It’s one of the cardinal laws of physics and the underlying principle of Einstein’s relativity itself: the fact that there’s a universal speed limit to the motion of anything through space and time, the speed of light, or c. Light itself will always move at this speed (as well as certain other phenomena, like the force of gravity), while anything with mass — like all known particles of matter and antimatter — will always move slower than that.

Image credit: Matt Howard.

Image credit: Matt Howard.

But if you want something to travel faster-than-light, you aren’t, as you might think, relegated to the realm of science fiction. There are real, physical phenomena that do exactly this, and yet are perfectly consistent with relativity.

Image credit: Les Bossinas (Cortez III Service Corp.), 1998 / NASA.

Image credit: Les Bossinas (Cortez III Service Corp.), 1998 / NASA.

Brian Koberlein explores a few of them in great detail, with spectacular results. Check it out!


  1. #1 Ken Ziebarth
    United States
    May 26, 2015

    The one I like is sweeping your laser pointer across the moon. The edge of its beam moves across the lunar surface faster than c!

  2. #2 Sinisa Lazarek
    May 26, 2015

    @ Ken
    in that scenario nothing actually moves. The spot from a laser is not a material object, and the photons themselves that make it and your eye perceives, don’t move faster then c.

  3. #3 Michael Kelsey
    SLAC National Accelerator Laboratory
    May 26, 2015

    @Sinisa #2: I think Ken knows that. The edge of the beam isn’t an object, and the FTL movement at the lunar surface can’t be used for FTL communication. Ethan had a guest post on exactly this topic a month or two ago, if I recall correctly.

  4. #4 PJ
    Perth, west Oz
    May 26, 2015

    Would this also depend on the rate of the sweep? If the laser is a diode (ie; not pulsed), the photons are going to arrive at the moons surface (average) 1.28 seconds after leaving the laser. If the sweep is very slow compared to the moons half degree apparent width, it would be less than C. Even if the sweep was (faster than C), some of the photons are still going to arrive 1.28 seconds after their departure

  5. #5 Sinisa Lazarek
    May 27, 2015

    @ Michael

    Don’t know what Ken knows or not. As for guest post.. i know, i was here.. lot of misleading info in that guest post.

  6. #6 DJ
    May 27, 2015

    I’m curious about that entangling thing. There’s no way to send information based on what state it ends up in, but is there a way to send information based on the fact that a state has been read, ignoring the state itself? Like, when you measure one particle, is there any way for the other side to know THAT you measured it, that “their” side just got “locked”, by making some sort of device that responds when it gets locked?

  7. #7 Sean T
    May 27, 2015


    I must confess that QM is confusing to me, so I may be wrong here. However, I don’t believe that there is any way for an observer working with one member of an entangled particle pair to know that the other member of the pair has been measured short of actually communicating with the person doing the measuring on the other particle. If you have one member of an entangled pair and you measure some attribute of it, you would notice absolutely nothing out of the ordinary. For example, if you measure the spin of an electron about some axis, you get either +1/2 or -1/2. Nothing about that measurement indicates entanglement. Even if you communicated your result with your partner, (who would say he/she got the opposite value), that would not really be conclusive evidence that your particles were really entangled. You could measure a pair of non-entangled electrons and get opposite results by chance. (50% probability of that occurring in fact). To really detect anything strange, you’d need to measure a large number of electron spins. If you measured 20 of them, for example, you’d get some string of + and – values, occurring randomly. You would not notice anything until your partner gave you his/her values. If you find that your partner measured the opposite value for all 20 measurements, you now are fairly confident that you were measuring entangled pairs since there is only a (1/2)^20 probability of that occurring by chance. That is a bit less than a 1 in a million probability, so it’s very likely you have entangled pairs.

    The key point is, though, that you really only can detect anything strange happening after the fact. You can’t look at the measurements and conclude anything about the other member of the pair. In particular, you generate a random string of values whether you go first or your partner does. The second set of measurements, looked at in isolation, is just as random as the first. It’s like looking at a series of coin flips and seeing HTTHHHTHTHTH. If you saw that sequence, you would have no reason to believe it’s not a random sequence. If you saw THHTTTHTHTHT, you likewise would not think anything of it. It’s only when you compare the two after the fact that you’d notice anything non-random about the situation.

  8. #8 Sinisa Lazarek
    May 27, 2015

    @ DJ

    Einstein had a bit of a naive representation about entanglement, but for this is purpose it is good. In short..he said.. imagine you had a socks in a drawer .. and there’s a red one and a yellow one.. i.e spin up and spin down of entangled particles… if you take a sock and you see it’s red. you know that there’s a yellow one in the drawer.. same thing for entangled particles.. if you have an entangled pair and you measure one particle.. you.. and only you know the state of the other particle.. but there’s not much you can do with that info. You don’t affect the state of the other particle…. you just measure… the states were defined at the moment of entanglement.

  9. #9 Russell Seitz
    May 27, 2015

    You see superluminal red light every time you look at the world through a glass window with less than perfectly parallel surfaces – the dispersion of its refractive index with wavelength means the red light in the scene propagates faster then the blue light.

  10. #10 Denier
    United States
    May 27, 2015

    I scoff at the petty FTL imitators. The true King of FTL is the off-shell virtual photon. The virtual photon can create a link between point A and point B instantly. This is no slowpoke tachyon which takes time to travel from here to here. How pedestrian. The virtual photon’s entire plane wave is created at the same time and is completed in under Planck time.

    So you have a laser pointer aimed at the moon? Bwahahahah. I’ll bet that old thing uses batteries. Not so with the virtual photon. The virtual photon links two points in space, faster than light could traverse the distance, and borrows all of its energy from space-time. No external power source needed.

    Although the virtual photon is bound by perturbation theory, it is completely exempt from the conservation of energy, obviously exempt from 186,000 mph speed limits, and exempt from Special Relativity. The off-shell virtual photon is too cool for school.

    All hail the King!

  11. #11 Wow
    May 28, 2015

    “The virtual photon can create a link between point A and point B instantly. ”

    We’ve been there before.

    No, it doesn’t.

    You didn’t listen at all last time, and I’m not expecting any different here.

  12. #12 Denier
    United States
    May 28, 2015

    @Wow #11

    Of course it doesn’t. It isn’t a real thing. What dark matter is to gravity, and dark energy is to cosmic expansion, virtual particles are to subatomic processes. A virtual photon connects two points only as fast as a physicist can draw the wavy line on a Feynman diagram. In the real world it isn’t real. The processes are real. The math describing the properties of the force carriers are real, but until they figure out sting theory, or loop quantum gravity, or however the universe actually works on the smallest scales, there remain some unknown areas of exactly how the communication works.

    Those unknown areas are precisely what allows the V-I-R-T-U-A-L particle to do nonsensical stuff, because the nonsensical stuff isn’t mathematically ruled out. That includes faster than light travel. If you look up almost any collection of possible FTL travel or communication (like here, or here), virtual particles are listed.

    If you continue to cling so desperately to the concept of virtual photons as bullets of matter flying around instead of the mathematical placeholders they are, then I say run with it. Cross the universe faster than a speeding laser pointer.

  13. #13 Wow
    May 28, 2015

    “Of course it doesn’t.

    Then why did you say it did????

    What dark matter is to gravity, and dark energy is to cosmic expansion, virtual particles are to subatomic processes.

    No it isn’t. All they have in common (which so many other things do: people can be quite inventive) is that many people don’t understand but feel this should be no impediment on making claims about it.

    Those unknown areas are precisely what allows the V-I-R-T-U-A-L particle to do nonsensical stuff

    It’s only nonsensical if you make up what they “are doing” out of thin air and hope and make out that this is what they are doing.

    If you continue to cling so desperately to the concept of virtual photons as bullets of matter flying around instead of the mathematical placeholders they are

    They aren’t either bullets (I certainly don’t call them that) nor are they mathematical placeholders.

    Or, at the most generous disposition toward your ludicrous claim, that is definitely not a settled issue. Some theories for dark matter propose a change to the form of newtonian mechanics, but there aren’t many usable theories that manage it, unless they are only a small part of the reality, meaning that dark matter as ordinary matter, not a mathematical placeholder.

    They are a *mnemonic* placeholder. Like “thingy”. A meta-syntatic variable. A placeholder for some thing that will be properly described, in the same way as electrons are described.

  14. #14 Rose
    May 28, 2015

    I look forward to seeing your plans for a perpetual motion machine.

  15. #15 em
    May 28, 2015

    crappy article. imprecise on so many points.

  16. #16 Michael Kelsey
    SLAC National Accelerator Laboratory
    May 29, 2015

    Ethan et al., there’s a really cool preprint out which analyzes, in a very simple special relativistic structure, superluminal travel and closed timelike curves:

    The math is quite straightforward algebra, and the authors construct their scenarios in a reasonable way (similar to most undergraduate relativity courses) to avoid “infinite energy” concerns.

    I found it quite enlightening, and gratifying, to see actual worked-out examples for the trope that superluminal motion is equivalent to time travel. It turns out that is true, but only for particular regimes of superluminal speeds.

  17. #17 Alexander Roth
    March 29, 2016

    This was a good review of some super-lumic examples but a little short on practical applications.

    Is it possible to PROVE that light speed, “c”, is the ultimate speed? It can certainly be demonstrated that by assuming a signal speed greater than c unacceptable paradoxes and violations of causality result.

    But is there a direct physical proof of the speed limit? Depending only on paradoxes weakens the argument.

    Basic to special relativity is the idea that there is no special frame of reference. Does this carry over into the universe at large? From any point in the universe we can measure the sum of the vector velocities to a large number of galaxies. If the measurement is other than zero, we can adjust our velocity to make it so. Have we not thereby found a special frame? We can also deduce the Hubble constant by noting the galaxy velocities vs distance. The Hubble constant thus derived gives us the age of the universe as the time since the big bang, TBB. This process, finding the special frame and the absolute TBB gives us points of absolute time and space as we wander around the universe. Of course the special frames vary smoothly as one changes position.

    In this view, accepting the existence of special frames, paradox analysis becomes a little different. It can be shown that an infinitely fast signal, IFL, under the right conditions, will definitely not lead to a paradox. Is that enough to provide some hope of very distant high speed communication? See my web essay at Fermisquestion(dot)com and note the subsection “Instantaneous Communication”.

  18. #18 Sean T
    March 30, 2016

    There is experimental evidence that relativity with it’s maximum speed is a better description of the universe than classical mechanics, which lacks any maximum speed. Consider this experiment: electrons are accelerated through a known potential difference. They are then detected by two separate detectors with a known spatial separation. This setup allows the energy and velocity of the electrons to be determined.

    Now plot the velocity against the energy. Classical mechanics says K = 1/2 mv^2. Therefore this plot, assuming CM is right, should look like a parabola with velocity increasing without limit as energy increases. This experiment has been performed, but the result is completely different. The plot is not a parabola and velocity does not increase without limit as energy invreases. Instead, the velocity approaches c assymptotically as energy increases, exactly as predicted by relativity and confirming the maximum speed limit.

  19. #19 Wow
    March 31, 2016

    “But is there a direct physical proof of the speed limit? ”

    You can easily prove it wrong: demonstrate something going FTL.

    There was one thought that an experiment had, but it turned out to be a machine error.

    This is how you do science. Overturn accepted science WITH BETTER SCIENCE.

  20. #20 Alexander Roth
    April 2, 2016

    Thanks Sean T, but that does not answer my question. Einstein showed us in SR, What the experiment you described showed us, that mass increases with velocity reaching infinity at “c”, thus creating a barrier at c. But can not the uncertainty principle allow tunneling from a little less than c to a little more than c? If so, what is the tunneled particle like?

    Wow: The world is still waiting for a verifiable demonstration of a signal (information) exceeding c. Entanglement doesn’t exactly do it.

    My question asked not about a particle exceeding c, but a signal — without defining it, exceeding c. We can call it a tachyon and sidestep any further description of it. But even the tachyon fails at FTL speeds because of violations of causality. In other words, paradoxes. My question is: Even dealing with unknown particles, how can we prove that FTL cannot exist in any form, other than by showing it results in paradoxes. A direct proof showing that our Universe cannot support any type of FTL would be a better proof than showing that paradoxes result from FTL.

    However, there might be a cosmic loophole (no relation to a wormhole). This is discussed at my website: (Click on subsection, “Instantaneous Communication”

  21. #21 Michael Kelsey
    SLAC National Accelerator Laboratory
    April 2, 2016

    @A. Roth #17 and #20: Ethan has made it pretty clear that flogging self-published, non-peer-reviewed personal theories is not appropriate in the comments on this forum. Your rhetorical question used as incitement to introduce your own hypothesis falls under that category.

New comments have been temporarily disabled. Please check back soon.