feed for tomorrow Friday, September 23, 2011 from 16:00 to 18:00 (Europe/Zurich)

  25. If neutrinos travel faster than light, then we should’ve detected the neutrinos from Supernova 1987A before we saw the explosion itself

    We did detect those before the light, by three hours. But that’s because the light is produced at the end of the process with a Type II supernova whereas the neutrino burst occurs towards the beginning. That’s the point also of the Supernova Early Warning System to let astronomers know if a close supernova is about to have the light arrive. So this did happen, we already knew it, and it doesn’t tell us much.

    The more interesting issue is that a few hours before the SN 1987A neutrino burst the Mont Blanc detector detected a burst of neutrinos. This has generally been considered to be disconnected from the SN 1987A event and simply be a statistical fluke. These results raise the possibility that some of the neutrinos traveled faster than the speed of light and those were the ones detected by Mont Blanc. This seems problematic though because why then would only Mont Blanc have detected them?

    So overall, probably still an error. But if it isn’t could be quite neat.

  26. I think it’s a mistake to harp on the scale of the numbers involved. They’re not easy numbers to achieve, but they’re not on the boundaries of believable. 60ns is on the scale that many a hardware geek has needed to measure, and the right carrier phase GPS hardware with suitable dwell time (which, admittedly, is measured in weeks) could get you to within a few cm. There could totally be other systemic issues, and it’s not easy to measure numbers like that, but it’s not like the measurement scales instantly jump out as “that’s impossible” territory.

    (To Brian’s earlier GPS comment: Since Garmin wrote that there has been a lot of advancement in GPS, 95% of samples in 3 meters is now a totally reasonable number for a sub $100 tracker in a pocket, and carrier-phase enhanced real-time corrected differential GPS in surveying systems is routinely used for accuracies of less than an inch).

  27. Dan M.

    Sorry – just had to chime in on the GPS error issue – we use a fairly “off the shelf” real time kinematic GPS (RTK GPS) in our surveying work and it easily does centimetre/sub-centimetre accuracy while moving around (not on point averaging), and I am fairly sure there are better GPS solutions than the cheapie that we have. So I doubt the GPS would be the source of the error. Hope that helps.

  28. Lewis

    Having a story like this in the mainstream press is a great opportunity to teach people about the scientific method and how it works. I know exactly what I’m doing with my scouts next week 😛

  29. Sion

    There is something called scale factor to take into consideration when using professional-level GPS. GPS works on a curved surface but the neutrinos and photons will travel in a straight line. The two measurements could be out by a significant amount if they have failed to factor this in to the distance measurement. In the UK it works out at about a 40cm per km discrepancy in the two measurements.

  30. It is cutting edge news. Because how can we know that nothing can move faster than a photon? Most photons we observe are from one and the same source. We have little information about photons’ behavior, for instance, when they are confronted in an environment of many photons that move in a different direction. The movements now seen of photons that moved along a star, located in-between the origin and us, show signs of having been changed, and this is now proclaimed to be caused by gravity. But that is too simple an answer for which we do not have enough data. What about the other photons in perpendicular direction coming from the star in-between? Did that slow the distant photons?

    The central question may be: Does a photon from a distant star reach the surface of another star? I believe the answer is always no, and propose the photon’s path will be bent slightly due to photons from the in-between star coming straight at it in much larger quantities and pushed aside by photons with outward directions existing in much larger quantities in that in-between star environment as well. We see the result of the distant photons bent around the in-between star by the enormous amount of photons that do not come in our direction (we do not see them) — and not because of gravity.

    And how do we measure the speed of extrasolar photons? Are they travelling at the same speed when all we have is their arrival time? Though we should expect them to all move within the same ball park of time, how do we know all photons in the universe move indeed at the same exact speed ? Then, 60 nanoseconds is nothing to be amazed about. The neutrino may have gotten some push from our natural world (but in a different manner than mentioned in the article, and yes, I am awaiting verification of all input, too).

  31. John

    This hit the mainstream news. I am very worried that if it turns out to be erroneous, they will not report it and there will be millions of people walking around saying “Einstein was wrong “(full stop).

    : (

  32. James

    Surely, the Earth moved during the experiment …changing the distance between the two points, slightly.

    Hell, even the solar system and the galaxy would have moved a little… And if these particles are unaffected by matter and already traveling at the speed of light, you can’t add the speed of the Earth to their speed. Like you would if you were running down a moving train.

    So if the detector was moving back towards the incoming neutrino, it would make it seem like it was traveling faster than it actually was.

    Basically they are trying to accurately measure the distance relative to the Earth… But the neutrinos aren’t relative to the Earth.

  33. AJ

    Ugh, much of the media-reporting on this find is excruciating.

    They even quote the researchers’ warnings that we shouldn’t get carried away at this point, for example Reuters quotes Antonio Ereditato: “for the moment, everybody should be very prudent.” In another article, John Ellis is quoted as saying, “This would be such a sensational discovery if it were true that one has to treat it extremely carefully,” and even alludes to the supernova discrepancy that Phil points out.

    Yet the writers take no heed andstart immediately jumping to sensational conclusions. It’s sad.

  34. It all boils down to one thing: Einstein is god and he cannot be proven wrong. We know he is wrong, but no one wants to say the emperor has no clothes.

  35. PeteC

    Count me in in that group that suspects strongly that they’ll find some fault in equipment, components or computation (or, indeed something forehead-slappingly doh! like forgetting that the Earth is curved when they assumed a straight line distance from the gps measurement – which come to think about it would mean they thought the distance was further and the neutrinos faster than expected).

    Also count me in that group that really, really hopes they’re right. For all the dedicated and hard work done, we’ve not really seen a proper breakthrough in physics – not one that’s been experimentally shown to be true rather than string-theory philosophising – for the better part of a century now. Plus I want my warp drive.

  36. SLC

    Re Sion @ #29

    Actually, due to the curvature of space in the neighborhood of the earth, photons and neutrinos don’t travel in an exact straight line in its vicinity.

  37. SLC

    Re Tensorguy @ #24

    Tachyons also have imaginary mass.

  38. Bill

    I like your article, but I am taken back by your suggestion that gravity pulls and to think otherwise is wrong and against fundamental laws of physics. Actually, I don’t believe that gravity pulls us. I think push is a more accurate concept. A heavenly body such as the earth displaces the fabric of space-time, and the result is an effect of tension against the surface of the earth. Matter is pushed down in an elastic-like effect. The tautness of curved space-time holds us down. There are many reputable names in physics that think the same way. So your quip is off I think.

  39. BachFan

    @4 Jeff Johnson

    D*mn you, I’m going to be following that #mundaneneutrinoexplanations hashtag and giggling all evening!

  40. Sion

    Well, yes, but that error would be swamped in scale by the one I described, so I didn’t think it was worth muddying an already complex issue. Sorry.

  41. Bill Nettles

    These must be arsenic eating neutrinos from California. That’s the only way they can get their energy high enough.

  42. Booie

    Could this be explained by the speed of light in air being lower than in a vacuum whereas the speed of neutrinos should not vary.Also when coming from a supernova both would be travelling in a near vacuum most of the way.

  43. david

    why don’t you look at Firmilabs study that was done a couple years ago. Pretty much the exact same test and had the same results but wrote it off because they didn’t believe their own findings.

  44. Jess Tauber

    During tonight’s Republican debate Rick Perry will use the news reports to claim that Einstein and that whole cosmology thing was wrong so we should trust only the Bible, and some of the other Republican candidates (not Huntsman) will fall all over themselves to be on the latest antiscience bandwagon. Bachmann will have some old lady whisper to her that her daughter was rendered retarded by having to take high school physics. But Mitt Romney will save the day, because in his flip-flopping he is the living instantiation of quantum statesmanship.

  45. AstroPaul

    @30 (James): Remember that special relativity ensures that regardless of our motion (relative to anything!) we should not see anything massive move faster than c. To get this kind of effect from the cosmological expansion, you’d need to put 100kpc of empty space between your source and detector; to get it from a point mass, you’d have to get within 40,000 Schwarzschild radii, i.e. within 120,000 km of a solar-mass black hole. So I think both relative motion and general relativity can be ruled out as explanations — unless one or the other relativity theory is in serious error.

    @8, 9, 15, 16: I’d be very surprised if anyone is racing light and neutrinos to do this measurement. They’re most likely assuming that we know c very well from experiment and comparing the arrival time to the predicted arrival time for “ideal light” in a perfect vacuum. So considerations about EM interactions affecting light and not neutrinos should be beside the point.

  46. GR@Y_M@TTER

    Folks, Einstein said that Light was a universal constant. That is true. Period. Yes It can be slowed down, but It can’t exceed its natural threshold, as defined by nature/God. The scientists will find out what I already know; that the neutrinos took a short-cut.

    Now you see them, now you don’t. They should conform, but they wont. Cause there be holes in space and time. So you sometimes get there before you arrive.

  47. In the article I read (universetoday) it stated that time it took the neutrinos to travel the distance was 60ns, whereas that same distance at the speed of light would’ve taken 2.4 thousandths of a second. By my calculations, that’s 400,000 times faster than the speed of light.

  48. Juan

    38 comments and only one “warp drive” reference? Come on, boys! we’re better than this!

    Bring them on!

  49. James

    @AstroPaul – Thanks for clearing that up for me.

    I love how science turns ‘being wrong’ into a positive experience 😉

    Experiments like these… and you guys debating over them… is humanity at its best.

  50. Poul-Henning Kamp

    You overlook another possible interpretation:

    The constant ‘c’ in relativity is not speed of light, but slightly higher, because photons have rest-mass, which must be slightlig larger than neutrinos.

    The photons rest-mass being zero is based on the fact that we havn’t been able to measure it, not on the fact that it is theoretically impossible for it to have one to begin with.

    The reason this is worth considering is that until a couple of years ago, we were pretty damn convinced that neutrinos were mass-less too.

  51. Brian Too

    Imagine this turns out to be true. Just for kicks. I’m not sure we can exactly start building our interstellar rocket ships yet.

    After all, the increase over c is still only 1 part in 40,000. And while it would overturn the idea that c is some kind of absolute speed bump, there’s no mechanism for getting large multiples of c here. Not in any practical way, and maybe not even in principle. And yet that’s what we’d need in order to make interstellar travel practical on anything like a human timescale.

    So even if true, this might turn out to be applicable only to massless particles like neutrinos.

  52. T

    For your info, they said results will be published… Let’s see what community has to say about it.

  53. CB

    Thanks, Phil. I was waiting for you to give your initial reaction to this, and as usual you didn’t disappoint.

  54. Infinite123Lifer

    Phil stated:

    “Now first off, if it were this would overturn so much physics that they may as well have discovered that gravity pushes, not pulls.”

    My understanding varies on this.

    On one hand Earth “pulls” objects down at about the rate of 9.8 meters per second squared.

    On the other hand I have gathered that Relativity actually explains that the mass of the Earth has warped the space around it, essentially causing the space-time around the Earth to actually “push” the object into the Earth. In which case…gravity actually causes what essentially is a push?

    Yeah? Nay? Ehm?

  55. Oops, didn’t see your parenthetical about how the neutrinos did arrive faster. I still think the Mont Blanc issue may still be relevant.

  56. GR@Y_M@TTER

    MASS SMASH, Anything that enters our (matter-real) material plane, has to take on the shape of matter to exist in the first place. Anything that exists in our universe has a mass. Despite how tiny a given particle may be, it has mass, might be near impossible to measure but it has a mass none the less.

    Consciousness has no mass, but affects the material world thru a material body. Our consciousness exists out side of time and space, and is locked in phase with your genetic code(via junk DNA), to create a quantum connection, that forces matter to simulate, resonate and align with its higher vibration.

    Neutrinos that travel faster than light would be like saying, a car is faster than a bike. And its a true statement under the right circumstances. What are the circumstances. The neutrinos are not moving faster by velocity. But they are by actual distance traveled.

  57. Yousuf

    It’s possible that it’s just a calibration error in the distances. There was an earthquake in this part of Italy a few years ago, the Aquila earthquake of 2009. The land could’ve have stretched a bit, or the land could’ve even risen or fallen in some places putting the source and destination a little further apart vertically. I don’t know if they recalibrated their distances after the earthquakes. And if earthquakes don’t explain the error, then here’s a tachyon joke. :)

  58. Chris A.

    Okay, surely I’m off on this, but here goes:

    According to the Heisenberg uncertainty principle delta(p)delta(x) >= h/4pi

    Meaning, there’s a minimum product of the uncertainty in a particle’s momentum and position. Thus, if we know its position well (where it was emitted and detected), our knowledge of its momentum (and therefore, its velocity) is poor. Is there a chance that the intrinsic uncertainty of the velocity is enough to allow a superluminal v?

    Without seeing their paper, I don’t have any numbers to plug in for the neutrinos’ momenta. And we don’t have a very good number for their masses either.

    But, if the difference between a neutrino’s velocity and c (call it epsilon) is less than delta(v), aren’t superluminal velocities allowed?

  59. Keith Hearn

    Phil wrote:
    “this would overturn so much physics that they may as well have discovered that gravity pushes, not pulls.”

    You mean something like discovering that the universe is expanding instead of being pulled back together by gravity? Inconceivable!

    Also, did we have any neutrino detectors in operation four years before SN1987A was detected? Do we know for sure that there wasn’t a burst of superlimunal neutrinos passing through in 1983? Hmmm, SN1987A was detected at 3 neutrino detectors. Kamioka was completed in April 1983, which would be very close to Phil’s “almost 4 years” before February 24, 1987. Phil, can you be a bit more precise on when we would have seen neutrinos from SN1987A if they were going the same speed as those in this report? On the other hand, the second detector was IMB, which apparently was operational as early as 1982. And Baksan started operations in 1977. So there were detectors in existence early enough to spot them. We don’t know if they were actually in operation at the right time, but I’m sure that could be checked.

    But in the end, I suspect they’ll eventually figure out that it was some sort of measurement error. Hmm, my back of the envelope calculations says that if two points are 730km apart measured along the curved surface of the earth, then they’re about 147m closer measured in a straight line. That’s too much, so that’s probably not the explanation. But it’s probably something like that.

  60. Spection

    As we skeptics have learned to regrettably expect as “quasi-journalistic business as usual”, this story is being hyperluminously hyped by nearly all authors and readers alike at nearly all other sites (:sigh:)… Thus I am once again delighted (though not at all surprised) that our beloved Phil Plait has penned one of the needed but all-too-rare pro-science cautions against falling into yet another mindless-media-driven “science by press-conference” fiasco (lo! what hath Pons and Fleishman wrought!) Well done, sir!

    I’d like to share some quotations from the ScienceNow article Phil cited. Regarding the probability of systemic measurement-related errors producing inaccurate FTL results, “… Chang Kee Jung, a neutrino physicist at Stony Brook University in New York, says he’d wager that the result is the product of a systematic error. “I wouldn’t bet my wife and kids because they’d get mad,” he says. “But I’d bet my house.”

    Jung then goes on to focus his skepticism on one key aspect of the OPERA group’s claims that other web articles posted today haven’t highlighted (so far as I’m aware) — specifically, the suspicious precision of their stated value for the max GPS imprecision. He states: “I would be very interested in how they got a 10-nanosecond uncertainty, because from the systematics of GPS and the electronics, I think that’s a very hard number to get.”

  61. Chip

    Following up on what Brian Too@53 wrote – Yes, I agree – however we wouldn’t need to go FTL to explore [dramatic music] “where no one has gone before” [/dramatic music] – if the technology ever existed, we’d just need to get to just a little under the speed of light, among many other staggeringly difficult technological accomplishments I’m overlooking, and we’d also have to accept that as explorers or settlers, we’d never again see our friends and relatives who remain on Earth.

  62. AndyC

    I was mostly glad to read that they want their peers to dog into their findings. This showed that they are willing to admit mistakes or oversights on their part if any are found. This will be very interesting.

  63. SLC

    Re GR@Y_M@TTER @ #57

    Actually, an upper bound can be put on the mass of a photon. In computing the anomalous magnetic moment of the electron, it is assumed that the photon has mass 0. The result of the calculation is that the computed value agrees with the observed value to 10 significant digits. If the photon had sufficient mass, this agreement would not hold.

  64. Wzrd1

    Of interest is, how many times did they repeat the experiment? Was it ALWAYS 60ms, no more or less? If that is the case, EVERY parameter should be re-examined, including the actual particle release time, not the initiation time.
    If there is a degree of variability to the measured result, has tidal effects been accounted for? The crust DOES move a bit in response to the moon’s passage…
    I’d love to read whatever paper that they produce as well. I’m seriously thinking a minor procedural error is causing this anomaly.

  65. Joe

    Another cold fusion?

  66. Poul-Henning Kamp


    Are you sure that the theory underlying that calculation is not also premised on zero rest-mass for the photon ?

    QED falls apart of the photon has rest-mass and I’m not even sure charge-conservation would survive.

    It is a very fundamental assumption, but as far as I know, we have no hard data that it has to be a mathematical zero mass, only that it must be very small mass indeed.

    But the assumption throughout is that it is zero mass (as in: “speed of light” for instance.).

    Last I heard, the observed limit was measured to mumble*10^-16 eV and that is pretty far from zero, considering how many of the darn things there are.

    Thinking more about it, if both the photon and the neutrino have non-zero rest-mass, it is not obvious to me that the 60ns tells us which one is heavier.

    Anyway, an experimental issue is far more likely, I just wanted to point out that there is another interpretation that does not invalidate the ‘c’ being an absolute upper bound in relativity.

  67. @ Wzrd1,

    Tidal issues would be way too small. They would need to alter things by close to 18 meters (the approximate length that light travels in 60 nanoseconds). Tidal forces don’t move solids that much, and if something was moving that much you’d sort of notice.

  68. I don’t thin the idea of giving the photon non-zero rest mass works. We’d have to explain then why the photon always seems to be traveling so close to the actual limit. Worst, you get the speed of light from Maxwell, and Maxwell with a small bit added in implies most of SR, so the “c” in SR should be the same as the “c” you get from Maxwell. And Maxwell is talking about electromagnetic radiation.

  69. Wzrd1

    Joshua #71, my point, poorly made, was intended to suggest cumulative errors and possible sources.
    Sorry for not being clearer, but I was also having a conversation here at the house while I was trying to post AND being called from the other room… :/

  70. Kim

    Man, there are levels and levels of awesome in this! I think the highest level is not the possibility that neutrinos indeed travel faster than light, but that those scientists are *dealing well with the press*! Admiting that they possibly are wrong, that they had to check and recheck, and that they are opening their data for everyone to analyze is great for True Science©’s image to the public!

    But! What did they expect with the experiment? To find neutrinos arriving *exactly* at the speed of light or a little less, due to interaction with matter?

    I’m expecting eagerly for they broadcast, as well to a Sean Carrol’s post. Tomorrow will be a great day, and all the next to come!


    Way to kill my buzz, Phil. Haha. Skepticism’s a bitch.

  72. Speed of light in vaccuum: 299,792 km/s
    730 kilometer distance:2.43551 milliseconds

    With refraction of air:2.43624 milliseconds

    difference: 73 nano seconds

    Following on my comment at #16 (and karls at 15 you bastard!), that difference is close enough to be within the claimed time measurement error to call into question the entire experiment.

    However, this is seems such a basic and obvious answer to the problem, its hard for me to believe it was not accounted for in some way. Further the calculation above is for a specific wavelength of light in air. They used neutrinos, but it went through rock for much of the journey (curvature of the earth, plus the detector is underground).

    Could they just have assumed no refractive index for neutrinos through rock? Is that a correct assumption? I’m not a physics guy, so I truly dont know.

  73. wzrd1 @68

    from the article…

    “Over 3 years, OPERA researchers timed the roughly 16,000 neutrinos that started at CERN and registered a hit in the detector. They found that, on average, the neutrinos made the 730-kilometer, 2.43-millisecond trip roughly 60 nanoseconds faster than expected if they were traveling at light speed”

    Yes it was done multiple times. Assuming they are competent, i’m going to guess that the median and variance was far enough away from the speed of light to get excited.

  74. techskeptic,

    Neutrinos refracting in rock would be very weird and would almost require a fifth force. Furthermore, that would make the neutrinos move slower not faster.

  75. Chad

    I say we let the gamers figure this one out.

  76. VinceRN

    Great article, a much better explanation of this than the couple I’ve read earlier.

    Is their equipment actually accurate to measure the time of leaving and time of arriving that accurately?

  77. Viadd

    Among the things that might have gone wrong are the geodetics of the distance measurement. Specifically, the Earth is not a sphere or even an ellipsoid, but something bumpy (called a geoid). Also, a latitude of thus-and-such doesn’t mean that the angle to the center of Earth w.r.t the equatorial plane is given by the latitude.

    Also, you can’t use GPS under a kilometer of rock, so you have to get the location somewhere that you can see the sky, then accurately measure the displacement from your experiment. Plumb lines and levels are going to do complicated things due to the local mountains and other masses. Forget about magnetic compasses.

    I am confident at the 99% level that the science team knows all this. But 1% is still much more likely than FTL neutrinos.

  78. Peter B

    David Harrison @ #49 said: “In the article I read (universetoday) it stated that time it took the neutrinos to travel the distance was 60ns, whereas that same distance at the speed of light would’ve taken 2.4 thousandths of a second.”

    Unfortunately the author of that article is wrong, and several commenters have pointed that out.

  79. Peter B

    David de Hilster @ #36 said: “It all boils down to one thing: Einstein is god and he cannot be proven wrong. We know he is wrong, but no one wants to say the emperor has no clothes.”

    Would you care to expand on that, please.

  80. if true this is absolutely amazing. However, this does need to be confirmed for some kind of systematic error. There could be some kind of error in the distance calculations or something. However, there’s really only one place in the world I can think of that could take this on. The Tevatron at Fermilab is the world’s second largest accelerator of this type and can produce similar neutrino streams. Fermilab has a detector setup some distance away.

    The best way to confirm this would be to try to observe the same thing at Fermilab. If they did, that would just about clinch it and prove it was not some kind of bug in the system at CERN.

    Oh… except the Tevatron is set to be shutdown on the 30th of this month. Fermilab could not secure $35 million in federal funding to keep it open another three years (yes, 35 million with an M. Like,less than is spent on the National Centers for Complimentary and Alternative Medicine in the same time period.

  81. Robin

    There’s a whole lot of excitement that’s way out of place right now. Given the scientists who did the experiment aren’t sure of their results, it’s at least premature to start making proclamations about emperors and their new clothes, Einstein’s god-like status, and so on. The scientists are asking for other scientists to essentially help them with error analysis, theoretical analysis, and etc. I think publishing a paper with tentative results would have better, especially in light of well things went for Pons and Fleischmann.

    There are a lot of things to consider here:
    –how accurate and precise was their event timing?
    –what uncertainty is there in the assumption(s) that they were timing the correct neutrinos?
    –what is the real error in distance measurement?
    –how well does the data from a set of runs correlate with the rest of the data?
    –much more……

    Certainly, the assumptions that have been made about what parameters were constrained and known need to be closely examined and tested. Given the measurement resolution of the test and its intent, it certainly cannot be assumed that the source frame and the measurement frame were inertial. In fact, it’s likely they weren’t inertial at all. After all, there will be some difference in acceleration of gravity between the two frames. I’m sure there are going to be many scientists expert in GRT and SRT analyzing the setup and data.

    It is fun to think about. We understand well, just from mathematical analysis, what happens when vc…..well, things get funky. Suddenly a pesky i pops up, and it’s not clear what happens then. In a lot of science, we’d discard the imaginary part, but here maybe that can’t be done. Long ago a smart person or two looked at the seemingly pedestrian Pythagorean Theorem and wondered what happened when c (you know from the a^2+b^2=c^2) had an imaginary solution (we conveniently threw those out all through junior high, high school, and much of college). They then realized that perhaps that solution represented a path not in our common 3D space. That idea blossomed a bit and had impact and influence on some ideas about strange physical phenomena today (wormholes, black holes, and etc.). What impact will or does that i in some Lorentz Factor have on what we believe? What could it mean? It’s certainly fun to think about and could lead to valuable science and technology. That’s far off the path of where things are now and what needs to be done and analyzed with the test in question.

    I’ll wait until the results are released in a paper and others repeat the test. The Scientific Method has been pretty damned reliable so far.

  82. Man, I get into these discussions late…

    GPS can be made accurate to within 1cm using surveyor and military grade equipment.

    Has anyone done a resurvey since the last big quake in Italy?

  83. Sander Salazar

    Would it be possible to share an article about what would it imply if, indeed, the speed barrier have changed to the neutrino’s?

    Since my knoledge of physics isn’t quiet as high as I wish I can’t think of any effect other than relactivity and possibly eletromagnetism, though how would it change and how it would affect anyone isn’t something easy to grasp. I think that would be a good reading for the weekend.

    Well anyway, good article, it is something that more people need to read, the amount of misleading and dreamstate statements I have already read is scary, many StarTrek references and even some ballistic!

  84. Sander Salazar

    Would it be possible to share an article about what would it imply if, indeed, the speed barrier have changed to the neutrino’s?

    Since my knowledge of physics isn’t quiet as high as I wish I can’t think of any effect other than relativity and possibly electromagnetism, though how would it change and how it would affect anyone isn’t something easy to grasp. I think that would be a good reading for the weekend.

    Well anyway, good article, it is something that more people need to read, the amount of misleading and dream-state statements I have already read is scary, many Star Trek references and even some ballistic!

  85. Doug A


    The official paper notes that they detected a 7cm shift in their baseline due to the 2009 earthquake in Italy.

  86. Beelzebud

    I’m just glad particle physics is back in the realm of making real tests, and observing real reactions. For too long it’s just been a lot of math, with not much based on what we see as reality. I just fear that something like String Theory is so ingrained in people’s heads, that they’ll continue “bending the math” to fit whatever observations are made.

  87. Paul

    @Steve Packard

    Neutrinos at fermilab are created by the Main Injector, not the Tevatron. The neutrino program there is continuing for the forseeable future (MINOS+, Nova, Minerva, etc)

    As for confirming this, MINOS also reported, a few years ago, that their neutrinos were arriving around 60ns earlier than expected on a similar baseline to CNGS, but their systematics were also around 60ns.

  88. Wzrd1

    @#81. techskeptic, thanks. Only barely had the chance to skim the article. Too many distractions here at the time…
    Hmmm, if the distance were greater, hence, the time, I’d suggest neutrino switching and some potential special physics involved in it, but not at this close a distance.
    I can’t see diffusion of rock, not dense enough to properly interact with that miniscule neutrino.
    I noticed a kind poster sent the URL for the report. I’ll review to see what the distribution is. Certain artifacts can be technologically induced, but may be outside of the view of the scientists (such as network issues, noise factors in circuitry, etc. I’m both a network guy and an electronics guy, so I’m conversant with some potential issues.

    The abstract removes the probability of geographic errors: “Dedicated upgrades of the CNGS timing system and of the OPERA detector, as well as a high precision geodesy campaign for the measurement of the neutrino baseline, allowed reaching comparable systematic and statistical accuracies.”
    “The kicker magnet trigger-signal for the proton extraction from the SPS is UTC (Coordinated Universal Time) time-stamped with a Symmetricom Xli GPS receiver [10]. The schematic of the SPS/CNGS timing system is shown in Fig. 3. The determination of the delays shown in Fig. 3 is described in Section 6.” Hmmm, removes ONE source of error.

    “The proton beam has a coarse bunch structure corresponding to the 500 kHz of the CERN Proton Synchrotron (PS) (left part of Fig. 4), on which the fine structure due to the 200 MHz SPS radiofrequency is superimposed, which is actually resolved by the BCT measurement, as seen in the right part of Fig. 4.” Removes ANOTHER source of error. Looking at the graphic, I’m becoming excited with a GOOD problem. :)

    Figure 5 is giving me heartburn, it looks like TWO signals before the peak from CERN. Though, the first two seem weak looking.

    “Fig. 7: Monitoring of the PolaRx2e GPS antenna position at LNGS, showing the slow earth crust drift and the fault displacement due to the 2009 earthquake in the L’Aquila region. Units for the horizontal (vertical) axis are years (meters).” ANOTHER suggested error source in the circular file.

    Interesting! “The total statistics used for this analysis consists of 7586 internal (charged and neutral current interactions) and 8525 external (charged current) events. Internal events, preselected by the electronic detectors with the same procedure used for neutrino oscillation studies [29], constitute a subsample of the entire OPERA statistics (about 70%) for which both time transfer systems at CERN and LNGS were operational, as well as the database-logging of the proton waveforms. As mentioned before, external events, in addition, are requested to have a muon identified in the detector.” Could they be oscillating and tunneling when oscillating? THAT could create a solution AND give some serious heartburn for some of the quantum guys and gals out there…

    “The schematic of the SPS/CNGS timing system is shown in Fig. 3. A general-purpose timing receiver “Control Timing Receiver” (CTRI) at CERN [30] logs every second the difference in time between the 1PPS outputs of the Xli and of the more precise PolaRx2e GPS receivers, with 0.1 ns resolution. The Xli 1PPS output represents the reference point of the time link to OPERA. This point is also the source of the “General Machine Timing” chain (GMT) serving the CERN accelerator complex [31].
    The GPS devices are located in the CERN Prevessin Central Control Room (CCR). The time information is transmitted via the GMT to a remote CTRI device in Hall HCA442 (former UA2 experiment counting room) used to UTC time-stamp the kicker magnet signal. This CTRI also produces a delayed replica of the kicker magnet signal, which is sent to the adjacent WFD module. The UTC time-stamp marks the start of the digitization window of the BCT signal. The latter signal is brought via a coaxial cable to the WFD at a distance of 100 m. Three delays characterise the CERN timing chain:
    a) The propagation delay through the GMT of the time base of the CTRI module logging the PolaRx2e 1PPS output to the CTRI module used to time-tag the kicker pulse ΔtUTC = (10085 ± 2) ns;
    b) The delay to produce the replica of the kicker magnet signal from the CTRI to start the WFD Δttrigger = (30 ± 1) ns;
    c) The delay from the time the protons cross the BCT to the time a signal arrives to the WFD ΔtBCT = (580 ± 5) ns.”
    The kicker signal is just used as a pre-trigger and as an arbitrary time origin. The measurement of the TOFν is based instead on the BCT waveforms, which are tagged with respect to the UTC.” I’m out of potential sources for error from the circuitry or network.

    I’d LOVE to see their baseline signal resolution, without CERN input…

  89. Jason

    From what I have seen they are using the aggregate results of three years worth of expierments. I would guess that the first couple of times the times where off one of the very first things they would have done is to 1) ensure their clocks were right and 2) double-check the distance. And then probably do it all over again.

    Perhaps they even re-calculate the distance each time to make sure. I haven’t read the paper to see if they included the procedures followed included this step or not. Given their invitation for peer review and confirmation I would assume they at least believe all their ducks are in row. (Very very fast ducks it appears)

  90. Rick

    *yawn* one of my other infinite selves in a different trouser leg of time read this 6 years in the future.

    Me in this world says, ‘cool’. Looking forward to the peer review stuff. And, lets all give subatomic-particle-physicists some credit for probably thinking of each of the ‘inaccuracies’ we’ve thought of.

  91. Crux Australis

    It almost seems odd, to me at least, that the difference (1 in 40,000) is so small. I mean, the electron’s mass is 1/1800 of the proton’s mass, h has no other constant even near so small (AFAIK), and c is so high. It just seems more likely to me, that such a small difference is more due to experimental uncertainty, than to some ground-breaking New Physics.

  92. Wzrd1

    Jason, someone posted the URL for the paper earlier.
    Just skimmed through it, as is in my post above.
    I’m wondering if neutrinos lose mass during oscillation or if they’re perhaps tunneling for some odd reason…

  93. Read about this on the internet, was skeptical. Came here to get explanation of whats’ going on.

    Did not leave disappointed.

  94. PatrikD

    Keep in mind that 60 ns is a lot, given the kind of accuracy that is routinely achieved with atomic clocks in particle physics. Likewise, 18 meters (60 nano light-seconds) is a huge distance even with fairly simple methods.

    The magnitude of that effect rules out a lot of likely failure modes. Which means they’ve either done something really stupid, OR there’s some other interesting non-FTL effect at work (like, the neutrinos were actually generated 60ns earlier than they assumed?), OR they’re about to overturn physics as we know it.

    Also, I’m sure the discoverers are aware that those are their options – including the first one…

  95. punk_rock_physics

    Lets suppose this result is valid for a minute. Could one of you guys explain the paradoxes that would imply? Could we use neutrino beams to send information about today to ourselves in the past?

  96. Tom Brown

    Hi Phil,

    Great piece. Your point about SN1987A is excellent, and I’ve reposted it elsewhere.


  97. Physics is inconvenient sometimes. It just really ought to be possible to get out and see the universe. There SHOULD be a way to travel FTL. Damned reality always standing in the way of our fun.

  98. Even if the results are wrong and neutrinos travel at c. neutrino’s have mass. Particles without mass travelling at c I can deal with, but particles with mass?

    Maybe neutrino’s predilection to oscillation might cut in here, but got me.

  99. tqft, It has been known for a long time that neutrinos generally travel very close to c (generally at least (1-10^6)c ). They have so little mass that it doesn’t take much to accelerate them to very close to the speed of light. So that isn’t new.

  100. Parlyne

    @Poul-Henning Kamp/#73 Amusingly enough, performing the calculations you’re asking about actually requires one to assume a photon mass to avoid a nasty infrared divergence in certain individual terms which would prevent them from being evaluated at all. In the final result, the infrared divergences cancel, and it’s consistent to take the photon mass to zero, which is how one arrives at the standard result that agrees so well. But, if you take seriously the idea of photon mass, you can ask how large it can be without changing those results too much.

    As for this experiment, though, the neutrinos aren’t being raced against light. Their travel distance and flight time are being measured directly. The comparison to the speed of light is simply a mathematical one.

  101. Chris L.

    Unless I’m mistaken, the neutrino’s mass should have gone to infinity just as it passed the speed of light. Also, it would have required an infinite amount of energy to get it there. Wouldn’t have those two things been noticed at the time?

  102. Joseph G

    @52 Poul Henning: I was wondering if it could be something like that – that it’s not a case of neutrinos “breaking the light barrier”, but simply a case of photons not actually moving quite at the speed of… for want of a better term, the speed of causality?

    I wonder if neutrinos have ever actually been timed before this? Neutrinos from the sun, supernovas, nuclear reactors etc can’t really be timed with the degree of certainty we’re talking about here? I mean, I’ve heard of neutrino detectors picking up nuclear weapons tests, but I still kinda doubt anyone had the data necessary to time them to the nanosecond.

  103. Joseph G

    @60 Yousuf: I laughed, but you know I’d already seen that one, right 😉

    @78 Kim: The scientists are dealing well with the press, but is the press dealing well with them? I anticipate many facepalms in my near future as I follow coverage of this.

    Anyway, I’ll try to add some crazy speculation, ‘cuz I want to be part of the group :)
    Maybe they’re accidentally creating tachyons which are hitting the target 60 ns early and creating neutrinos that travel exactly as fast as we’d expect them to?
    Or maybe there’s a loose bit of metal 18 meters up the beam path, and that’s where the neutrinos are getting generated?
    Silly, I know, but this is fun…

  104. James

    @Chris_L unfortunately the scientists didn’t realise their experiment had used an infinite amount of energy… until they got the bill from their electricity company. 😉

  105. Neutrinos have a very small amount of mass, is it enough for them to be affected by the Earths gravity on the way through? Does that play a role? Gravity after all can be strong enough to bend light.

  106. Joseph G

    Also, I’m assuming they allowed for altitude? Even if they’ve got the XY coordinates pinned down to a fraction of an inch, they might be doing their distance calculations XY to XY. I kinda doubt they’d forget this, though, as presumably both machines are underground (I’m not familiar with OPERA, too lazy to look it up before posting).

    Also, the earth isn’t perfectly spherical (even beyond its oblateness), so even coordinates that are locally accurate in all three dimensions might not line up quite the way we think they will?

    Hell, I think I’m going to bet my money on this second scenario. Do I have any takers? 20 bucks each 😀

  107. Adam

    @ #14 a typical Garmin GPS for your car, those cost less then $100 and have low accuracy +/- 15 m, the GPS used by the scientist is likely a sub-centimeter accuracy GPS that cost hundreds of thousands of dollars and can accurately locate a point to about the size of your thumbnail.

    i can’t wait to hitch a ride with Kirk on the Enterprise.

  108. Keegan

    Hey Phil, I love your work but I have one question… Why do you hate my dreams of time travel?

  109. Bill Clack

    Take-aways are: photons probably have mass; C is at least 8 km/s faster than we thought; C is not light speed; C > neutrino speed. New value for C may explain Pioneer spacecraft positional anomalies. E=mC^2 : we all thought the speed of light was C. Light speed is still a constant, but it isn’t C. The reason we got confused is the speed of light is so amazingly close to “C”. 5 miles/sec; 8 km/s. If C in e=mC^2 is larger than we thought, then more energy hides in mass than we thought. We may have found the missing mass of the Universe–in front of our eyes. Photons…@billcad15

  110. Brian Weaver

    How sure are we that superluminal neutrinos WEREN’T detected sometime in 1983-84? No one was really sure you could even detect supernova neutrinos until 1987A happened, so at the time period when the hypothetical superluminal neutrinos would have arrived the science of detecting neutrinos was still relatively young. One of the neutrino detectors that picked up on the 1987A supernova, Kamiokande II, didn’t go didn’t have the requisite sensitivity to detect supernova neutrinos until upgrades were done in 1985. The other, IMB was still pretty new during this period, only having made it first observations in 1982. With the equipment still getting shaken down, you have to wonder if any superluminal neutrino busts that may have been recorded but without an accompanying supernova might have been throw out as spurious data, especially without Kamiokande II available to back up the results.

    Is it possible the data was actually captured, but it was never considered good enough quality to publish? If the CERN results pass initial scrutiny, I’m curious what a re-examination of the raw data from the collectors of the period would reveal.

  111. Andreas H

    I just love the process.

    Scientists figured something out and to the best of their abilities made sure there was no error. Now this claim is out in the open and everyone can take a good hack at it. Just think about the brain power pondering over this issue right now, magnificent!

    No matter the end result, this is just pretty damn awesome.

  112. Rasputin

    I haven’t been this excited since I saw Princess Leia’s slave girl costume for the first time.

    It’s probably not true but holyzomgawesome what if it is?

  113. Aleksandar

    I’ll try some mind test. Let’s suppose the results of measurement were true (I wouldn’t bet on it, but suppose they are true). What we get? The speed of neutrinos measured through rock and air is a fraction of c greater then the speed of light measured in vacuum (c). And we know from other measurements (supernova’s, GRBs, etc) that the speed of neutrinos is approximately or exactly c, measured through the medium of vacuum. The difference here is the medium (rock and air versus vacuum). For photons, the difference is measurable and known. Speed is 0.00m/s through rock and they are slowed down in air, a fraction less than c. In vacuum, light speed is, well…c.
    With neutrinos, we would get the other effect, they are sped UP, not down in presence of mass/gravity. If these is true, it CAN be measured to even greater accuracy if we put a detector on the other side of Earth (New Zealand, perhaps). What would this mean for all the theories of forces and gravity, is mind boggling. I repeat, only and only if these measurements come to be true (which I highly doubt).

  114. Duncan Young

    GPS position solutions are typically calculated in a earth-centered earth-fixed reference frame, so the concerns about curvature and rotation are probably not valid. One catch is the wet tropospheric correction, which can lead to elevation biases of 10’s of centimeters.

    The critical citation is number 26, G. Colosimo, M. Crespi, A. Mazzoni, M. Jones and D. Missiaen, “Determination of the CNGS global geodesy”, OPERA public note 132 (2011), which does not appear to be on the web.

    There is this from 2002:; citing issues with traffic constrained access to survey points in the road tunnel near the detector.

  115. Evil Merodach

    While I am skeptical of these findings, this is the second experiment that has intimated that neutrinos are somehow moving at superluminal speeds. Fermilab’s Minos experiment showed similar results but the findings were within statistical error ranges.

    There is an explanation that doesn’t violate Einstein’s speed limit but it may be even more mind-bending.  It’s been suggested that there is a fourth type of neutrino called a “sterile neutrino.” The sterile neutrino is suspected to be a type of particle called a Kaluza-Klein particle which, unlike normal matter, has the ability to escape our three spatial dimensions and travel into higher dimensions.  The LHC is expected to prove or disprove these extra dimensions.  This may be the first inklings of higher dimensions to come out of CERN.

    If there are extra dimensions, there is no reason to expect that the speed of light there is the same as it is in our three-space.  All three known types of neutrinos (electron, muon, and tau) have mass, and thus cannot reach the speed of light.  The sterile neutrino is also expected to have mass. But as neutrinos “rotate” through each of their different types, the sterile neutrino may be straying into dimensions where the speed limit is higher.  When the sterile neutrino rotates back into one of the known three types, it may be forced back into our more mundane dimensions.

    As for the neutrinos from Supernova 1987A, those neutrinos are far less energetic than those used by CERN.  It may be the energy of the neutrino contributes to the effect.  

    This news falls under the heading of extraordinary findings so it definitely requires extraordinary proof. This is just one experiment out of many that are needed to confirm this result.  Still, this is exciting stuff!

  116. Joseph G

    @121 Andreas: I know, isn’t it great? I can practically hear the gears grinding. And by contrast, imagine a not-so-skeptical community (lets say homeopathy) that when faced with new information would either dismiss it outright or get really excited and wax imaginative without ever bothering to do anything useful, such as confirm or disprove the discovery.

    @123 Evil Merodach: Fascinating! Particularly the bit about CERN producing higher-energy neutrinos then a supernova! Or is this a case of the neutrinos all rotating to a lower energy state en route to earth?

    *mental note for my hard sci-fi novella: FTL communication but no FTL spaceships* 😛

    And as for

  117. Evil Merodach

    I agree with Brian Weaver (#120) that we may not have had the means back in ’83-84 to detect neutrinos from Supernova 1987A. It could be that in 1987 we were detecting only the lower energy neutrinos from that supernova? It might be interesting to see if there are any neutrino bursts unassociated with any event that might herald a pending (to us) supernova.

  118. Julie

    The opening headline on this mornings French breakfast TV news was devoted to this story. “Scientists run experiment and have unexpected results. They may replicate this, or they may not”. They interviewed two scientists, one in Geneva and another in Paris. They showed pictures of of products being scanned at a supermarket as proof of the importance of Einsteins theory. . Quite exciting.

  119. andrew

    when i heard about this on the radio, my very first instinct was “theres a mistake somewhere”.

    i tried to think of a few things from what i had heard about it on the radio, one of which was the distance thing Phil mentioned. But seeing as 30cm = 1 ns, GPS can easily nail it to less than that! are they using straight line distance or distance along the curve of the earth?

    maybe it was the exact timing of when the neutrino left and arrived?

    the true test lies in whether their results can be reproduced by a different team, in a different place! it may take a while, but i cant wait!

  120. kamacitrus

    Yes, I’m sure they’ve used a Garmin GPS unit to measure the distance. Give me a break.

    I think one thing that may be a point of conversation is; what part, if any, does the observer with foreknowledge of the arrival have to play in this experiment? We all know about Heiblum’s study. Just a thought… This, if true, will change the fundamentals of our daily lives.

    Cause and effect.

    Think about it…

    We’ve all had the thought “This is a bad idea”. What happens after a thought like this crosses our minds? Usually something bad. Are our action determined because of the probable outcome? If so, where does freewill exist?

    Amazing study.

    Perhaps we will finally have a unified theory.

  121. TerryEmberson

    One of the first things I did when I heard this was to try and find a real science blogger to clarify it. At least I’m getting skeptical instincts.

  122. Ralph

    The neutrinos are produced in ten microsecond bursts from a proton beam. So the time-distribution of the neutrino burst would appear to need to be known to part-per-thousand accuracy to get the claimed timing accuracy.
    The paper indicates that the time-distribution of the *proton* burst is understood (not sure how accurately).
    However, if the geometry of the proton beam changes during the burst, then that might influence the direction of the subsequent neutrinos, biasing the distribution of neutrino arrival times at the detector?
    I’d be looking very very carefully at the pulsed kicker magnets that extract the protons out of the SPS, and also at mechanisms for protons early in the burst to disturb protons later in the burst.

  123. prianikoff

    Neutrinos are detected indirectly, using Cherenkov radiation.
    So, if they’d arrived at the photo detector faster than light, wouldn’t they produce a characteristic double image, after the event?
    One of the particle arriving at the detector and one of it departing, in the opposite direction.

  124. Pete

    I think today’s xkcd sums up everything nicely…

  125. Eric Pepke

    I see a much more basic problem with these results.

    It’s possible that the known value of c is slightly wrong. However, ignoring that possibility…

    Let’s say that a neutrino, which has mass, is going faster than c. That means one of two things:

    1) It has imaginary rather than real energy under relativity (plug it into gamma), or
    2) Relativity is off.

    If 1, I’m not sure how a detector could detect imaginary energy.

    If 2, then the GPS, which relies on relativity, might be off, too.

    Pedantic note: If c were a little larger, they wouldn’t change the number, just make the meter a bit longer. Same result.

  126. Mark Bruton

    Just a thought as I am not a physicist, can they launch both a photon and what is it neron at the same time and see which one arrives first,

    is it that simple , i suspect not

  127. b

    This reminds me of when Cold Fusion came out

  128. ponyface

    Is the 730km or so along the surface or detector to detector (ie through the Earth)? Surely that could account for 20 metres.

  129. I don’t know the math on this, but I’ll be interested to see how it holds up under the scrutiny of the scientific community.

    xkcd has an entertaining take on it….

  130. haha

    If using GPS, they will use Einstein’s relativity theory, but if you find a faster particle the GPS is not valid anymore because relativity does not apply. So, can we find a faster particle using GPS?

  131. Nigel Depledge

    PeteC (37) said:

    For all the dedicated and hard work done, we’ve not really seen a proper breakthrough in physics – not one that’s been experimentally shown to be true rather than string-theory philosophising – for the better part of a century now.

    Manhatten project.

    Quantum Electrodynamics.

    Quantum Chromodynamics.

  132. DennyMo

    I love XKCD’s treatment of the report:

  133. Dave Steckler

    My first area of concern after my first read of the article is that the distance measurement might be incorrect (it is, in fact, not a “measurement” – it is a calculation). While the GPS positioning on the surface of the earth is very, very accurate, calculating a point-to-point distance is derived from simply using the etrf2000 geoid.

    Since the etrf2000 geoid is an approximation of reality, so is the “known” distance the neutrinos travelled.

  134. MarkHB

    Move at c
    Move at c
    Whisper words of wisdom
    Move at c…

  135. Nigel Depledge

    Gray matter (59) said (and, no, I’m not impressed by your use of “@” symbols in place of the letter “a”):

    MASS SMASH, Anything that enters our (matter-real) material plane, has to take on the shape of matter to exist in the first place. Anything that exists in our universe has a mass. Despite how tiny a given particle may be, it has mass, might be near impossible to measure but it has a mass none the less.

    Sound waves have no mass.

    Are you saying they don’t exist?

    Consciousness has no mass, but affects the material world thru a material body. Our consciousness exists out side of time and space, and is locked in phase with your genetic code(via junk DNA), to create a quantum connection, that forces matter to simulate, resonate and align with its higher vibration.

    This makes no sense at all.

    A recipe for rice pudding has no mass, but it still exists in the real world, and we interact with it in exactly the same way we interact with other forms of information. Neither vague “quantum connection” nor “aligning with higher vibration” required.

    Neutrinos that travel faster than light would be like saying, a car is faster than a bike. And its a true statement under the right circumstances. What are the circumstances. The neutrinos are not moving faster by velocity. But they are by actual distance traveled.

    So what is velocity, if not distance travelled per unit time?

  136. jupiterisbig

    I love all the posters here commenting about the earth’s shape and how accuracy – that’s what I would have discussed at high school, man … I think they threw it open because they were deep in n-space or some such and still couldn’t’t find a solution! one for messier or cromwell …. 😉

  137. Eric Pepke


    Remember the mirror on the Hubble telescope? I’ve heard that people were so busy looking for fine problems they didn’t notice a gross calibration one.

    Oh, and a recipe for rice pudding has mass. It has information, which means it has a temperature, which means it has energy, which means that unless it’s moving at c which would make following it difficult, it has mass.

  138. Christian Treczoks

    I bet that the distance is somehow “off” by the right amount.

    Earth is not a perfect sphere, maybe they mapped GPS coordinates on a sphere and calculated the distance based on this (No, I did noch check whether this would provide for a sufficient difference, I’m just thinking aloud) . Or they just checked everything over and over again with the exception of the one very measurement that is off (hi, Murphy).

    I just imagened the following scene:

    Jamie Hyneman (looking at a neutrino detector on a table at M5 industries): “Thats an interesting device! So what do we have today?”
    Adam Savage: “CERN claims that they found neutrinos being faster than light, and I thought when we can confirm THAT…”

  139. Georgia

    Not sure if this has been addressed … but their talking about comparing the speed of neutrinos through dirt (I think it’s all underground) with the speed of light through a vacuum (i.e. nothing). They can’t send light the same way…

  140. GPS accuracy can be boosted to centimetre levels. It uses signals from stationary radio stations but is commonly use in geodesy. So at least distance they have got accurately.

  141. flip

    While skeptical of the results – and from Phil’s reckoning, it sounds like it’s more likely that the results are off a little – the potential for coolness is huge. Sounds like it also needs lots more study to discover out what’s really going on! I do hope that an update here of what happens next will occur. This is just too interesting not to follow. (Sadly, I suspect I’m going to be disappointed)

    #20, BJN summed it up nicely.

    New word learned today: dilithium. Thanks Phil! I always learn something here, even if it has nothing to do with the actual topic.

    Thank you Nigel (#156) for deconstructing those weird comments. I was waiting for someone to do that.

  142. Mesh

    Can’t wait till they invent Neutrino based internet! Imagine the ping!

  143. @47. AstroPaul,

    If they are not racing photons and neutrinos to the target, it’s makes it harder to eliminate experimental technique /measurement errors. That would mean they are assuming their control, “We already know the speed of light in a vacuum, so we don’t need a control in this experiment.”

    In college, we always performed the control experiment, even if it was a known value, such as the speed of light. Part of the purpose of control is also to ensure you don’t have some error in the techniques used. “Well we’re faster than c, but so are the photons, there must be an error somewhere.”

    It seems to be an interesting coincidence that the distance used and the velocity supposedly measured were just enough (relatively speaking) for the results to be indisputably significant (if shown to be correct). If you think about it, it’s somewhat anthropicly lucky. There’s no a prior reason to suppose the degree to which the velocity of these neutrinos exceed the speed of light would be detectable at the distances over which we are able to experiment, nor is there reason to assume the opposite, either, it would seem the experimenters lucked out.

  144. Infinite123Lifer

    127. Andreas H Says:
    September 23rd, 2011 at 1:01 am

    “I just love the process.
    Scientists figured something out and to the best of their abilities made sure there was no error. Now this claim is out in the open and everyone can take a good hack at it. Just think about the brain power pondering over this issue right now, magnificent!
    No matter the end result, this is just pretty damn awesome.”

    Here here. Awesome statement. It’s one I can ponder at least and appreciate.

    Robin says:(92)

    “It is fun to think about. We understand well, just from mathematical analysis, what happens when vc…..well, things get funky. Suddenly a pesky i pops up, and it’s not clear what happens then. In a lot of science, we’d discard the imaginary part, but here maybe that can’t be done. Long ago a smart person or two looked at the seemingly pedestrian Pythagorean Theorem and wondered what happened when c (you know from the a^2+b^2=c^2) had an imaginary solution (we conveniently threw those out all through junior high, high school, and much of college). They then realized that perhaps that solution represented a path not in our common 3D space. That idea blossomed a bit and had impact and influence on some ideas about strange physical phenomena today (wormholes, black holes, and etc.). What impact will or does that i in some Lorentz Factor have on what we believe? What could it mean? It’s certainly fun to think about and could lead to valuable science and technology. That’s far off the path of where things are now and what needs to be done and analyzed with the test in question.”

    This kind of thinking is of the highest degree. I appreciate the statement in full. I could not agree more heartedly. I would even venture to say that its never far off the path when considering unknown variables especially based on new observations. Very intriguing, highly intensive thought produced blog. I think the same thing every day. For instance:

    If the definition of 0 is the absolute absence of the universe. And the definition of 1 is the universe itself. I have some very suggestively crazy ideas that I care to share to the right sort of people…first, that 0 (zero), in this world, in our math, in working out real life problems…where mathematics results in an undefined, incalculable or undetermined state that; zero, actually does not exist in some instances, or at least would better suit the base 10 system better if infinity were to replace by the 0 when talking about certain aspects of…the universe and its fundamental parts. After all if you have absolutely nothing, than you have infinitely nothing, and if you have a universe (meaning 1, always) than it is impossible to have nothing or 0.

    I know it sounds stupid, but I’ve been thinking along those lines for a long long time, and most everything totally outrageously possibly and true or not does sound stupid to most the first time it is offered (i guess thats what makes progress fun, or at least slow and excruciating) An incredible list of actually “of no relevance” things are perhaps just “not relevant” as well.

    Please somebody mathematically inclined consider the ramifications of devising a new # system based on the only place where 3 primes show up in a row…arguing that 1 is a prime number. For there is only 1 universe as the number is defined and nothing can multiply together to make it except for itself and 1. Thus I offer that 1,2 & 3 are all prime numbers…of which, this only occurs once in the entire numeric system and that zero does not exist but should be replace with the symbol of infinity and that by using triangles, anything can be calculated and understood.

    I am not a cookoo. I just have this idea. I wish I could grasp the mathematics to really check into this further. Somebody please just think about for a second. I know how truly “divine” almost mathematics is & now that I started learning about the science of Logic I am even more impressed by its natural world continuity.

    But to really prove my point! What if we only had 3 fingers and not 10. I am highly intrigued in profound and simple logic forming from the complicated darkness and things which are not understandable.

    Awesome stuff, I have not looked up this many terms since my last chemistry class.

    I ll let you guys get back to it.

  145. M. Pu

    In which amateurs think they know better than CERN physicists.

  146. Nigel Depledge

    Eric Pepke (158) said:

    It has information, which means it has a temperature

    How so?

  147. Joseph Hoshen

    GPS is very accurate for stationary points especially if GPS measurements are done continuously. A centimeter level of accuracy is achievable. (GPS is used to measure earth plate motions to sub-centimeter level accuracy.) To get GPS position at a given point in space at least 4 GPS satellites have to be visible, which is almost always achievable. Based on data from these satellites, the GPS receiver calculates X, Y, Z and the time t of the GPS fix. The X, Y and Z are universal GPS coordinates. Earth curvature has no impact and is irrelevant to the GPS coordinate calculations. The distance can be calculated from the coordinates of the source in CERN and the detector coordinates at OPERA Italy using a very simple Pythagorean formula. Furthermore, because the time of each GPS fix is also calculated, any variation in distance as a function of time can also be determined with high level of accuracy. So the question of distance accuracy between the 2 points is not an issue. The time of flight of the neutrinos is a more complex problem and could be in question.

    If the tests were repeated for various distances, say 1460 km; and if the neutrinos arrive 120 milliseconds earlier than the speed of light in vacuum predicts, it would give a significant boost to the results of the current test.

  148. KennyBloggins

    Warp Speed Engage!

  149. Joseph Hoshen

    Correction to my previous posting; the time unit is nanoseconds; not milliseconds!!!

    If the tests were repeated for various distances, say 1460 km; and if the neutrinos arrive 120 nanoseconds earlier than the speed of light in vacuum predicts, it would give a significant boost to the results of the current test.

  150. Betz

    For a fun thought experiment, assume the results are actual and true. So what’s special about the path these neutrinos take? They go through several km of rock. They have to be passing through N atomic nuclei. Is there any mechanism that could act as if they are being accelerated & decelerated as they pass through the nuclei, kind of like gravity slingshot of a space probe? OK, granting that c is indeed the speed limit would rule out any acceleration; then the actual distance travelled must be shorter. Can there be any kind of de facto shortcut for a neutrino so that it doesn’t really travel the width of a proton or quark, but sort of instantaneously appears on the other side?

  151. Kaychsea

    It’s through the definition of entropy in relation to information theory. In actuality it doesn’t.
    Of course where the statement misses the point is where he assumes that because something has energy it has mass, if this was so a photon would be unable to move at the speed of light because it has energy. It does because what it has is mass equivalence. Which is different.

  152. Kriton

    Phil admittedly points out that his argument about the neutrinos from SN1987A is a bit weak. Keith Hearn and Brian Weaver already make some good points about this. I’ll point out a couple more.

    The analysis of SN1987A makes the assumption that the measured speed is consistent throughout space time. However, if this is not true, but turns out to be, for example, some faster than light tunneling exclusively at the start of the process, both the conclusions made above and the experiments may be consistent.

    Secondly, has any analysis been attempted for correlating detection events with others years in advance of the supernovas? How do we know the spike attributed to SN1987A might not be attributable to another supernova that is yet to be seen? I doubt, until now, that the question has ever been seriously considered. I agree with Brian – review the data. The above argued its absence, but if no analysis has been done, then how do they know it is absent?

    So be careful about believing that an experimental result has been debunked via clever, but ultimately irrelevant analysis.

  153. koza dereza

    Of course neutrinos do not travel faster than light.

    Yes, there are theoretical mechanisms for superluminal speeds, but they are extreme theoretical constructs, not anything we see on earth.

    I suppose some people still believe the magnetic monopole was actually discovered in San Francisco in the early 1980’s t00. That was a little bit before the web became popular, so it did not get some much coverage.

    Like the monopole, this “event” will not be confirmed. It is rather sad it gets blasted over the internet, however.

  154. Viadd

    Looking at the paper, it appears that they probably did the geodesy right. (Doing old-school surveying from GPS base-stations at both ends of the tunnel, ETRF2000.) Knock the Bayesian prior of that source of error down to below 1e-3.

    The time transfer includes a 8.3 km optical fiber, which means a ~part-per-thousand error in the propagation delay can cause the effect (the stated error is a part in 40k).

    The wrong way to measure the propagation delay of a fiber is to bring it into a lab measure it with a (light pulser & photodetector) at one end and a photodetector at the other, going into a timing circuit. That tells you the what the propagation delay is in the lab, but as-installed the cable will stretch, it will be at different temperatures, it will generally be different.

    The less-wrong way is to put a reflector at the far end of the installed fiber and measure how long it takes for the signal to go down and come back and divide by 2. The propagation delay is not necessarily the same in each direction (e.g. Faraday effect, change in polarization from the reflector) so that can screw you up.

    I assume (I Am Not A Metrologist) that the right way to do it is synchronize a portable atomic clock to the signal generator at one end, carefully transport it to the other end and check the delay, bring it back to the signal generator to check for drift, and repeat to build confidence. You do have to worry about both special and general relativistic effects (from moving things and from spending time under a mountain).

    I put the Bayesian prior for getting this wrong at 1E-3. (If I just read all the references, it probably says explicitly how they did the time transfer.)

  155. Kriton, valid points, but the idea that the neutrinos that we thought were from SN 1987A actually coming from another supernova that hasn’t reached us yet seems extremely unlikely. That would mean that by an accident of timing the neutrinos happened to show up exactly when we would expect them by conventional theory if they came from SN 1987A. That would almost be evidence for an evil deity that likes to mess with astrophysicists.

  156. Viadd

    I just read more of the paper, and they state that they did time transfer both by two-way fiber and by portable atomic clock, so my post #175 is invalid. Knock the Bayesian prior for time transfer error back to 1E-5.

  157. Jamey

    Issue re: neutrinos in this experiment, vs. neutrinos from SN1987a…

    If neutrinos are in fact tachyonic, then they have an imaginary rest mass, and adding energy *slows them down*. Neutrinos from SN1987a would be expected to have much higher energies than those from this experiment, and therefore their speed should be much closer to c – probably enough to wipe out most of the 4 year issue.

  158. PeteC

    I wish I had time and still had some of the ability to go through this paper myself; while I did physics at university I moved into IT after graduating, so after a couple of decades I’m more than a little rusty.

    Did they consider the weaker gravitational field the neutrinos passed through? Earth’s gravity decreases linearly as you head towards the center of the planet (there’s a shell of matter over your head effectively pulling you back up again). I know it’s not a lot, but as they travelled in a straight line underground they would have moved through a lower gravity field, which if I remember corrctly would result in a very slightly different time.

    It’s remembering the Hafele–Keating experiment, where they flew an atomic clock around in an aircraft, that reminds me of this; though of course, that was moving as well, so it suffered from increased dilation.

  159. Al Cibiades

    This all sounds pretty interesting — Even if “FTL” is NOT being demonstrated, there may be some startling discoveries in the wings — possible exception for Equivalence (10.1103/PhysRevLett.107.081102); Minimization of Uncertainty via entanglement (Science 316, 726 (2007);; Science 324, 764 (2009))

  160. M.L.

    “Now first off, if it were this would overturn so much physics that they may as well have discovered that gravity pushes, not pulls”

    Some have interpreted dark energy as evdience that gravity does indeed ‘push’ on large scales.

  161. SkyGazer

    Ok, my eyes can handle the impact of photons hitting them at the speed of light!
    But those neutrinos are scary.

  162. bigseph

    Perhaps if anyone in history had been able to test the one-way speed of light without using mirrors or redirection or hokey gamma experiments, this result wouldn’t be so surprising to most people.

    Photons themselves travel faster than our supposed SOL.

    Physics and cosmology’s reliance on each other as well as manipulation for the purposes of “propping up” various ideology has caused a pure science (physics) to be a slave to the whims of other tertiary sciences that actually DEPEND on physics for proper baseline assumptions.

    Interesting that the article mentions Galileo, because Galilean velocity fares better than Einstein’s when it comes to faster than light calculations.

    But oh no, we wouldn’t want to find that c isn’t constant, because we’ve long ago assumed that it is, and numerous experiments show that it can be slowed down. Now we see that there is no actual limit to potential speeds in physics, there was just a limit to Einstein’s theory. One day someone will measure the one-way speed of light from a distant star or planet and use a proper experiment to do so. The speed of this neutrino will pale in comparison.

    But by all means nitpick this research. People will keep getting similar results as technology improves. And the earth won’t explode by a neutrino reaching infinite mass because it moved faster than c. That’s an erroneous theoretical HUMAN boundary. Not a boundary of our physical universe.

  163. TedK

    To Karl Withakay, who asked if neutrinos would be affected by passing through Earth’s air: I’m glad you admitted you know nothing about physics because that question shows you know nothing about neutrinos. Those particles can pass through rocky planets without hitting a single atom along the way.

    Everybody thinks we’re going to have the USS Enterprise on our hands next week. God, would someone do something about the science education levels in the USA, please!!

  164. Infinite123Lifer

    Neutrino Neutrino had a great fall
    Neutrino Neutrino went through the wall
    Neutrino’s watchers expected the fall
    The Neutrino with its mass if at all
    Detected by CERN and that is the call
    If by mystery it traveled at all
    Faster than light! Response to the call
    The brightest and best and smartest of all
    Responding to CERN the craziest of all
    Testing to see the Bayesian priori fall
    All around the world they hear this call
    Its written in stone as up on a wall
    Sweet wisdom i cant let help you to fall
    The answers hidden and not in us all

  165. I would not look towards the distance measurement nor the timing system are the source of error, although advanced and precise they are relatively trivial technologies to comprehend and test and have been validated many times. I would look at the timer triggering mechanism and how it is coupled with the generation of the particles. The generation side is where a phase delay could create this result.

  166. Scott


    “…A heavenly body such as the earth displaces the fabric of space-time, and the result is an effect of tension against the surface of the earth. Matter is pushed down in an elastic-like effect. The tautness of curved space-time holds us down…”

    Uh, that description of the “fabric of space-time” sounds suspiciously like the “ether” which Einstein maintained does not exist. If the pressure of the displaced fabric is pushing objects, shouldn’t the pressure be higher on one side of the Earth than the other, due to the headwind encountered as the Earth travels through space due to the movement of the galaxy, the galactic rotation, and the orbit of the Earth around the Sun?

  167. Torbjörn Larsson, OM

    So … has everybody caught where they goofed yet?*

    It is an easy one. According to the paper the distance measurement procedure use the geodetic distance in the ETRF2000 (ITRF2000) system as given by some standard routine. The european GPS ITRF2000 system is used for geodesy, navigation, et cetera and is conveniently based on the geode.

    I get the difference between measuring distance along an Earth radius perfect sphere (roughly the geode) and measuring the distance of travel, for neutrinos the chord through the Earth, as 22 m over 730 km. A near light speed beam would appear to arrive ~ 60 ns early, give or take.

    Of course, they have had a whole team on this for 2 years, so it is unlikely they goofed. But it is at least possible. I read the paper, and I don’t see the explicit conversion between the geodesic distance and the travel distance anywhere.

    Unfortunately the technical details of the system and the routine used to give distance from position is too much to check this quickly. But the difference is a curious coincidence with the discrepancy against well established relativity.

    * Extraordinary claims need extraordinary evidence. Other outstanding concerns are:

    1. This needs to be repeated.

    2. It is not a clear photon vs neutrino race. Physicist Ellis and others here noted that the time differential for the supernova SN 1987A was a few hours, but at the distance of ~ 200 000 ly it should have been years if the suggested hypothesis would be correct.

    3. Analogous to the experiments where light waves seemingly travels faster than photon speed in vacuum, they don’t measure travel times of individual neutrinos but averages over a signal envelope. That must be carefully measured to establish that particles (or information, for that matter) travels faster than relativity allows.

    Especially since the neutrino beam oscillates between different kinds of particles!

  168. Scott

    “…A heavenly body such as the earth displaces the fabric of space-time, and the result is an effect of tension against the surface of the earth. Matter is pushed down in an elastic-like effect. The tautness of curved space-time holds us down…”

    Uh, that description of the “fabric of space-time” sounds suspiciously like the “ether” which Einstein maintained does not exist. If the pressure of the displaced fabric is pushing objects, shouldn’t the pressure be higher on one side of the Earth than the other, due to the headwind encountered as the Earth travels through space due to the movement of the galaxy, the galactic rotation, and the orbit of the Earth around the Sun?

  169. cruf

    A pretty straightforward generalization of Einstein’s rule is that nothing can accelerate to the speed of light, either from above or below. So subluminal things like snails and rocketships and electrons can’t speed up to c and superluminal things (mabe like these neutrinos they’re detecting) can’t deccelerate down to c. The problem with superluminal particle is it’s very hard to iinteract with them and detect them. Maye that’s why neutrinos are so hard to detect in the first place. Incidentally, the “rest state” of superluminal particles ought to have infinite velocity with respect to our reference frame. So they are basically everywhere at once. This sounds suspiciously like dark energy to me – we can’t detect it but it is apparently all over the place.
    If 73% of the total mass-energy in the universe is dark energy and another 23% is dark matter, that only leaves 4% for everything traveling slower than the speed of light. It would seem more natural for there to be equal mass-energy on either side of c. If the velocity of all the tachyons (> c) relative to us was infinite, they would appear to be everywhere all the time and their portion of the mass-energy budget would tend toward 100%. The fact that it is only 96% is a signature of their actual velocity distribution.

    So tachyons appear to have more energy than they actually do. And the excess grows as their speed increases, asymptoting to infinity as their velocity (relative to us) goes to infinity. This is probably a result of how we are interpreting the indications of dark energy’s presence – using inappropriate subluminal physics.

  170. jimthompson

    The SN 1987a argument is just really, really dumb since Kamiokande wasn’t sensitve enough to see them until 1985…Come on people, you have google, look it up!

  171. @188. TedK

    Please actually read my comment before going off on your rant. I know quit a bit about physics and neutrinos. I’ve actually tutored college physics.

    I never said I didn’t know anything about physics. I said I was not a particle physicist . Although they do not interact electromagnetically, neutrinos do interact through the weak force (and gravity) or there’d be no way to detect them. I am well aware of the massive numbers of neutrinos that pass through every square centimeter of this planet every second.

    It would seem that you either did not really read my comment, or you did not understand my meaning.

    “Is the index of refraction of the atmosphere the same for photons, which interact electromagnetically, as it is for neutrinos, which do not?”

    My point was exactly that neutrinos would not be expected to be slowed down by traveling through matter (have any refractive index) while light would, therefore it might be possible for a neutrino to exceed the speed of light in a particular medium. Perhaps I might have phrased it better so that it was easier for some people, such as yourself, to understand my meaning.

  172. Robin

    @ Bigseph (#185): Do you have any actual evidence to support your claims. You seem to display a lack of understanding about how the speed of light has been measured throughout history. That the measured speed of light seems to converge nicely over the history of its measurement seems to indicate the techniques for measuring it are appropriate. Certainly there is nothing inherently better about measuring the speed of light “from a star”, and in fact trying to do so presents quite a few challenges, one of the biggest of which would be constraining the experiment such that you were actually measuring the speed of light (directly or indirectly) and not some velocity or time that included factors not held constant or that were unknown.

    Physics is not slave to the whims of anything. Science is not slave to the whims of anything. People maybe slave to the whims of other people, and certainly people can willfully misinterpret or misstate scientific data or results, but that is something that is not part of science itself.

  173. Joseph G

    So is the accelerator actually creating a beam of neutrinos that are aimed at OPERA, or is OPERA just detecting neutrinos that shoot out omnidirectionally, as in a nuclear detonation? OP makes it sound like a beam, but I was under the impression that when neutrinos are created, they pretty much go every which way? In so, they’re not charged (obviously), so you can’t collimate them or anything…

  174. Chris Winter

    Doug Natelson wrote: “I vividly remember a physics colloquium at Stanford sometime in the late ’90s where the speaker put up a plot of some parameter. The plot was a straight line, and the y-intercept was supposed to be mass^2 for the neutrinos in that experiment. The intercept was clearly negative, though within the error bars. Someone asked about this, and the speaker (can’t remember who this was) basically said that with better stats this would almost certainly go away.”

    (Emphasis added.)

    Blas Cabrera, perhaps? I remember a lecture by him about neutrino detectors.

  175. Joseph G

    @185 bigseph: But oh no, we wouldn’t want to find that c isn’t constant, because we’ve long ago assumed that it is, and numerous experiments show that it can be slowed down. Now we see that there is no actual limit to potential speeds in physics, there was just a limit to Einstein’s theory. One day someone will measure the one-way speed of light from a distant star or planet and use a proper experiment to do so. The speed of this neutrino will pale in comparison.

    Erm… Showing that light can be slowed down has nothing to do with disproving general or special relativity. Special relativity has been experimentally proven literally tens of thousands of times (proven trillions of times, if you want to count technologies like GPS).
    If you’re looking for a way to catch those know-it-all scientists in a mistake, you could scarcely pick a worse theory to rail against. Try ranting about the MACHO model of dark matter, if you want a somewhat mainstream scientific idea to tear down (while not coming off as a total crank while doing it). You’d be a bit late on the bandwagon, as it’s already been pretty well disproved, but you’d at least be on the side of reality.

  176. bigseph

    @ Robin-

    The same “science” that has held to it’s dogmatic claim that the matter has been solved for the past 50 years? I’d also suggest that the speed of light has been measured at different speeds throughout our history, and there are at least a dozen cosmological problems when using our standard c in equations to explain universal phenomena.

    I’d rather not do your homework for you, there are numerous articles of people slowing down the speed of light as well as the previous neutrino study that was just barely found lacking. Feel free to cozy up to your laptop and visit a search engine.

    It seems to converge nicely that all the times I have looked for the sun at midnight, it’s nowhere to be found in the sky. Yet on the other side of the earth, they can see it just fine. Does that mean they are wrong? Hardly Having a restricted perspective causes people to look at an experiment testing the speed of light, bounce it off a mirror, halve the distance, and say “Voila!” How can you measure the one-way speed of light unless you are either equidistant and closeby the source or in some universally quantifiable position away from the actual receiver of the light?

    Ironic that you would be against a SOL test from light source to destination from a universal perspective, but you think we have accurate measurements from a light source to a mirror from an earthly perspective. Light exists in this universe independent of human existence. What makes you think it should bow to your theoretical ideas because it just so happens to be what YOU experienced here on earth?

    True science is observable/repeatable/testable. When multiple experiments record particles moving FTL or demonstrating the ability to slow light down, perhaps it’s time to scrap the plan of plugging c into all of your equations. We’ve manipulated it ourselves from a limited perspective on earth. That alone makes c hardly a universal constant.

    I have nothing further to say on the subject, assail my comments as much as you like. Toodles.

  177. Tim

    Just to be clear, the Kessel run wasn’t significant because it was fast. It was significant because it was short.

  178. CScott

    Before ruling out a surveying error did they consider that the detector is a Km or more below the surface. At that latitude a verticals will be out by 600-700 seconds of arc due to the Earth’s rotation. Over a depth of a Km that could give an error of a few meters.

  179. Joseph G

    @202 Tim: Eh, I think that’s a clumsy retcon. The most plausible explanation (other then the writers just not giving a womprat’s ass), is that Han was bulls***ting, trying to impress people he thought were ignorant rubes with some made-up impressive achievement. Given his personality, I think that explanation makes the most sense.

  180. davem

    What Geoid are they using to measure the distance on the GPS? WGS84? If so, how do they know that their particular part of the globe conforms to it? We know that WGS84 is an approimation to reality. Has it ever been measured accurately? While GPS can get to cm accuracy, that’s over smaller distances, where the actual shape of the planet doesn’t matter so much.

  181. Ema Nymton

    205 comments, and the small handful of crackpots are exceedingly obvious. Even with disparate ideas, the tenor of their rhetoric is like a big blinking beacon.

  182. Wally-Bob

    Not being the caliber of many that have posted here but having studied relativity in college, looking for an explanation of the following:

    I am an observer at position=origin @ velocity=0 (by definition as all is relative)
    To my right I observe an object “A” traveling away from me @ 0.6c.
    To my left I observe an object “B” traveling away from me @ 0.6c.
    A, B and me are collinear points.
    A sees me as B does also.
    A and B do not see each other as each is beyond the other’s luminal event horizon. (red shifted to wavelength = infinity and/or light never arrives at the other and/or each is the other’s dark energy, which ever one prefers)
    Changing the relative reference frame to A, I am departing at 0.6c and B is departing at 1.2c.
    Obviously my next question is: Why not?

  183. Mind

    Of course there is something faster than light…it is our mind…in my mind i can travel through one point in the universe to another point in the universe in less than one second….thats really fast, isnt it?…;)
    The Mind

  184. eduard

    what would happen if the lab’s budget was to be cut and then they claim this?

  185. Joseph G

    @209 Wally Bob: I probably am less well read in this area than you, but I’ll take a wild stab at it. My understanding is that light appears to travel at the same speed to observers in all reference frames. The light passing between A and B would be redshifted (a lot) but it would still get there. Time dilation doesn’t increase linearly, it increases asymptotically with proximity to C so even if the relative speed of A and B is 1.2 C the redshift would not be infinite (in fact, I believe the redshift would only be by a factor of about 2).

    Hopefully someone can correct me there if I (almost certainly) am wrong.

  186. Voltaire2

    If another intelligence were trying to get someone’s attention, might not playing with some fundamental forces of physics be one way? It’s still a cool idea, regardless of its reality or lack thereof. :^) Perhaps we all better read Einstein’s Bridge.

  187. Michael Bryant

    I’m puzzled. I’m an accountant but I read years ago about neutrinos travelling FLT. How come I seem to be the only one who remembers?


  188. redshift

    Comment 192 suggests arc vs. chord error to br 22m, equating to 60ns time error.
    Comment 203 suggests depth below surface to have measurable effect on distance.
    Comment 205 discusses choice of geoid as a source of error.

    Let’s just scotch these with some proper maths:

    We are assuming they made the schoolboy error of measuring an arc instead of a chord.
    Mean radius of earth 6371km.
    The guys measured 731.278km between the two points.
    Angle subtended from earth centre = 0.11478rad.
    Chord length (2rsin(angle/2) = 730.877km. Error 0.401km.
    Time for light to travel 401m at 300 000 000 m/s = 1339ns, not in the same ball parks as 60ns.
    As for being a km or so down or using the wrong geoid, that’s nothing compared to the variation in earth’s radius. I used the mean 6371km. The actual range is 6353km to 6384km. Repeating my calculations based on these figures gives errors between 400m and 404m, which equate to time differences in the range of 1346ns to 1333ns, therefore it doesn’t matter what geoid is used or if they were a few km down, the variation is miniscule.
    Read the report. There are no obvious errors.

  189. Al Viro

    re photons having non-zero mass: that would make velocity dependent on
    frequency, so we’d have X-ray detectors catch supernovae before optics
    and radio… SN 2006gy was ~7e7pc away and it had been observed both
    in X-ray and optically; if optical-range photons were, indeed, moving
    at about (1-1/(4e5))c, the X-ray ones would be at about 1-1/(4e10) or
    so (for nanometer wavelength ones; harder ones would be faster). For
    that kind of distances that would amount to optical part reaching us
    ~50000 years after the X-ray observations…

    Moreover, I strongly suspect that we’d have noticed already simply for
    red vs. blue light – 1-(v/c)^2 is changing as E^-2 and for velocities
    that close to c it means that (c – v) is proportional to E^-2. So just
    switching from red to blue light would decrease c – v by factor of 3
    or so. Not to mention that VLF radio would be *seriously* slower than visible
    light, with all kinds of interesting effects caused by photons slower
    than hyperbolic relative to Sun 😉

  190. redshift

    Einstein wasn’t comfortable with the discrepancies between relativity and quantum physics. If relativity is actually flawed, maybe it’s the gateway we need to finding a grand unifying theory.

    What if the c in e =mc2 were not the velocity of light but simply taken to be velocity of the particle being measured? Sub-atomic particles with lower mass than a photon (e.g. neutrinos) could then have a higher velocity for a given energy. A theoretical zero-mass particle would be pure energy and could have infinite velocity (gravity?)

  191. Wzrd1

    From the graphs in the report, it seems as if some of the neutrinos may be arriving far before the main train, another group in front, then the observed reported main groups.
    For all the world, it looks like a phase shift of an arriving signal, added to the premature arrival.
    Now, WHY would they arrive early? If there were a fourth flavor of neutrino that was massless, it would have no effect, they’re not accelerating. Tunneling, perhaps, during oscillation?

  192. BSteele

    Good point about the supernova. What if neutrinos travel faster through matter than through a vacuum, though? Has this ever been tested? I’m asking you because you seem like you might know.

    Thanks a bunch!

  193. HowardW

    Comment 130 (Duncan Young) asks for reference [26] Colosimo et al. concerning geodesy. It may be found at this location:

  194. gayathri
  195. karishma

    can i know more about this?

  196. Deepak

    Apart from the implications to the physics world, can someone kindly explain the implications on the day-to-day life if this turns out to be true?

  197. T. Peterson

    Perhaps someone has already made this point in the comment section: these researchers weren’t making any claims (as you say twice in the article). They were presenting their results for others to verify.

  198. hollypahl

    I beleive that “Entanglement,” proves the transfer of information at vastly grater speeds than the Speed of Light.

  199. Hrune

    I have a logical conundrum: Results that are against relativity are based on GPS measurement which is based on relativity:).

  200. Albert

    I’m just an amatuer but is it possible that it’s going thru a warped space so it’s traveling less distance than the actual distance hence the faster time???

  201. Giancosimo

    Perhaps they have measured the distance on a circumference arc but neutrinos travel on a chord internal the Earth?

  202. CScott

    700 seconds of arc over 2 Km gives 6.79 M – not enough, but it is a start. Since you can’t take a GPS measurement 2 Km underground extrapolating from the surface is going to be a weakness. How exactly was the location of the underground detector determined? Conventional surveying techniques are really designed for use on the surface.

  203. Until now it has been assumed that the maximum speed of information propagation (c1) inside subatomic particles is equal to the speed of light in a vacuum (c0) but there has been no means of testing that assumption.
    Photons cannot travel through nuclear particles but neutrinos can, and usually do. So the observed timing may be giving us the first indication of the value of c1. It may be showing that neutrinos move very rapidly across nuclei traversed in rock. We need to know the average proportion of the neutrinos’ paths through the Italian rock that are within nucleons, to see whether this would account for the phenomenon.
    If this is correct our ideas about neutron stars and black holes would need revising but the space-time in which we live remains Einsteinian.

  204. Switzerland to Italy… Maybe Italy’s horrific fiscal deficits combined with Switzerland’s strong franc have created an economic gradient down which those Swiss neutrinos accelerate? Perhaps if they fired them toward Greece, the overage could be well north of 100 ns. Just a hypothesis from someone whose portfolio just got poleaxed to the tune of 20% this week.

  205. Alan D

    It’s nice that they have put their results out for scrutiny by other physicists. Sometimes, I think, people are too close to something, and some avenues of checks or verification can get overlooked.

  206. Wayne Jones

    Twitter version: GPS calc -> straight line distance btween source and target – what if path of neuts was curved due to gravity’s small pull on them?

    FB version:
    Setting aside any difference between the mass of a neutrino and that of a photon:
    1) the path of the neutrinos was a straight line, there was no tunnel because none was needed, and there was no “frictional” interaction between the neutrinos and the atoms in the earth’s crust through which they passed — right?
    2) Gravity is a warp of space/time and has a vector component for each center of gravity — right?
    3) Although the crustal materials and all other visible and dark matter in the universe exerted gravitational force with individual vectors calculated by the centers of gravity of whichever aggregates suit the convenience of calculations, the largest vectors working on the neutrinos were the vectors between each neutrino and the center of the earth, along the line that swept through a computable angle with the center of the earth as its vertex — right?
    4) Oh . . . maybe number 1 above is not correct — maybe the neutrinos followed and flight path that approximated an arc because on average the force of gravity felt by each neutrino was constant; however, that would make the path of the neutrinos longer than the straight line indicated in the diagram and assumed in the +/- 20 cm accurate GPS calculation of the distance between source and detector, and if longer, the problem of the neutrinos going faster than c would get even worse — oops (maybe – but we’re talking about tiny differences here any way, so I would think such considerations ought to be addressed and dealt with as inconsequential [too small to matter] or additive or misguided).
    5) The question I have been trying to get to: if each neutrino can be thought of traveling in its own gravity well (the warp of space/time that is gravity), is the measurement of its speed between points A and B affected by the fact that the vector associated with that gravity well sweeps through an angle relative to the greatest source of the gravitational field (here the center of the earth)? It’s clear that gravity affects photons – hence, gravitational lenses – the standard example of gravity as a space warper. Given the gravitational lensing example, what is the effect on the flight path of a neutrino in the experiment of the interaction between the gravitational deflection of the neutrinos and their inertia (the light we see that demonstrates the lensing makes it around the deflecting gravitational fields – except those passing a black hole within its event horizon – where their inertia is not great enough to overcome the gravitational vector of the black hole)?

  207. Hmm..

    On @Quora: How does one verify that the Superluminal neutrino in question is infact was from CERN?

  208. Completly Ridiculous Hypotosis

    I think this is just a result of virtual pair production by light. When it produces a pair its energy is transferred slower giving it an average velocity less than c. While light is actually traveling at c, in between pair productions but its energy is transferred by the pair at a slower v depending on the energy of the photon, so its averaged velocity is slower than c. But neutrinos don’t pair produce so they can have a “superluminal velocity” which is slower than c. Presuming the experiment stands. QFT for the win.

    219 you’re talking about kinetic energy and Newtonian theory which is equivalent to relativity with c=infinity.

    209 you’re using Galilean velocity addition formula but in relativity they would appear to be traveling away from each other at 0.88c so the light frequency is halved and the wavelength doubled.

  209. harry

    astropaul@47, while the observed speed of a photon would not have changed due to the motion of the observer, these experimenters didn’t measure the speed (of the non-photons). They measured the arrival time of them and then computed a speed from it, based on a presumed distance. That distance changes as much as 71m in the 2.4ms flight time of the neutrinos due to the Earth’s orbit around the sun. Astronomers have to correct for the aberration of light due to the Earth’s velocity when they measure the direction or frequency of photons. Arrival time of a wavefront is also dependent on the motion of the observer. The Opera team did not report on accounting for this motion which could cause a much larger effect (depending on the relationship between the vector from CERN to Gran Sasso and the Earth’s velocity vector at the time of the neutrino creation). It is a glaring omission.

  210. Bamber

    Umm, dunno much about quantum physics, but did the CERN folks take into account Earth’s 30km/s movement on the orbit around the Sun? As far as I know there is the invariance of the speed of light regarding light and electromagnetic waves, which light is, but when we’re talking about neutrinos, do they fall into the same invariance law’s yard?

  211. Hah

    I like how far off the mark the estimates are in the original article. They need to know the distance to an accuracy of 3m, if it’s off by 20m it’s too much. 20m is 3 orders of magnitude larger than the 2cm to which they know the distance. This is the problem with very smart people from outside the field commenting on experiments, models and results, such as we see all the time in Climatology.

  212. David Kahana

    Regarding the photon having a non-zero rest mass: this is certainly a possibility. But there are some pretty stringent experimental bounds on the photon mass. These bounds are both direct and indirect. By measuring the frequency dependence of the speed of light in a laboratory the best bound available is m_photon < 10^-10 eV. The indirect tests are all based on the point that a non-zero rest mass for the photon would modify Coulomb's law by a Yukawa like screening factor and also, similarly, change the behaviour of various static magnetic fields. So there is a stronger limit of m_photon < 6×10^-16 eV from measurements of Jupiter's magnetic field. There are even stronger limits from galactic magnetic fields, but those are generally considered not to be trustworthy, since the mechanism of generation of the galactic magnetic field is not well understood. The strongest earth bound laboratory limit that I know of is one that determines the product of the photon mass squared and the ambient cosmological vector potential measured extremely accurately using a specially designed toroidal Cavendish balance and finds m_photon < 7×10^-17 eV for specific estimates of the ambient cosmological vector potential. That experiment was done in about 1998 by Roderic Lakes. Photon masses this small would certainly imply a value of c in the Lorentz transformation that is far too little different from the currently measured value of c to accommodate the supposedly observed anomaly in the neutrino travel times. So that can't be the explanation. Similarly, the mean free path of a 1 MeV neutrino in solid iron is on the order of 1 light year, so the fact that there is rock in the way of the neutrinos is not expected to make any difference at all.

    Personally I'm in the camp that holds that the experimenters have simply missed something in the analysis, that they have missed some source of systematic error. The statistical nature of the analysis, the fact that the experiment is not actually capable of determining the travel time of any individual neutrino since there is no very firm knowledge about where or when any individual neutrino detected at Gran Sasso was produced by decays of pions and kaons in the secondary beam, and the fact that the zero point for time is established only by a measurement of the proton beam pulse timing and pulse waveform at the CERN SPS, all of these leave the method open to many possible sources of systematic error. In my view, what is needed for such a measurement to be convincing is to have two neutrino detectors in the neutrino beam: a near one and a far one. Then you can measure the transit time from the near to the far detector. That will eliminate the need to know as much about the time structure of the primary beam.

  213. Guy who lived in Japan

    Hey, great article, and thank you for explaining how science works — I wish our education system did as good a job as you do.

    Anyhow, you said “scosh” to mean a little.
    I think it should be spelled “Skosh”, as it comes from the Japanese word “sukoshi” which means “a little bit”.


  214. Stephen

    @#51 James “I love how science turns ‘being wrong’ into a positive experience ;-)”

    You have it wrong.
    What is good is finding out that you were wrong.
    What is bad is staying wrong.

    Faith keeps you wrong.
    Science moves you forward to the truth.

  215. Kieran Miles

    1st Question: –

    When the LHC went online they said that there were possibilities that a wormhole might be produced and such wormholes would either dissipate or move to the centre of the earth, if such wormhole were to move to the centre of the earth would this have an effect on the neutrinos in anyway? Wormholes have no mass they consume mass and cause space and time to warp funny, this is just a question not trying to scare anyone here.

    Please see link below: –

    2nd Question:

    Taking into account the link I just posted would the fact that space time is being distorted by the earth itself causing an arch (dip) effect cause neutrinos to react/act different as the earth causes space/time to distort. Could this distortion cause neutrino travel times to distort as they are passing through the earth? Would this not be a likely explanation?

    3rd Question: (not really a question more of a theory about the speed of light which I posted to a friend on facebook)

    My theory of how one could travel faster than light involves part of the experiment I posted above and surfing. Take a massive object like the size of a sun, have it spining as fast you can then drop it as hard as you can and detonate it, if you could cause a ripple (wave affect) in the fabric of space and if you could surf that wave you might be able to travel the speed or faster. What you have to remember is that the wave is the fabric of space itself not matter, not particles, the universe is stretched like a rubber band, so this idea is allowed in the theory of relativity as space is not matter. Then you would need to find something which reacts with the fabric of space to be the surf board and then all that’s left is to survive the ride. Have you ever had a skipping rope and flapped it up and down the harder you flap it to begin with the quicker the first wave travels down the rope, it also dissipates as it moves down the rope.

    So what do you think?

    Please come back to me on these points, i welcome it, if it’s sensible and objective.

  216. ab

    @181: The effect is a number of orders of magnitude too small (fortunately or unfortunately)

    Too all who are hoping this is true and implies a future warpdrive (as I do): Think on the implications regarding the Fermi paradox

    Potential ideas from top of my mind:
    * speed of light in vacuum is “dressed” and below a “naked” speed of light due to interaction with virtual particles. See Scharnhorst / Barton effect. Maybe neutrinos interact more weakly, thus are less “delayed” – since it does not carry charge there is no Cherenkov radiation
    * neutrinos being able to partially go through some additional dimension. If so, why only neutrinos? I do not like that
    * neutrinos having negative squared mass and i being a real thing with regard to mass: not as mad as it sounds – a number of “calculation helps” in physics have been found to have real meaning, e.g. see Aharonov Bohm effect

  217. Dave Nixon

    The sun orbits the galactic center at a speed of 250km/s and the Earth orbits the sun at 30km/s. Due to earth rotation the actual direction the neutrino path is pointing relative to the direction of the Earth’s motion varies, but obviously the Gran Sasso target moves some hundreds of meters during the 2.4ms flight. Neutrino path is north-south approx at 42 degree latitude (Gran Sasso) so any net motion of the earth to the north results in decreased distance traveled since the target is closer when the neutrinos hit it than when the neutrinos left CERN. Galactic center is 29 degree declination so is always in the northern hemisphere (ignoring precession). Sun’s vector is apparently 60 deg approx away from the galactic center so on average the Sun (and the Earth) is moving more to the north than to the south. Need to work out the exact geometry but this seems a plausible explanation of why the neutrinos are traveling a smaller distance than terrestrial geodetic measurements.

  218. Harry

    It seems to boil down to this: If they know precisely when the particles left CERN and they know precisely when they arrived at the detector and assuming the particles measured were the same particles at both ends then the results are accurate. That being the case we would have a new branch of science with the goal of determining how they did it, reproducing it and up-scaling it for some practical use. Maybe one day some brave soul will do to light what Chuck Yeager did to sound. Who knows, it could be just what Stephen Hawking needs to bag the theory of everything.

  219. amber

    Not fair!! i was just doing a physics report on neutrinos and i find THIS! maybe it could mean a supernova will occur which is why there are neutrinos whizzing around faster then speed of light or maybe this can explain something about blackholes..

  220. I truly wish this wonderful, dedicated team of physicists to be finally proven right! Here is their current full length publication with about 60 names on it:

    I read it and the physics is clearly well grounded and professional.
    The determination of the distance is taken from a professional geodesy
    group, who constantly report the arc distance between the starting point
    at CERN and on the surface near the detection point along the Gran Sasso

    The paper states that the GPS coordinates are measured with cm accuracy.
    No word is given on the difference between the arc distance along along the
    Earth’s perimeter (D[Earth Arc], which is longer, then the distance between the
    same two GPS points (D[Neutrino]), the one traversed by the neutrinos.

    Given the extremely important scientific implications the full paper on the
    “dedicated geodesy on the ETRFS 2000 reference frame” should be disclosed!

    As an extreme example, if the two points were at opposing half of the earth surface at the equator, then D [EarthArc] would be R*3.1415 ~ (R*π) while the straight line cutting through the Earth would be D[Neutrino] = 2* R. The difference in this example is R*1.1415 ~ (R*(π-2)).

    The distance between the 730 km arc “as the crow flies” and the 730 km
    point-to-point straight line distance can be calculated. Its approximate value is ~2800 cm.

    Useful arc calculations are here:

    Here is a reference on how ETRF89, ETRS2000, and ETRS2005 geodesics work (in detail). All presume finite steps of stright line on Earth surface.

    It would be worth to dig here a bit more and see how these different ETRF, ETRS geodesic systems link first on national level within Italy, and then with Switzerland!

    This is worth reading. Neutrino science and geodesics are mastered by different people.

    4.1 ETRS89 and its realizations
    The local reference system suggested for mapping applications
    in Europe is the European Terrestrial Reference System 1989 –
    ETRS89 (Boucher and Altamimi, 1992), based on a densification
    of the global frame ITRF. The ETRS89 definition has its rise in
    the cartographic requirements: an European site generally moves
    of about 2-3 cm/y with respect to the global system ITRS; this,
    even at the scale of only ten years, produces a displacement that
    is significant at least for the large scale cartography: one purpose
    of ETRS89, therefore, is to reduce this issue in order to obtain
    reference frame as crystallized as possible. ETRS89 is then de-
    fined as coincident with ITRS at 1989:0 epoch, but it moves and
    rotates with the so called stable part of Europe.
    At the continental scale, ETRS is realized and monitored by the
    EUREF Permanent Network (EPN) and is updated in line with
    ITRF. An exception is the realization 2005: for complex technicalities, not discussed here, the ETRF2000 is still the current
    European frame. ETRFyy is distributed with the estimated coordinates and velocities of the EPN permanent stations, at the reference epoch and the transformation parameters between ITRFxx
    and ETRFyy (Boucher and Altamimi, 2007).
    4.1.1 The Italian national ETRS89 realizations Many nations have their own official ETRS89 realizations: in Italy, until the last year, it was materialized by the IGM95 network of
    the Istituto Geografico Militare Italiano, (IGMI); the network is
    composed of about 2000 markers, surveyed in the nineties and
    adjusted in ETRF89 (Surace, 1997); IGM95 is distributed by the
    IGM monographs. In Italy, differences of some decimetres exist between IGS05 and ETRF2005, with an increasing rate of
    1-2 cm/y and deformations of some cm between the Northern
    and Southern regions, due to differential geodynamical deformations a the national scale (Jimenez-Munt et al., 2006); moreover,
    due to surveying instrumentations, data analysis techniques and
    elapsed time, IGM95 presents additional, spatially correlated deformations and randomly sparse errors, whit a RMS of 3 cm in
    planimetry and 5 cm in altimetry.”

    The trivial (simplistic) explanation can be that these neutrinos KNOW their speed and the CERN-OPERA team ended up measuring the TRUE distance between CERN and OPERA!

  221. Marcelo Miller

    I agree with Dave Nixon. From the distance between CERN and Gran Sasso you know that light takes about 2.4×10^-3 s to travel. Then, at those latitudes the earth rotates with a surface velocity of about 3.3×10^4 cm/s, what means that the laboratories moved about 22 meters in the time the light travels from cern to sasso ( 2.4×10^-3 s).
    Close to the speed of light that distance of 20 meters means ~70 nanoseconds… VERY close to the “measured” delay. Of course these are very rough numbers using newtonian physics and no relativistic corrections. But this order of magnitude estimate makes me wonder that the faster-than-the-speed-of-light neutrino measurement is just about some mistake in the corrections for the rotation of the earth.

  222. DaveW

    …. I’ve always thought that time was discrete and not continuous… these guys have discovered quantum aliasing?

  223. Dave Nixon

    Marcelo #261 raises the rotation of the earth as an extra motion over and above the earth’s orbit around the sun and the sun’s orbit around the galactic center that I mentioned in #256. I think the target speed due to the earth’s rotation (330 m/s) is much smaller than the other two motions that are 30 km/s and 250 km/s respectively. Also the earth’s rotation is at right angles roughly to the neutrino path, so the change in the distance the neutrinos actually travel because of the target movement due to the rotation is quite small. However the earth’s rotation is crucial in analyzing the neutrino path direction relative to the absolute motion of the earth due to the other larger motions. If the earth is travling more towards the pole star (north) than the other way, then Gran Sasso is on average closer to the neutrino start point at CERN after the 2.4ms flight than it was when the neutrinos left CERN. So need to understand the geometry in detail, but it seem likely this makes the FTL effect go away.

  224. David Kahana

    Marcello: I think you’ve gotten things quite wrong there. First of all the rotational velocity of the earth is much smaller.

    Here’s a rough calculation: Equatorial circumference of earth C_e = 40,075.016002 km.
    Latitude of Gran Sasso ~= 42 N and cos(42) ~= 0.743144.

    So the rotational velocity at Gran Sasso = C_e * cos(42) / 86,400 s = 0.344693 km / s.
    Thus in 0.002433 seconds Gran Sasso moves about 0.000839 km towards the east along it’s line of latitude, or about 0.84 meters.

    A correction would need to be applied to that number though, which reduces it, because the chord from CERN to Gran Sasso has a significant North-South component. The coordinates of CERN are about 46 N and longitude 12 E, while the Gran Sasso mountain is at 42 N and about 6 E.

    At any rate, the size of any ground motion from rotation during the neutrino transit is at most 0.84 m.

    But now consider the following: the earth also moves in it’s orbit during the transit of the neutrinos from CERN to Gran Sasso. Now, how big is that motion?

    Well, the mean radius of the earth’s orbit is 150 million kilometers, and it’s average orbital speed is about 29.78 km/s. So in 2.43 ms, Gran Sasso moves 72.454 meters along the earth’s orbit.

    Moreover, the sun is moving relative to the galaxy, and the galaxy is moving relative to other galaxies, and these are much higher velocities than the orbital velocity of the earth.

    Now, ask yourself: do you really need to apply all of these “corrections” for the “ground motion” of Gran Sasso to the neutrino arrival times? Or is there something simply very wrong with this whole line of reasoning?

    Here’s the trick: you should carry out the whole analysis in one reference frame, the frame that makes the analysis the easiest for you to do. A reference frame that moves instantaneously with the orbital velocity of the whole earth in its orbit at the time of emission of a neutrino at CERN might be a reasonable first choice. It’s clear that in such a frame the relative distance between CERN and Gran Sasso is essentially fixed, up to tidal and other potential motions of the earths crust, and the neutrinos are – theoretically that is, I’m going to blithely assume that special relativity is still correct, here, despite what these crazy guys from Gran Sasso are suggesting – emitted with an initial velocity that is the relativistic composition of their velocity due to the earth’s rotation at CERN with the velocity at which they are produced. Now, a couple of milliseconds later, when the neutrinos arrive at Gran Sasso, the earth’s orbital velocity will be slightly different, it’s true and the earth moves in a very slightly curved path over that 72.454 meters that I calculated … but it will be only very, very slightly different from a straight line. So this will only produce tiny corrections — proportional to the curvature of the earth’s orbit. In this inertial reference frame, the neutrinos will simply move in straight lines.

    But better still, you could carry out the whole analysis in a frame that co-rotates with the earth. An earth fixed frame is not quite an inertial frame of course, and there will be corrections to straight line motion for the neutrinos, because in an Earth fixed frame the neutrinos don’t move exactly in straight lines. There will be a Coriolis effect – this is one of the fictitious forces that exists in a constantly rotating reference frame in Newtonian physics.

    There will even be general relativistic corrections to the neutrino paths due earth’s gravitational potential, and other people above have pointed this out correctly.

    But the basic point to understand is that all of these corrections are calculable and they are very, very small relative to just assuming straight line motion at constant velocity for the neutrinos. Because when the neutrinos are emitted, the velocity of CERN has to be added to their velocity, and CERN and Gran Sasso are BOTH moving with the rotational and orbital velocities of the earth in such a way that their relative distance doesn’t change. It’s the relative velocities and distances that matter – that’s the meaning of relativity whether it’s Galilean relativity, or Einsteinian special relativity.

  225. darck

    Eisenstein teori was only for the FIZICAL reality the mass univers not the quantum , the thing is that the world that we all know and observ isnt all that real . I am surprised that the faster that light article was even published. …

  226. Dave Nixon

    Re: #264 David Kahana – I think you’re neglecting earth’s orbit around the sun and the sun’s orbit around the galactic center which are not so small – 30km/s and 250km/s respectively. In the galactic frame, during the 2.4ms from the neutrino production at CERN, to them hitting the target at Gran Sasso, Gran Sasso has moved several hundred meters in some direction that depends on the orbital geometry and state of rotation. Also I don’t think CERN’s velocity matters. Once produced, the neutrinos are going to move at the speed of light no matter what CERN’s velocity.

  227. Lemme Tellya

    Nothing here points to exceeding the limit of the speed of light. Assuming everyone checked their math it means one of two things. Either the distance or timing is off. Einstein came to his conclusions not by using measurement, but by logic. Nice try, though.

  228. Jeff P.

    Brian Cox has suggested that this may not invalidatate Einstein, other than the fact that the neutrinos may have passed through more than the standard 4 dimensions. I would think the discovery of a 5th or more dimensions very significant in itself.

  229. David Kahana

    Re: #267

    Dave Nixon, no. I’m not neglecting anything. The motions of the earth with respect to the sun, its rotation on it’s own axis, the motion of sun with respect to the galaxy, the motion of the galaxy with respect to the center of mass of the local group of galaxies and of the local group center of mass with respect to the cosmic microwave background radiation are only relevant to this calculation of the expected neutrino flight time from CERN to Gran Sasso to the extent that a reference frame in which every one of these individual motions I listed is set to zero deviates from a perfectly inertial frame.

    The non-inertialities in this case will be extremely small: of the order v^2/R for the motion of the earth’ in its orbit. Similarly for the solar and galactic motions. They are very close to constant motions over the time scale of the neutrino transit. The biggest deviation may be that due to the earth’s gravity in fact. But in 2.43 milliseconds (approximately) neutrinos will not fall very far towards the center of the earth. It doesn’t make sense to talk about “ground motion” of the earth unless you say precisely what it is that you’re measuring that motion with respect to. Specifying that precisely will define your frame of reference, and then you’ll need to do the whole calculation consistently in that one frame.

    Now let’s suppose for just one moment that we are back in the world as it was before we heard this rumour that neutrinos might go faster than light. Back in that world, the world that I suggest we’re actually still living in, the world where special relativistic kinematics has always worked to extremely high accuracy, neutrinos are known to have a mass, because they display flavour oscillations.

    So in that world, the one in which special relativity is strictly true, neutrinos do not move at the speed of light. Instead they move almost but not quite at the speed of light: they actually go a little bit more slowly than light, depending on what their energy is. So that’s the time of flight calculation you’ld need to do if you wanted to be absolutely precise.

    If I chose to work in a frame in which all of the motions you list, except the earth’s rotational motion, say its rotation with respect to the background stars of the milky way, rather than with respect to the direction of the Sun are all zeroed out, then I’ld need to use the relativistic velocity addition law to add the velocity of CERN in that frame to the velocity of the neutrinos in that frame, to find the actual velocity of the neutrinos in that frame. The resulting velocity will have a speed which is almost, but not quite, the speed of light. Of course, the actual deviation from the speed of light would be very, very small. But it wouldn’t be zero.

  230. Hannibal Four-Vector

    Suppose that some particle can travel faster than light. Then all adjacent events on the particle’s world line are space-like separated. Since the neutrino travels freely, its world line is straight. There exists therefore a frame of reference in which all events on this world line are simultaneous. This means that the particle is not, then is(everywhere at once), and then again is not – like the an antimatter version of the beast in the revelation of St John the Divine.

  231. Nigel Depledge

    Betz (171) said:

    They go through several km of rock. They have to be passing through N atomic nuclei.

    Well, I suppose this is technically correct, but N is probably a lot more likely to be zero than you might expect. The nucleus is typically only about a millionth (give or take an order of magnitude) of the cross-sectional area of an atom.

    And your neutrino would have to interact with the nucleus for it to make any difference, and I have no idea what the chances of that actually are. Never forget that billions of neutrinos pass straight through the entire Earth every day.

  232. noreneherman

    For instance, even the speed of light might turn out different in different directions in space, depending on your viewpoint, and depending on how you do the timing.

  233. prianikoff

    Amazing how many competing theories are now being aired on the basis of one experimental result that’s yet to be replicated!

    More likely there’s an error; either in measuring the distance S from A to B, or determining t, when the neutrinos were first produced at A.

  234. Robert

    I’ve seen several people claim that the experiment might be explained if photons had some small rest mass, and were moving at less than the relativistic speed limit, while neutrinos move at this limit, or at least faster than the photons.

    If this were the case, relativity predicts that photons of different energies would move at different speeds and for a far off event like a supernova we would see the high energy photons arrive before the low energy ones. This is not the case.

  235. SN1987A

    Wow, I am suddenly a superstar again. Everybody is talking about me. Everybody seems to know where I live and uses simple arithmetic to calculate how long would it have taken for my magnificent radiance to arrive in the wake of 24(!) neutrinos back in 1987 if the CERN guys wound their clocks up regularly. 3 hours instead of 4 years? Well, maybe I live a lot closer than you think. Or may be neutrinos can travel even faster.

    What you call light does have a speed limit in its what you call medium. So does the sound in its own. Bang on one end of a lumber and it will take time to hear the sound from the other. Flash a torch at one end and it will take a lot shorter but still a finite time to reach your eyes peering over the other end. Push one end and the other will slap your bewildering face (Ouch!) with no delay. Simultaneous signals, different mediums, different speeds.

    You learned how to measure the speed of sound long before you learned how to measure the speed of light. “M” limits the speed of sound wave exactly as “C” limits the speed of electromagnetic wave. It does not stop you from flying at supersonic and even hypersonic speeds. It is just time to learn about the first of many mediums beyond the time-space continuum and how to observe and measure the processes happening in it rather than discard the observations only “because Einstein said so”.

    Einstein, a truly great scientist gave you a perfect tool to model and predict the three-dimensional world and this model will be valid for you as long as you choose to have the three-dimensional perception of the Universe. The Universe is what you think of it. Any model you device will be valid within its framework and does not gave to be so without.

    Just lift your gaze off the calculator and look at the stars. We are not far away, we are not close we are you and you are we. A moment we send our light, your world changes with no delay (a kiss rather than a slap) but you may choose to observe this light through the models of space-time curvature, strings or gravitational lenses – totally up to you. All these devices distort the awesome beauty of the real Universe and create the illusion of space. Then you divide this illusion by time that does not exist to compute speed (DivideByNothing exception).

    Welcome to the 4th dimension.

    Sincerely yours, SN1987A. Love, Light.

  236. Nigel Depledge

    Bigseph (185) said:

    Photons themselves travel faster than our supposed SOL.

    [yadda yadda]

    But by all means nitpick this research. People will keep getting similar results as technology improves. And the earth won’t explode by a neutrino reaching infinite mass because it moved faster than c. That’s an erroneous theoretical HUMAN boundary. Not a boundary of our physical universe.

    OK, so I haven’t yet read all the comments, but I assume you are going to back up your claims with some evidence and reasoning, rather than simply make them and expect us all to believe you.

  237. Dave C

    I haven’t yet read the paper but I’m sure these headlines would be better stated as, “CERN Scientists Discover Error in Survey Quality GPS Measurements”.

    Whilst the best GPS measurements may have a resolution of around 10mm over short distances, I doubt that its absolute accuracy is anywhere near that over the distances involved in this experiment.

  238. Nigel Depledge

    Bigseph (201) said:

    , there are numerous articles of people slowing down the speed of light

    This is not relevant, and if you understood the meaning of c, you would know why.

    I could follow your unhelpful example and simply say “I’m not going to do your homework for you”, but consider this a freebie:

    c is the speed of light in a vacuum. No-one has ever slowed it down, nor claimed to do so. Light slows down in various media because it interacts electromagnetically. Hence, transparent materials all have a refractive index.

  239. Nigel Depledge

    Wally-Bob (209) said:

    Changing the relative reference frame to A, I am departing at 0.6c and B is departing at 1.2c.
    Obviously my next question is: Why not?

    I did not study relativity at college, but I think I can answer this one.

    Nothing prohibits the relative velocities of A and B in your example (I’m ignoring time dilation and distance contraction here, which will probably come into play to some extent, however . . . ), but neither A nor B is exceeding c relative to its local spacetime.

  240. Karl Love

    They say they measured the distance between the two sites using GPS. If we assume that their GPS-derived measurement was accurate, wouldn’t that still imply that their measurement between the sites was along the curvature of the Earth’s surface? Wouldn’t the neutrinos have passed between the two sites in a straight line, cutting through the Earth’s surface / crust? Surely that would imply that the measured velocity between the two sites would be faster than expected?

  241. Teokox

    In my opinion neutrinos cannot travel faster than light, but if their speed is so close to the speed light, their mass grow up (neutrinos have mass, very little but they have) and the space is shortened according to the relativity laws. So the effect is the distance is smaller at that speed and the neutrinos go faster than light. Could be this the answer?

  242. Dave Nixon

    I guess I am learning something about relativity that I thought I understood – only 40 years since I took that course – and the rust is showing. Thanks David Kahana for your detailed explanation. My bet is now on some subtle statistical error in the way they figure the arrival time of the leading edge of the neutrino pulse.

  243. David Kahana

    Cheers, Dave! Special relativity is pretty strange, indeed. If you had actually followed your own reasoning all the way to the end, working in whatever frame you chose, done the whole calculation there and then finally transformed the results back to an earth fixed frame, you would of course get the same result for the travel time as if you calculated in the earth fixed frame taking account of all the deviations from inertiality of the earth fixed frame.

    By the way, having read the experimental paper in some detail now I’m inclined to agree with you that a very likely source of error here is in the determination of the zero of time, which is coming from a measurement of the detailed time structure of the primary PROTON beam. The zero of time is when the leading edge of the proton pulse hits the target that produces the secondary pions and kaons. There is no direct experimental knowledge of when or where any individual neutrino is produced in the secondary pion+kaon beam, thus there is no direct measurement of the time of flight for any individual neutrino. All they really have is the arrival times of the neutrinos they detected at Gran Sasso, and the times when the particular pulses of protons that produced the pis and Ks that produced each detected neutrino were on target at the CERN SPS.

    The experiment would be a lot more convincing to me if there were a near and a far neutrino detector in the neutrino beam, like at MINOS and Super-K. In that case you wouldn’t need to worry much at all about the time structure of the primary beam, you could just keep that information as a consistency check.

    Not being a surveyor, I’m not really competent to comment on the determination of the distance, which seems like another possible source of error. They claim to know the baseline to within +-20cm. I really do wonder if such accuracies are possible over a distance of 730 km given that they have to propagate the distance from their GPS measurements, taken at the surface, down into the detector halls … I mean, I remember the story of how long it took and how difficult it was to get a good measurement of the elevation of the top of Mt. Everest. Of course, that was in the days before there was GPS.

  244. QED?Solid-state,expansion,light?

  245. A quick question over Phil’s statement: “If neutrinos travel faster than light, then we should’ve detected the neutrinos from Supernova 1987A before we saw the explosion itself.”

    1. Didn’t we receive the neutrinos some 3 hours before the light from the event was spotted? Since the neutrinos are generated by the progenitor and the light when the shock wave reaches the surface this is what was expected.

    2. Why do some websites and papers report that the first neutrinos reached Earth 18 hours prior to first light? Is this event documented? How can this event be explained?

    3. Given the distance to the LMC, site of SN1987a, shouldn’t the results from CERN dictate that the neutrinos arrival should have some 4 years prior to the first light?

  246. We do not see any reference neither in the original paper nor in any of the discussions to Heisenberg Uncertainty Principle! The uncertainty of Location [x] and Velocity [v] of the traversing (muon) neutrino can be determined. From the CERN – OPERA paper we learned that the studied particles energy was ~ 20 GeV. This can be applied into ∆X ∆ mv ≥ħ/2. Nobody has mentioned that we also have to part with the determination ambiguity. This needs serious and thorough analysis. Another pandoras box?

  247. Flash

    does this mean i am no longer the fastest in the world?

  248. Chris

    Regarding the comparison with supernovae observations, this argument usually assumes that the neutrinos and the visible light departed at the same time. However, isn’t it possible that neutrinos could have departed the supernovae at some arbitrary time before or, more importantly, after the visible light departed? Maybe a supernova generates a burst of neutrinos at some particular point in its timeline that is different to when the burst of visible light is created.

    In your article above, you suggested we should have seen the neutrinos 4 years before the visible light. But what if there was some process that caused a burst of neutrinos 2 years after the burst of visible light? In this case, the neutrinos might have caught up to the visible light, passed it and then arrived only 2 years before the visible light. What if the process that causes a burst of neutrinos occurs 5 years after the burst of visible light?

    My point is simply that this argument is only valid to the extent that you accept that neutrinos are created at the same time the burst of visible light is. /apologies if this was already suggested in the 291 replies above!

  249. Anonny

    There are GPS units so accurate they use them to measure continental plate movement, way better than “a meter or two”. If they used a military model and not ye olde garmin mapping unit, their locations should be close to spot on.

  250. Panchito666

    I see here a terrible mistake. “The travel time at the speed of light is about 2.43 milliseconds, and the neutrinos appear to have outraced that speed by 60 nanoseconds. If true, that means they were traveling just a scosh faster than light, by about 1 part in 40,000” WRONG. We have to assume, that the neutrinos surpass the light by only 60 ns, not in this particular “race” but in every race. So, to know the “factor” between how fast is the neutrino than the light, we just divide 1/0.000000060. If we apply the “error” the CERN is reporting of 10 ns, que just divide 1/50 ns and we have that the neutrinos are faster only in a part of 1 in 20 million. That matches up, with the result the CERN gave in their press release: “…and appears to indicate that the neutrinos travel at a velocity 20 parts per million above the speed of light…” SO, WE MAKE THE WHOLE OPERATION AGAIN: 160,000 (distance to supernova 1987) x 0.00000005= 0.008 YEARS! So the neutrinos should arrived only 2.88 days BEFORE the supernova…

  251. Matthew Domville

    I’m hoping that this translates into more fantastical science-fiction style technology in the near future. Here’s a comic that encapsulates my thoughts:

  252. SN1987A

    Here we go again. You have chosen an ARBITRARY number of 168.000 light years to my home, an ARBITRARY number for the speed of light and use scalar arithmetic to compute distance and time. The difference in arrival time of the neutrinos and the light violated your dogmatic model of the Universe, so you had to come up with an elaborated abstract mental model of the supernova process which you cannot observe. You can get away with this as there are no witnesses.

    Any violation of the theory you compensate for by RLATIVISTIC corrections. For what cannot be corrected you device a special case that always involves some kind of non-linear distortion of space. You MUST distort space because you assume that time is linear and the relationship is locked by the space-time continuum model. You try to twist the reality in favour of your “sanity” exactly as you wrinkle your trousers in favour of storage convenience. Reality however cannot be twisted because it is REAL, only your perception based on your believes can.

    Why would you apply the light speed limit to something that is NOT light? Just treat what you call neutrino as a phenomenon since you do not know what it really is. Make an observation of this phenomenon without translating it to particle/object/thing and try to expand your awareness beyond the three-dimensional space-time box. Sooner or later someone will come up with a new dogma that makes FTL phenomena legal for superluminal entities and imposes a new, higher limit on Reality – just another step in the evolution of human consciousness. Do not worry; TOR will still be governing the GPS operation in space-time and the ballistic computers in your tanks will still be using the Newtonian mechanics.

    GPS is touted as one of the most distinguished practical applications of TOR. If in this particular case both the timing and the position are derived from the same source and are locked by the space-time continuum relationship. Any change in one part shall be RELATIVISTICALLY compensated for by a change in the other. No error is introduced as long as it is not already present in the system itself.

    Anyway there is only one REAL Reality, only one Truth and it is ABSOLUTE. It can only be observed (and enjoyed), not perceived or interpreted. Any RELATIVE (as in relativity) concepts are NOT Truth and are of no consequence in Reality. They however limit your ability to enjoy the unlimitedness of the Universe and for instance observe the life of a supernova as it is (not what you think it might be) and marvel at its perfection.

    Always sending you Light and Love. SN1987A.

  253. Joseph G

    @289 SN1987a: Um… what?

  254. David Kahana

    Chris #293:

    There’s certainly no assumption made that the neutrinos depart at the same time as the light. That would be the wrong assumption to make in every proposed mechanism for a core collapse supernova I’ve ever seen. The theoretical assumption is that the neutrinos depart first, before the light.

    I suggest it’s well-nigh impossible to imagine any reasonable mechanism for the explosion that produces a neutrino pulse of the correct magnitude at *any* time after the light pulse begins to go out, never mind at a time *years* later than the light.

    At a time of several years, let’s say four years, the matter from the exploded star will have spread out to a large radius: some non-negligible fraction of a light year, say 1/5 of light year if the shocked stellar matter finally achieves 1/20 of the speed of light. Chinese and Arab astronomers saw and recorded a supernova explosion in 1054 CE in our own galaxy, and the remnant (the Crab Nebula) and the pulsar have both been detected. The current expansion velocity of the Crab Nebula is much smaller: more like 1/200 of the speed of light. But the point is that the matter is even now expanding very rapidly, so rapidly that the expansion was eventually seen in successive photographs of the nebula.

    The very dense core of SN1987A would still be there, four years after the collapse, as a neutron star, although there some, crazy in my opinion, theorists who say that it may have gone into a black hole after it produced a supernova, because no pulsar is observed at the position of SN1987A. Probably no pulsar is seen, because the pulsar is spinning and has magnetic poles that are oriented in a such way that its outgoing pulsed radiation, which is expected to be fairly directional,
    just doesn’t head towards us. But in four years, all of the neutrinos that were produced would have transported out of the core. Any model that exists has that happening on a time scale of a few seconds at most.

    Neutrinos interact only very seldom with matter at ordinary densities. But the visible light from a supernova explosion only begins to escape when the density of the leading edge of the outgoing matter in the explosion has dropped to a point at which it is basically transparent to light — when it is no longer an optically dense plasma but is instead at low enough density for the light to begin to free stream.

    As the material from an exploded star gradually spreads out, its density is inevitably dropping with time. There is just no plausible way for matter at low and dropping density to produce so very many neutrinos and anti-neutrinos as are contained in the pulse from SN1987A.

    The point is that the light is released at a stage when, for practical purposes, the main part of the explosion, the part that produced the neutrinos, is already over
    and done with.

    The time scale on which the neutrinos are released, is certainly no more than a few seconds, and the neutrinos must come from the very dense inner core of the star, not from the low density exterior. Remember: it’s known that the time spread for the arrival of the neutrinos at earth was no more than a few seconds. The neutrinos
    were certainly not produced over a time scale of many days, like the light … they
    were produced very quickly.

    So to do what you want to do, you’ld actually need to think of a mechanism that could trap all of the neutrinos inside the collapsed core for a few years after they were produced, and then release them -all at once. That is simply crazy – what could possibly do that? At a minimum, neutrinos will be escaping all the time from the surface of the collapsed core. There’s nothing can stop them from transporting out.

    Barring such a mechanism, all that’s really needed for the argument that bounds the difference in the velocity of the neutrinos and the velocity of light is the distance to the supernova and the time difference in the arrival of the light and neutrino pulses at the earth. Of course, we also need to make the assumption that light and neutrinos travel at constant velocities from SN1987A to earth.

    The distance to SN1987A is known by a pretty direct method, more or less by simple geometry, thanks to a very famous series of observations that were done using the faint object camera of the Hubble telescope, beginning in 1990, the first observation having been done 1278 days after we saw the supernova itself.. Due to these observations, there was a way of determining the physical size of a large ring of matter that surrounded the precursor star, which brightened and then dimmed in a very specific way with time, as the light pulse from the supernovae passed through it Due to the 1278 day time lapse, this matter had to have been pre-existing, since it was at too large a distance to have come from the supernova itself. As Hubble continued to observe SN1987A, the ring of matter gradually brightened up, the illumination proceeding gradually around the entire circumference of the ring from one side to another, at which point the entire ring was lit, and it reached maximum brightness. The ring of matter glowed for a period of about 400 days, the light fading away gradually. From that behaviour, we can estimate the ring’s actual size, and its inclination to us. And since it clearly surrounds SN1987A, we get a good measurement of the distance to the precursor star … unless you imagine that the ring lighting up was caused by something other than the light pulse from the supernova …

    I’ll give a brief account of the prompt mechanism for a core collapse supernova, here, in which the initial shock that forms in the core collapse is sufficiently strong that it never stalls on its way out, becoming an accretion shock. Certainly it’s true that not all, maybe even not most theorists believe in that mechanism. However I think it’s probably the most reasonable mechanism. It’s a bit of a joke, if you ask me, if as some people say the shock stops at first and then is re-invigorated a second or so later on by the outgoing neutrinos, or if convection or some 2D or 3D hydrodynamic instabilities are really needed to explode stars. In fact, I’ld say that since the equation of state of the matter in the collapsing core is not at all well known then the prompt mechanism is the most reasonable to consider. But none of the mechanisms will differ very wildly from all the other ones about the time scale on which neutrinos begin to come out. The (anti)-neutrino data from SN1987A is not really sufficient to distinguish the mechanisms. But the order of magnitude of the observed pulse confirms the idea that the explosion was caused by the collapse of the core … there is nothing else imaginable that can provide that much energy.

    Then I suggest that you just think about it for a while and see if you can come up with a different and detailed mechanism that does what you want to do with the relative timing of light and neutrinos, so that the results from SN1987A would agree with this one questionable experiment from Gran Sasso. Remember that you’ll need a pulse of anti-neutrinos and neutrinos of the right size to come out about 4.2 years *after* the light from the explosion comes out. Remember that neutrinos interact extremely weakly with matter at ordinary densities. For myself, I can’t imagine any such mechanism is possible.

    It’s actually the pulse of neutrinos and anti-neutrinos that is the prompt signal of a core collapse supernova, and it is an absolutely enormous neutrino pulse, containing an almost inconceivably large number of neutrinos. Consider that SN1987A, at a distance of 168,000 light years, was about as bright in anti-neutrinos at the earth, as own sun is, in neutrinos. In fact, the neutrino flux in a core collapse supernova is so big that, if you were near enough to a star which underwent such an explosion, you would die of acute radiation poisoning – from neutrinos! before you even got the chance to see the star explode.

    A core collapse supernova happens within a couple of days after the inner core of a high mass star begins to burn silicon into iron. The iron builds up in the inner core very quickly: silicon burning doesn’t last for more than a couple of days. At the beginning of silicon burning the core collapses further and also heats up. The core is supported against further rapid gravitational collapse for the moment, by its luminosity, which comes from fusing silicon to iron, and it’s also supported in part by the pressure of the free electron gas, which is a degenerate Fermi gas at the densities that exist, which are close to the density of a white dwarf star.

    At the end of silicon burning the inner core will be a spherical object composed mostly of iron nuclei, photons and electrons, having a radius on the order of 3,000 km and weighing on the order of one solar mass – actually it’s somewhat more than a solar mass – it will be closer to the Chandrasekhar mass let’s say 1.3-1.4 solar masses. It will also have low entropy.

    Now iron cannot fuse into any heavier element without a net input of energy, so the luminosity of the core, in photons, begins to drop when only iron is left. Further collapse then leads to more heating, and some of the photons in the core are so energetic at this point, due to the temperature, that they can actually break up iron right into 13 heliums plus two neutrons and then also break up helium itself into protons and neutrons. This process of photo-dissociation removes more energy from the core, destabilizing it further by reducing the pressure. Moreover, the free electrons in the core, which were providing some of the pressure which was holding it up, begin to be captured on the protons that are being released, and as they are captured they release electron neutrinos, removing still more pressure from the core. This point in time is the very beginning of neutronisation of the core and it’s also when the very first neutrinos from the supernova begin to come out. But the neutrino flux will not yet be at its highest point, by any means. This point in time is also the end of the star’s life, one way or another. It will now become either a neutron star or a black hole, because there simply isn’t enough pressure to hold the core up any longer. The inner core now falls in on itself, and very rapidly.

    The collapse occurs more or less in free fall, and the free fall time for the core can be calculated to be on the order of 10-20 milliseconds. It starts at a radius of about 3000 km and it falls freely all the way down to about 30 km, when densities of at least 2-4 times nuclear matter density are reached, and neutron degeneracy pressure finally begins to stabilize the core against further gravitational collapse.

    It’s after this rapid, essentially adiabatic collapse stage that neutrinos and anti-neutrinos start being produced very rapidly, mostly by e+ e- annihilation in the core, the e+ e- pairs having been created by the temperatures that have developed. Entropy now rises in the inner core. The densities reached are so high in fact, that at first the neutrinos are trapped, as the core falls in. But very soon, far less than a second after the collapse begins, neutrinos begin to transport out in earnest. In the models, there is typically a large increase in the neutrino flux within the first second after the collapse, and a tail of about four or five seconds at lower flux. And that’s about the limit, unless you admit the possibility of a transition to quark matter in the very deepest portions of the neutron star. If that happens, then the star will likely convert some of its up and down quarks into strange quarks, and there could be an additional burst of neutrinos associated with that. But the time scale for that conversion is only about ten seconds. The basic point here is that the collapsed core is just not very big anymore, and all of the neutrinos inside will transport 30 km within a few seconds.

    Now what about the supernova light?

    At some point in time during the collapse there appears a point within the collapsing core at which the speed of sound in the core matter is just equal to the speed at which the core is falling inwards (the collapse is homologous meaning that the velocity at which the matter falls in is a linear function of the radius from the center of the star – faster on the outside/slower on the inside). That point is called the sonic point, and a shock wave forms right there as the matter on the outside which is falling in at faster than the speed of sound reaches the sonic point. In a successful prompt explosion the shock wave eventually propagates outward and has enough initial energy to go all the way out, and blow off the outer envelope of the star, leaving only the collapsed core behind.

    But any shock will lose a lot of energy as it moves outward because it has to move through un-dissociated iron nuclei in the outer core and it breaks some of these up as it passes through. Each 0.1 solar mass of iron that the shock breaks up costs about 1 foe = 10^51 ergs of energy.

    Now the total energy made available by the collapse of the core is just the change in gravitational binding from when it was 3000 km in radius to when it is 30 km in radius. And that can be estimated to be about 10^53 ergs, or 100 foes. A healthy shock might have only a few foes, initially. A very healthy supernova explosion releases about 1 foe in kinetic energy of the outgoing matter and in supernova light. But there will generally be tens of foes released in neutrinos and anti-neutrinos in a core collapse.

    When the core collapses, the outer envelope of the star also falls inward, as well as the outer layers of the core. The smallest red supergiant star capable of burning all the way up to iron in it’s core has a mass on the order of 8 solar masses. So in such a star there would be roughly 7 solar masses of material still outside the collapsed iron core. The shock has to propagate through most of this material before any light can be released.

    Now the radius of the photosphere in such a star (the photosphere is the radius at which temperature and density is low enough for light to stream freely) would typically be at a distance from the center of the star which is on the order of the radius of Jupiter’s orbit around the sun, or about one light hour. So the light from the supernova doesn’t appear until the shock wave reaches about that distance from the core of the star. In a healthy supernova explosion, the velocity of the shock wave reaches about 1/20 of the speed of light. So this means the delay from the time of collapse, to the time at which light begins to escape would be about 20 hours.

    However, as we see, the neutrinos and anti-neutrinos have been streaming out from the core on a time-scale of only seconds, beginning in essentially the first milliseconds after the collapse, and unlike the shock, the neutrinos that get out are not impeded at all by the outer layers of the star.

    In the case of SN1987A, the precursor star, Sanduleak -69° 202a, was not a red giant, it was instead a blue supergiant, which may form when a red giant somehow loses its outer hydrogen envelope, leaving the inner layers of the star uncovered. So the photosphere was likely at a smaller radius than a light hour, and that could account for why the delay of the light was only three hours relative to the delay of the neutrinos.

    The number of (anti)-neutrinos observed from SN1987A suggest, when compared with models, that the total integrated energy released in neutrinos was 82 foe in the first 1.6 seconds. If those neutrinos have energies of about 10 MeV then that amounts to about 5.12e+57 neutrinos. If you can imagine a plausible way to delay the emission of all those neutrinos for 4 years — then my hat’s off to you!

  255. Nigel Depledge

    @ David Kahana (301) –
    Wow, those numbers are just mind-blowing!

    Thanks for the comprehensive* description of a core-collapse supernova.

    * To a layperson, anyhow.

  256. Nigel Depledge

    @ Joseph G (300) –

  257. Scott

    We ARE on a planet MOVING through space. In other words, the end points were in motion during this experiment. That could have some affect on this.
    And what’s to say you can’t move faster than light anyways? [only theories]

  258. SN1987A

    Mind blowing the Universe is. Numbers are brain blowing. Brain and Mind are not the same. Brain is just a biological computer that executes someone’s models you willingly load into it. The Universe does not have spatial dimensions – they are a product of these models. Your brain as any computer is object oriented and needs to keep objects separated by either space or time hence the space-time continuum.

    Objects are constructed by your brain by applying the model’s rules to your sensory input. This input is real, a direct experience of the Universe but no brainware is capable to apprehend the infinity of energy change cycles happening continuously. Yes, the cycles are continuous and have neither beginning nor end. There is no past, no future, only eternal NOW. Time does not exist but the continuous change does.

    Your brain samples the sensors (observation) and updates the object’s properties according to the models (perception). A sample is the sum of the infinite number of cycles and two consecutive samples are different, not the cycles themselves which are continuous. One of the properties in your models is position. If a model disallows objects to share space (sanity), so any position conflicts must be resolved by some other form of separation. You introduce the concept of time to make the model “sane”. It is quite easy to pick the most obvious and easily identifiable cycles like planetary ones, scale the fictional time and assume it to be linear.

    As any computer your brain uses scalar processing for everything. Any discrepancies in models due to non-scalar input are either ignored (scientific ignorance or (lack of education == inferior model)) or corrected by a special model extension (prudent science) . To some degree it can be corrected by a generalized theory but when it fails an even more specialized extension is introduced to make ends meet (how many quark flavors do you need to explain matter?). This is a domain of hardcore scientists who always chase discrepancies that will always have their own (quite fractal isn’t it?). In any case the space linearity is sacrificed in favor of maintaining time linearity.

    Your models no matter how brain blowing they are by no means govern the Universe – Mind does. But the models can be upgraded to accommodate higher dimensions not by extension though as they make space meaningless and retire time back to oblivion. You can use one model for everyday life of material values, particles and linear clocks. You can switch to other model in which photon is neither wave nor particle but a pure Idea that vibrates the whole Universe simultaneously. Somewhere in this Universe this vibration is experienced by you sensors and converted to particle/wave mutant by your model in use. Same is applicable to all other matter substructures. Niels Bohr shared with you that there is no “Deep reality”. Von Neumann looked hard and found no reason for “ordinary objects” to exist and John Bell proved that all communications in the Universe are instant. You already have foundations for new models. Now you are given an opportunity to experience and understand the experience beyond current models.

    The brainware is available to download, all you need to do is ask for one. Ask and thou shall receive – one of the Universal Laws. Until you do you will keep arguing about matters of no consequence.

    Just to give you some help in this world of constant conquest of brains, this bizarre bazaar of brainware here is a tip. Truth is always simple. The more complex explanation/model the farther it is from the Truth.

    A great master sends you this message: “Nothing has attributes, qualities, properties or values apart from ones assigned by you.

    And another one to cheer you up. “Mind over matter: if you don’t mind it does not matter.”

    Love, Light. SN1987A

  259. Mr. Immortal

    Actually, gravity does push instead of pull. Gravity is the pushing force or pressure of the fabric of space exerted against all matter and most, if not all, particles.

  260. MaDeR

    Wow, obvious crackpot is obvious. He even thinks he is physical star. Off meds much?

  261. Mr. Simple

    Another thing. If e=mc^2 then objects can go faster than the speed of light. Because the mass times the speed of light times itself again is when matter becomes energy. Or should we just change the equation to e=mc?

  262. skiffle

    I thought E=MC Hammer

  263. Hrvoje Crvelin

    I noticed following paper appeared on arXiv:

    Inside they claim neutrino wave did come 3h before light (though they clearly indicate most people do not consider this as fact for anything further).

  264. Ravison

    Please correct me if I’m wrong, but assuming a neutrino interacts with very little are there measurements to ensure the thing passed through the entire circuit. Couldn’t it just have taken a shortcut? Was it measured at multiple locations and does not the

  265. Mike

    The CERN scientists are very well qualified. Surely they accounted for Cherenkov radiation, GPS error, etc. That does NOT mean that the results are evidence of faster than light travel, but it does mean that these folks accounted for at least the most obvious errors.

    As far as evidence of neutrinos from 1987A….those neutrinos (If..and that’s a really big if neutrinos travel slightly faster than light) would have gotten to earth a few years before detection (say 1983).

    a)Neutrino detection was very new in 1983 and if we did detect some in those existing SQUIDS we may not have correlated them with an event from 1987.

    b) The research does not claim that all neutrinos travel FTL, only the ones detected in this experiment..and then the research does not even claim that (just that they appeared to travel FTL).

    Anyway…this is all interesting stuff.

  266. Hrvoje Crvelin

    And here comes another update which claims to have explanation which matches what has been observed. Relativity is a bitch.

  267. There is a saying in Science which explains that a thousand experiments cannot prove a theory right, but one experiment can prove a theory wrong.
    Angelo Molinaro Tel 570 489 3059 e-mail

  268. Rick Cordasco

    Gentlemen, please!!! When I herd this stuff I just laughed. This is not first order macroscopic homogeneous differential equations, high school stuff, v=dx/dt, This is laced with Neutrino stuff that goes back into the Hans Bethe postulate, the Sun is Fusion , and Fermi’s the missing energy, the baby neutron. There is and has been for the past 40 years experiments going on about neutrinos (see CERN, FERILAB, SLAC, USCLA, BROCKHAVEN etc), the numbers don’t match the theory, so the process has been to find a new theory… yeach, not to mention science hasn’t been the same since Feynman farted!!! QED changed everything. So whats going on!, In case you were asleep for the last 20 years, the Super Collider was cancelled, Big Science ended!!! Sorry. The problem is much, much deeper than Neutrinos traveling faster than the speed of light. Einstein was proclaimed to be out of touch, the last 15 years of his life at the Advanced Institute. The fact of the matter is that he was in touch exclusively with a guy named Kurt Godel. Kurt basically showed that mathematics in general (Incompleteness Theorm,, Good Luck!!) l, and specifically was incomplete, overturned all of the verifiable exactness of scientific method and Greek logic in general, crazy SOB, and of course was rejected by the entire world community. Besides the appartent,,, instrument error etc. its even beyond Brian Green or Steve Hawkins. As Nietzsche said man is a bridge, and there very well may not be another side! So I’m afraid my cynicism has overtaken me,,, its about money and funding.

  269. Fascinated

    Firstly, they announced that they achieved a result and they needed it to be checked, so there was no real claim. Secondly, there is an army of the best physicists in the world with the best gear possible, I doubt that they are so silly not to realise the bloody obvious when doing their calcs.
    Thirdly, Someone will always come up with some bodgy solution to make the data fit the theory. i.e. Oh it must have twisted spacetime with no evidence to support the claim. People are happy with claims like these as it keeps the status quo and billions in grants are kept.
    Lastly, To many ill informed opinions are being spread all over the internet, just like this one. If you read this article carefully, it is nothing more than a bunch of IFS and alternative reasons to keep the data fitting the theory. So we all need to shut up and wait for Fermilab and others to update their gear and do the experiment again. Only with more results can we move on to the real truth.
    In the words of Kip Thorne ” Lots of things in the universe travel faster than the speed of light; The speed of light is a LOCAL statement” ( search for Dr Karl, Kip Thorn and Sir Roger Penrose in the same search )
    There is only the mindset of limitations that govern our finite and controlled universe )

  270. Sen. Abba Aji

    I would rather give benefit of doubt to the chance that nuetrinos may have actually been found to be faster than photons. It is after all, in expectation of understanding how energy acquired mass to form matter (Higgs bosons), and possible discovery of unknown behavior of other elementary particles at the initial state of the universe, that such unprecedented resources were pumped in the CERN project. So let keep our minds open to surprises. Acceptance of passed Surprises were indeed what got us to our current stage of knowledge

  271. Sen. Abba Aji

    I would rather give benefit of doubt to the chance that nuetrinos may have actually been found to be faster than photons. It is after all, in expectation of understanding how energy acquired mass to form matter (Higgs bosons), and possible discovery of unknown behavior of other elementary particles at the initial state of the universe, that such unprecedented resources were pumped in the CERN project. So let keep our minds open to surprises. Acceptance of passed Surprises were indeed what got us to our current stage of knowledge

    Sen. Abba Aji

  272. The errors I believe are in the programming of the GPS timers. I believe they overestimated the distance by about 10 meters using the GPS system, and underestimated the coordinated timing by about 30 nano-seconds, also based upon the GPS timing system.

  273. kentee

    I do think that it can travel faster than light because it has happened and that time difference is actually quite large if compared to that speeds

  274. Ok, mistery resolved! :)

    Somebody in the first comments asked the question whether the earth’s movement around the sun was incorporated into the distance measurement:
    While the geographic distance between CERN and Gran Sasso is precisely measured by GPS to within cm, the earth is still moving around the sun while the Neutrinos are on their way.
    It is even possible the earth is moving towards the Neutrinos during this time (I spared myself from the trigonometrics of the relative movement, and only worked out the max movement, so this needs to be adjusted).
    As always, I am never sure about the correct fram of refernce, etc, but I surmise this effect might explain all of the whoopla, so smarter brains than mine please check.

    I did this calculation: How far does the earth move during the Neutrino travel time of 2.43 ms between CERN and Gran Sasso?
    The sun is 150 million km away, the earth takes 365 days around, meaning the earth moves at 30’000 m/s. During 2.43 ms, the earth travels 72 m.

    That is more than enough to explain the “gain” in speed of the Neutrinos against the photons!

    (It helped to remember that the photons are not running in parallel with the Neutrinos in this experiment, as the Neutrinos actually are able to pass through mass – being neutral – while the photons cannot, so this was not a side-by-side speed competition).

    That would explain it all.

    I wonder whether I have come across the correct explanation here – myself and my buddy from the earlier comment would certainly accept some sort of medal and free travel to Gran Sasso for the next Neutrino arrival party.

  275. “Neutrino”


    “Knock, Knock”

  276. Randy

    The speed limit of the universe is set at the speed of light. Now you check, and find neutrinos are traveling faster by 20%. Same experiment plays out every day on every highway.

  277. Jennifer

    A new set of scientists conducted same tests and experiments and found the same result! There are particles that travel faster than the speed of light. Heres the link!

  278. Fastrogeek

    Perhaps the accepted value for the velocity of light is wrong? c is speed of light in vacuo, but my guess is all measurements have been made in an environment that contains Dark Matter (try sucking that out of your vacuum tubes). So, perhaps photons are slowed down by Dark Matter more than are neutrinos. In which case perhaps the neutrino speed as witnessed by the Opera experiment is more indicative of the true value of c, rather than the one we’ve been measuring using photons?

  279. Wim

    1. the accepted light velicity is not wrong, it would make E=mc²
    into E=(m*’how much we were off’)²

    2. I understand they used gps for distance measurements, I believe their measurement will be accurate to the cm, otherwise they would not come out with these results.

    So theres still 2 problems I can think of,
    measuring time, which is tricky, if you where to set 2 atom clocks to the exact same time, and move one to gran sasso, then that clock will be off due to time dilation, this is certainly not what happened, it just explains how difficult a correct time measement is.
    The other thing might be a quality about neutrinos we don’t yet know. I could think of 1 that explain the whole thing. Maybe the reason why neutrino’s don’t interact with matter is because they phase through, being destroyed at one end of it and recreated at the other, highly speculative, but entirely possible.

    Whatever happens, Einstein will still be right, about everything he observerd and explained, neutrino’s being created or traveling in the circumstances they did, may just not be part of it.

  280. Wim

    It would be awesome if they didn’t travel faster than light, but still arrived early, my phasing example could explain this, OR CERN somehow made a space-like curving in spacetime. Meaning manipulate the space between LHC and grand sasso in such a way that the neutrinos didnt have to cross the whole length to get to the other side.

    as I understand LHC is only 32km (the collider itself) so most of the travelled distance is through the earths crust, which makes it impossible to do thesame thing with visible light, that way you could at least measure if light is travelling lightspeed, cause it its not, then your experiments are wasting a lot of money for nothing

  281. Thats Guy

    Once we realize that Gravity is actually atmospheric pressure coupled with electro magentism, Warp Drive and Interstellar travel will be like riding a bike!

  282. Question

    The reasons they have for the neutrinos not being faster than speed of light. Wouldn’t those same reasons apply to how fast light travels as well?

  283. zero

    is the speed of earth accounted for?
    do we actually know the “actual speed” of earth?
    the actual speed of earth depends on the following
    ———–the speed of the motion of solar system,if it is moving
    ———–the speed of milky way galaxy
    ———–we might me enclosed in a smaller universe of the many universes existing,making a larger we know its speed?
    and so on…
    we might never know the actual speed.

    what about the law,that ‘relativistic mass’ increases with velocity?
    does ‘relativistic mass’ increases ‘relativistic gravity’ ?
    does the rate of increase of gravity has any effect on its time?

  284. Ieuan Styne

    The Neutrinos most likely took a short-cut route (perhaps a sign post was moved ) between the two points. They’ll figure this out once they’ve slepton it a while.

  285. Brian Calla

    If dark matter is expanded via gravity in a black hole and light can’t leave it and or time or light is bended why is it hard to believe that something can control and “go faster” then the speed of light. These ideas that exist and laws must always be challanged with all the options in from of mankind I think to often we ignore all other choices. Dark matter can not be seen or reflected. The invesitgation into gravity and the FORCE that ex

  286. it appers to me that most of you are barking up the wrong tree ok you have just seen the first faster than light sub
    particle and have got just a little exited have any of you so called brains considred multiple 11 dimentional frameworks ie 11 times 11 framework or 11 by 11 matrix the numbers should not be that hard to crunch
    and will show many farster than light particles now consider a 0 * 0 matrix with .0000000000013578 degree offset
    at below 0 point or 0.000000856-42squared

    there is only potentiality dissolving into reality dissolving into potentiality

  287. henry

    just a theory

    if you have a long car , with a road on top of it
    and this car is travelling at 100km/hr
    then you have another car on the road on top of this car travelling at 100km/hr
    with reference to ground… would not the second car be travelling at 200km/hr ?

    e.g: walking on a travalator… your speed with references to stationary objects outside of the belt would be your walking speed + the speed of the belt…

    hence if LIGHT was a carrier, and on this carrier something is travelling inside of it,
    that thing would definitely be faster then light
    light traveling on top of light, would mean the inner light is travvling twice the speed of light

    if this partical of nutrinos in riding the light wave… then the nutrino would defnitely be faster than light….

    now all we need to do, is figure out how to ride a light and walk on the light …
    and its probably not that hard too

    light is a photon a photon is a particle
    if this particle is already travelling at light speed,
    on this constant stream of light, we push the light forward using another light… then the first light would have travelled twice the speed,,

    light particles are the same, hence they should push each other around, if a torch light is already shining to a distance, and we put another torch light, (switch on another bulb), the lit area becomes brighter, bcos more light are shone on it, but did we discount the fact that light on light could have been the cause of brighter light ?

    excited photons are brighter.. then normal photons so to speak,
    heat cause rapid movement in particles/molecules…
    more heat = brighter = faster

    you want faster than light travel ….
    simple… get someone to shine a light at a distance, (laser or whatever)
    and you, run towards the target with a laser..
    your light theorectically travals at the speed of light + your running speed, hences your light is faster then the other person’s light and therefore faster then the speed of light

  288. Javier

    Quiero recordar los datos.

    Segun Caren Hagner (líder de la sección alemana del proyecto OPERA):

    El ultimo metro de fibra óptica, según se subía o bajaba, metía hasta 100 nanosegundos de retraso.

    Solo espero que en este blog se pueda calcular.

    Soy Ingeniero Superior en Telecomunicaciones por la UPM, llevo mas de 22 años en Laboratorio en Multinacionales, y conozco perfectamente las capas medias y de alto nivel de los protocolos de comunicaciones.

    Es IMPOSIBLE que en 3 años no lo hubieran detectado.

    Mas aún, conozco los dispositivos fotónicos de emisión/recepción, se que pueden meter del orden de picosegundos de retardo, JAMAS 60 nanosegundos.

    Pero lo peor, a nivel estrictamente físico, los retardos introducidos por deformación de un conector (Estructura pasiva de unos 2 centímetros), de un metro de fibra óptica (Lo que dice la Sra, Caren), pueden introducir entre 50 y 80 picosegundos.

    JAMAS, 60.000 picosegundos (60 nanosegundos).

    Solo puedo decir que es mentira, y me gustaría que en este blog, se permitiera el calculo libre de estos datos.

    En otros blogs (Como ) llevan 20 días sin calcular nada .

    Se les pidió en :

    No solo no han contestado, sino que han llenado de insultos tipo ‘magufo’, y finalmente, han bloqueado el acceso y borrado todo planteamiento de calculo.


    Un conector de fibra óptica , como mucho, es de unos 2 centímetros de largo.

    Si la fibra óptica se aleja 2 centímetros , como mucho, y aun así sigue el enlace digital (Lo que es mucho suponer), el incremento de retardo por esos 2 centímetros, con un indice de refracción de 1.2 del aire a presión atmosférica a ras de suelo, es de:

    Distancia / (Velocidad de la luz/ Indice de refracción) = 0.02/(3e+8/1.2) = 80 picosegundos

    80 picosegundos , en esas extremas circunstancias, es el máximo tiempo que un cable óptico mal conectado puede añadir al camino óptico.

    80 picosegundos son 0.08 nanosegundo.

    JAMAS un conector de fibra óptica mal conectado, añade 60 nanosegundos.


    Están mintiendo, y lo saben.

  289. Artimus_Foo

    It is an idea to maintain a certain amount of skepticism when we talk about braking einstiens laws, but denouncing something just because it brakes a law that has been said to have been kosha for all this time can hold us back. We do not fully understand how the universe works and anything could be true. To say that a mistake was made with the distances and timing is a moot point. Something that simple would have been picked up in no time and with a fair few professional physassists and mathmaticians pouring over the data and equipment, I feel it will not be long until some laws will have to be revised and the advancement of the human race can undergo, I just hope I get to see something in my life time.

  290. Artimus_Foo

    The Neutrinos from 1987A arrived 3 hours before the photons, true, this is because time had to be allowed for the particles to work there way out from the central mass to “free space”. This particular case there were only 24 neutrino recordings, but what fascinates me is that 3 hours before again at Mont Blanc, they detected a 5 nu burst, if from 1987A it could be proof of FTL, or just proof that something nearby was throwing out nu. The problem is that you can not really judge from this as the neutrinos and photons dispersed at diffrent times, if it were at the same time, would we find that the neutrinos arrive 5 minuets sooner? 30 minuets sooner? or at the same time?

  291. I would like to pose a question to the many who seem to visit and comment here. It isn’t directly associated with the topic but I would like everyones opinion anyways.

    If space is a vacuum free from drag by oxygen and one was to follow a trajectory clear of gravitational forces, why is it not possible to reach light speed or greater? Even if the buildup was slow and determined what would prevent a craft or object from reaching these speeds thru space? With a constant or repeated forward push nothing would or could act to interfere with the object in motion.

    Please be kind with your comments as I have a very fragile ego. )

  292. fluxrider

    Wally-Bob (209) said:

    ” Changing the relative reference frame to A, I am departing at 0.6c and B is departing at 1.2c.
    Obviously my next question is: Why not?”

    I’m not a physicist, but I can tell you what Relativity’s response is. When calculating with velocities that approach the speed of light, the calculations need to account for dilation effects.
    The proper formula to “add” two relativistic velocities together (i.e. to determine the velocity of one with respect to the other) is not v1 – v2, but rather :
    (v1 – v2)/(1 – (v1*v2)/(c*c))
    (I apologize for the parenthesis, I do not have good math symbols to use here.)

    So to use your example, v1= 0.6c, and v2= -0.6c
    The speed with which B is departing from A is:
    (.6c – (-.6c))/(1 – (-.36*c*c)/(c*c)), or 1.2c/1.36, or ~0.88c (not 1.2c).

    Also just to note, at speeds that are much less than c, this formula yields essentially the same answer as “v1 – v2”.

  293. Can someone please tell me What happens to some of the neutrino as they travel to Earth?

  294. This is an old article, and I read that the fault lay in a loose connector. But I would think a loose connector would cause a slowing of the signal, not a speeding up, unless the signal going through that connector was something else, and not the actual neutrino detection signal.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar