Faster-Than-Light Neutrinos and the Dynamics of the Internet

In the last six weeks the physic world may have witnessed the overturn of Einstein’s theory of special relativity or we may have only witnessed an interesting example of the effect of Internet dynamics and sociology on scientific discourse. As of this posting we do not know which alternative is correct.

As has been publicized everywhere, in newspapers, in blogs and via scientific electronic publishing, on September 23, 2011 researchers from the OPERA experiment announced the astonishing possibility that muon neutrinos may acquire velocities larger than c, the velocity of light. [arXiv:1109.4897v1]

The layout of the OPERA experiment.

The production of the muon neutrino beam at CERN [arXiv:1109.4897v1]

The OPERA detector in the underground Gran Sasso Laboratory.

Their abstract summarizes the experiment and result simply.“The OPERA neutrino experiment at the underground Gran Sasso Laboratory has measured the velocity of neutrinos from the CERN CNGS beam over a baseline of about 730 km with much higher accuracy than previous studies conducted with accelerator neutrinos. The measurement is based on high statistics data taken by OPERA in the years 2009, 2010 and 2011. 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. An early arrival time of CNGS muon neutrinos with respect to the one computed assuming the speed of light in vacuum of (60.7 ± 6.9 (stat.) ± 7.4 (sys.)) ns was measured. This anomaly corresponds to a relative difference of the muon neutrino velocity with respect to the speed of light (v-c)/c = (2.48 ± 0.28 (stat.) ±0.30 (sys.)) ×10-5.”

Incidentally, I have a soft spot in my heart for this experiment because its first purpose was to detect the oscillation of muon neutrinos into tau neutrinos. This has been accomplished. [arXiv:1107.2594]

The OPERA paper also summarizes previous measurements of neutrino velocity. “With a baseline analogous to that of OPERA but at lower neutrino energies (Eν peaking at ~3 GeV with a tail extending above 100 GeV), the MINOS experiment reported a measurement of (v-c)/c = 5.1 ± 2.9×10-5 [MINOS Collaboration, P. Adamson at al., Phys. Rev. D 76 (2007) 072005.]. At much lower energy, in the 10 MeV range, a stringent limit of |v-c|/c < 2×10-9 was set by the observation of (anti) neutrinos emitted by the SN1987A supernova [K. Hirata et al., Phys. Rev. Lett. 58 (1987) 1490; R. M. Bionta et al., Phys. Rev. Lett. 58 (1987) 1494; M. J. Longo, Phys. Rev. D 36 (1987) 3276].

Since then there have been a great number of papers and blog comments on the result. Some recent papers take this result as new physics and connect it with known or speculative phenomena such as : violation of Lorentz invariance, dark energy, dark sector neutrinos and so forth. Other papers present arguments against the possibility of the measurement being correct, for example “New Constraints on Neutrino Velocities” by Cohen and Glashow, arXiv:1109.6562v1.

Some blog comments present analysis showing where the OPERA experimenters may have made a mistake. But the result is statistically strong and if the conclusion is wrong it must be due to an error or errors in the experimental method: perhaps in the timing measurement method or in determining the 730 km baseline. My experience is that it is often difficult for outside researchers to find errors in a well designed experiment exhaustively studied by the inside experimenters. One example of a possible error pointed out by an outsider is a special relativity correction to the GPS timing, discussed by van Elburg [arXiv:1110.2685v2]. Another problem in finding errors in the research results of others is that sometimes the suspicious result comes from a number of errors.

My piecemeal and anecdotal survey of opinion in the physics community is that the consensus of interested physicists is that the OPERA result is mistaken, there is no consensus as to where the mistake or mistakes lie. But as my Ph.D. advisor, Isadore Rabi, said “Physics is an experimental science”. The great number of papers on the internet do not replace more experiments.

However as of this posting, October 29, 2011, we have only one additional experimental result, that from the ICARUS experiment. [arXiv:1110.3763]. The ICARUS experiment is a large liquid argon Time Projection Chamber in the Gran Sasso Laboratories that sits in the same muon neutrino beam as OPERA. The ICARUS experiment measures the energy of neutrinos that interact in the liquid argon and also records a ‘picture’ of the interaction. It does not measure the travel time of the neutrino from CERN. If one applies the theoretical deductions of Cohen and Glashow to the ICARUS results there are no faster-than-light neutrinos in the beam. A strong but indirect argument against the OPERA result.

We would all like to see a speedy settlement of the question- is the OPERA result correct? How rapidly can the question be settled? What about the OPERA experimenters repeating their experiment. Since the statistical situation is so strong, a repeat would have to involve physical changes in the apparatus. For example a large change in the mean neutrino energy or a different neutrino detector timing system or a major change in the neutrino beam. It will take time for the OPERA experimenters to make such changes and study the
results,  even if the actual new running time is short. Probably a month or more. We should be patient.

An obvious settlement path is repeating the experiment using the other long baseline neutrino apparatus, MINOS [The MINOS Experiment and NuMI Beamline] with a 734 km baseline and T2K with a 300 km baseline. The MINOS experimenters at Fermilab have discussed two directions for doing this. One direction is to go through old data trying to improve the timing information . This is worth doing but I am skeptical as to how much clarity this will provide. The other MINOS direction is to upgrade their timing system and take new data. But this will take of the order of a years or more. The experiment must be carried out, the data analyzed and the MINOS experimenters be sure they are right.

The T2K experiment [arXiv:1106.1238v2] in Japan can also look at old data and upgrade their timing and detection systems for their next runs. The latter direction will also take a year or more.

Thus we are in the midst of a dichotomy between (a) the short times- hours to days- required for us to think and calculate and then to report on the Internet and (b) the long times-years- required to carry out an experiment, analyze it, and be sure the result is right. The Internet gives us the illusion of speed in physics research.

It would be marvelous if the OPERA physicists in the end are correct. Their result would produce a revolution in our understanding of the relativistic world, equal to the revolution produced by quantum mechanics in the classical world. If the result is wrong then the superluminal muon neutrino concept will fade away. The fading process will be similar to what happened to the gravitational fifth force experiments and theories of the late 1980’s [E. Fischbach and C. Talmadge, Nature 356, 207 (1992)], the excitement, the puzzles, the mistakes to be remembered by old-timers such as myself. An odd sidelight is that a recent paper by Dvali and Vikman [arXiv:1109.5685v1] discusses superluminal neutrinos in connection with a gravitational fifth force.

November 20, 2011 Update

The OPERA experimenters  have submitted their paper “Measurement of the neutrino velocity with the OPERA detector in the CNGS beam” to the on-line, referred Journal of High Energy Physics (JHEP). The conclusion remains about the  same,  relative difference of the muon neutrino velocity with respect to the speed of light (v-c)/c = (2.37 ± 0.32 (stat.) +0.34 (sys.) -0.24(sys) ×10-5. There is an important addition: the results of of a small data, low proton beam  intensity run show that the beam time structure is understood. This new paper is   arXiv:1109.4897v2.

This new paper and its submission to a referred journal appears to have changed few minds, most concerned physicists remain skeptical. But I am puzzled, the very experienced and competent OPERA experimenters have had time to consider the dozens of suggestions, mostly on blogs, as to why the results are wrong; suggestions ranging from GPS problems to timing problems to relativity problems.  There are  numerous comments that the next step is to wait for results from other experiments to be done, particularly from MINOS.

But I remain puzzled and patient and fascinated. I will continue to  follow this fascinating physics on this post.

This entry was posted in Uncategorized. Bookmark the permalink.

20 Responses to Faster-Than-Light Neutrinos and the Dynamics of the Internet

  1. I merely like the handy information an individual supply with your reports. I’m going to book mark your blog site and also test out yet again in this article consistently. My business is relatively confident I will be well informed many innovative products proper here! All the best for the following!

  2. Pingback: Creativity in Science and Engineering | Reflections on Physics

  3. Paul J. Stiles says:

    Dr. Perl,

    It is always interesting to hear (or read) your thoughts on a subject. I have been following the Opera experiment and those speedy faster than light neutrinos. As someone trained in engineering and having somewhat more than a lay person’s understanding of physics, I think the probability of some sort of experimental error is more likely than some outlaw neutrinos breaking the speed law.

    As a science fiction fan and supporter of our space program, I hope the Opera results hold up. I would love to see new theoretical physics that allows for the possibility of faster than light space travel.

    Cold fusion powered (with poly-water coolant) boosted with 5th force power taking us to the stars would be great. Seriously, even if the Opera experiment results in regards to faster than light neutrinos do not hold up, at least whatever flaw(s) are found will lead to sounder experiments.

    Thank you for creating such an interesting BLOG, Dr. Perl. I’ll have to regularly stop by and read what’s new.


    Paul J. Stiles

  4. Jeffrey Matos, M.D. says:

    I’m a cardiac electrophysiolgist (MD specializing in heart rhythm abnormalities), with a background in chemistry and physics. I’d appreciate knowing your assessment of my view.
    My thought, which I illustrate below, is that the true neutrino speed in the experiment under consideration may be equal to or less than c if:
    (i) we live in a world of four or more spatial dimensions; and
    (ii) the three “familiar” spatial dimensions are arrayed as a curved space with respect to at least one other spatial dimension.
    The result of (i) and (ii) would be that a measurement of the distance between two points measured in the familiar three dimensional space would be greater than the distance between the same two points measured in the four-or-more dimensional space within which the 3 dimensional space lies.
    For example, consider a 1-dimensional subspace situated on a circle of radius R lying within a 2-dimensional space. The distance measurement between two points on the circle, a and b, separated by 90 degrees:
    – in the 1-dimensional subspace, would be πR/2
    – in the 2-dimensional space, would be R√2.
    Thus the apparent distance measurement in the 1-dimensional subspace would overstate the true distance between the points.
    Clearly, an overestimate of the a to b distance would result in an overestimate of the speed of a particle moving from a to b. And clearly, this argument could be conceptually scaled up to show that:
    – a three dimensional distance measurement can overestimate the true distance in a world with four or more spatial dimensions… and thus,
    – that a speed calculation based on a distance determined in a 3 dimensional subworld would overestimate the true speed of a particle, i.e. the true speed being the speed based on a distance determined in a 4-or-more spatial dimension actual world.
    Assuming no supraluminal neutrino speed, the extent to which the measured neutrino speed seems to exceed c, could be used to obtain information about the geometric relationship between the everyday three dimensions and the putative additional spatial dimension(s).

    • David Brown says:

      I might summarize Dr. Matos’s idea by saying “extra spatial dimensions might cause a slight, unexpected increase in the gravitational redshift predicted by Einstein’s general relativity theory.” If the preceding idea is wrong, then I suggest that M-theory is wrong and Wolfram’s “A New Kind of Science” is also wrong.

  5. Dan Iezzi says:

    There has always been a move on to disprove certain aspects of Dr. Albert Einstein’s Theories of Relativity. For example, Dr. Einstein suggests that the singularity of infinite density is not a “physical reality.” Dr. Einstein calls the cosmological constant the greatest blunder of his career, but that only holds true if the expansion begins from a point. Is this another attempt to show that slaves to the theory of relativity are greater than it’s master?

  6. Mukund says:

    What if the OPERA experiment is actually correct? Changes the way we see the world doesn’t it ?(obvious to say but not so to accept). Everyone keeps saying/blogging/etc that the result is flawed, but when I look through the past, the way the human understanding of Physics evolved/was evolving, ‘radical’ ideas were laughed at and eventually they came to the fore as time went by. Is this one of such instances?

    Secondly if it is possible to accelerate particles faster than the speed of light, how can we use it? Yes Relativistic Laws theoretically prevent that as you’d be knowing very well but I find that nature always has ways and means to surprise us. The reason I’m saying so is that, the relativistic physics models actually puts a limit on everything which I don’t favor. Finally, at an immediate time instant after the ‘Big Bang’, the levels of energy were so high that there might have been a situation wherein the particles were accelerated to speeds faster than that of light. If CERN with its limited energy can actually drive a particle to faster than light, shouldn’t the energy of the whole universe be able to far better?

    Thanks if you do actually find some sense in my questions and also for the Tau Lepton.


  7. One example of a possible error pointed out by an outsider is a special relativity correction to the GPS timing, discussed by van Elburg [arXiv:1110.2685v2].

    That paper was obvious nonsense from the get-go, and the author has, subsequestly admitted as much. Lest you mislead the public, it should not be mentioned in the same breath as serious physics papers on the subject.

    That said, the most likely explanation of the OPERA result, which has been much-discussed, both in the blogosphere and elsewhere, is that they have somehow mis-modelled their bunch shape. A 1% error (1% of 1 μS is 100 ns) would easily account for 60 ns discrepancy in average arrival time, observed by OPERA.

    Further experiments will, indeed, be required. But, when the inevitable correction finally comes, I doubt that anyone will blog about it.

  8. Jorge Laris says:

    So all we have to do is wait to see the results. Thanks for teaching me something new.

  9. David Brown says:

    Consider 3 possible explanations for the OPERA neutrino anomaly:
    (1) experimental error;
    (2) new physics extending quantum field theory;
    (3) new physics extending general relativity theory.
    Is there a strong challenge to general relativity theory? The MOND pages (McGaugh) Pavel Kroupa: Dark Matter, Cosmology and Progress
    Is MOdified Newtonian dynamics (MOND) a plausible way to explain the OPERA neutrino anomaly?

  10. ObsessiveMathsFreak says:

    Thus we are in the midst of a dichotomy between (a) the short times- hours to days- required for us to think and calculate and then to report on the Internet and (b) the long times-years- required to carry out an experiment, analyze it, and be sure the result is right. The Internet gives us the illusion of speed in physics research.

    This is probably the Internet’s biggest problem: There is no quality control. There isn’t even the implication of quality control. Material is posted without any editorial or quality checks, and is then distributed across the globe in seconds in the same way as the most solidly verified sources. This is also one of the nets biggest strengths.

    Another related problem with the internet is that, in general, it is a hysterical place. Hysteria is the name of the game when it comes to news, fads, and trends on the world wide web. Very few really care if you just performed an experiment which found the anti-top quark, but perform an experiment in which reveals a 10^-4 variation in the speed of lights and kaboom: Light Speed Barrier Is Broken, Einstien was wrong, Warp Drive Around The Corner, Illuminati hold over science broken, Ron Paul 2012, etc, etc. This doesn’t just happen for physics either–far from it. It is as though the billion monkeys behind the billion keyboards of the net are simultaneously wired to a high octane sugar drip.

    As such, it’s completely impossible to rely on the internet for just about anything serious. I say this as someone who uses the internet extensively in everyday life. While it is useful to find resources, gathers news, access archives, ultimately for scientific research the internet is really still just a transmission medium. The real material, discussion, and substance is still contained in books, papers, and seminar talks(I make brief mention of the recent availability of many talks online nowadays which breaks this view of the internet slightly).

    Now, another important aspect of the internet is the tolls it provides to us, but that’s a whole different discussion.

  11. Adam Treat says:

    This post raises an interesting problem with modern experimental physics: the enormous resources required to _repeat_ experiments. Resources including time measured in years if not decades, money measured in percentage of GDB of entire nations, and last but not least very substantial portions of the careers of young physicists.

    And yet we must _repeat_ experiments for the scientific process to establish robust and confident results.

    What do you do if you are leading up the MINOS or T2K experiments? Begin right away at repeating OPERA’s experiment at the cost of substantial resources? Wait a bit in case the OPERA results public announcement leads to the many eyes of the world wide physics community discovering a subtle systemic error? Repeating and confirming experiments, especially at the cost of such huge resources, is not nearly as exciting as doing original new experimentation. And yet it must be done. A real conundrum for modern experimental physics.

    • Chris says:

      I disagree: throw a few million of your favorite currency (USD, GBP, EUR), some people who would gladly volunteer, and wait a year or two. By particle physics standards, this could be a very cheap and quick experiment to confirm or deny Opera. For example, as I understood the MINOS paper, there delays are mainly things like electronics response and cable delays. You can hire people to just measure those.

      Your comment may hold for collider physics, but neutrino physics moves quite fast even being financially limited. Roughly 10 years ago we didn’t think neutrinos oscillated.

      Also: confirming an experiment isn’t boring. If Opera is correct, then it’s the tip of the iceberg.

  12. Frank Heile says:

    Hi Martin, Great blog post! Thanks! Looking forward to many more… – Frank

  13. Jasper says:

    Thank you for the very clear post. I like that you give a lot of references and I’m looking forward to reading your next post!

  14. Ed Minchau says:

    There are a couple of possibilities. First, there may have been some overlooked systemic error in the OPERA experiment. Second, the neutrinos may indeed have been traveling faster than light. If the unknown systemic errors cannot be found, or if a repeat of the experiment produces the same apparent results, what are the implications? To my mind the major implication of FTL particles with mass is a second time-like dimension orthogonal to what we experience as time, as would be required by the Lorentz equations. (Don’t tell Julian Barber.) Interestingly, electronics transform analysis already requires that second time dimension as a mathematical convenience.

  15. Pingback: From the Tau to Dark Energy: Martin Perl’s Blog | Outlooks & Insights

  16. David Galiel says:

    Arrived here via Sean Carroll’s link.

    As an autodidactic in my 50′s with great interest in current physics exploration, the philosophy of science, and the nature of collaborative research in the networked age, I just want to welcome you to the wild and wooly world of the web. I have enjoyed reading your rigorous yet accessible first article, and look forward to what comes next.

  17. Sean Carroll says:

    Welcome to the blogosphere! I’ve posted a pointer to your new blog at Cosmic Variance.

  18. Pingback: From the Tau to Dark Energy: Martin Perl’s Blog | Cosmic Variance | Discover Magazine

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>