Fermi smooths out space

By Phil Plait | December 29, 2009 9:39 am

This news came out a little while ago but I didn’t cover it at the time, and it’s cool enough that it deserves to be covered. I got it from my friends with NASA’s Fermi satellite outreach group. I used to work on Fermi outreach before the satellite launched and was still called GLAST (Gamma-ray Large Area Space Telescope), and it was fun trying to come up with lesson plans and educational efforts based on gamma rays (the Hulk came up a lot).

Anyway, one thing Fermi can do is measure the exact time when high-energy gamma rays hit its detectors. Not too long ago, photons from a distant explosion slammed into Fermi, and it found that all these photons arrived essentially simultaneously from the event, irrespective of their energies.

So what? So, Einstein was right. Check it out for yourself:

Basically, the idea is that some quantum mechanics theories propose that space is irregular, foamy, and bumpy on incredibly small scales, and this means the speed at which photons travel may change very slightly if they are more or less energetic. The difference is so small that it takes very long trips to detect it — imagine two cars traveling at 50 versus 50.5 kph: after a few seconds you’ll hardly see any difference, but over an hour they’re separated by half a kilometer. So the longer the trip, the easier it is to measure.

After 7 billion years, if those specific QM theories are right, two photons should arrive at very different times, but Fermi found that the high energy gamma rays hit Fermi less than a second after the low energy ones. This means that space really is smooth, or at smooth at scales smaller than predicted by those quantum theories. QM is still a solid model for the Universe — after all, solar panels, computers, and nuclear bombs do work — but this means that we need to rethink certain aspects of them.

I love hearing stuff like this. We have lots of ideas on how the Universe works, but we need observations of the Universe to know if we’re traveling down the correct path or not. Fermi has shown us that some of these paths lead to dead ends, and we need to look elsewhere for our journey to continue. And I will guarantee that not only will that journey go on, but we’ll find ever-more roads to investigate as we travel.

CATEGORIZED UNDER: Astronomy, Cool stuff, Science

Comments (25)

  1. Einstein was right? Of course He was! He is a Divine Being, remember? As a scientist, you worship Him, along with His heavenly consort, Charles Darwin. Both are infallible.

    Sheesh! Don’t they teach this stuff in Sunday school anymore?

  2. Gib

    So, they arrived 0.9 seconds apart. Why was that ? Was Einstein slightly wrong ?

    Or are they assuming that they arrived 0.9 second apart because the two different frequencies were generated at slightly different times/places in the original explosion ? And hence it would have been 0.9 second apart anywhere along the path between us and the explosion ?

  3. Kee

    Very interesting! Quick question: How do they know that the different energies of photons were emitted at the same time and space? I don’t mean they could have come from different GRB’s, but maybe they could have come from different events within the GRB’s domain. Maybe the nature of the collision of the neutron stars (or whatever caused the GRB) could have meant that the release of the different photons were fart apart, and the fact they arrived less than a second apart is evidence that their speeds actually were different. This could add evidence to the possibility that the QM predictions are merely off by a certain factor, rather than fundamentally wrong.

    But still, I suppose that a difference of 0.9 seconds after a 7 billion year interval does say something about the stability of lightspeed regardless of energy.

  4. Plutonium being from Pluto

    Are these conclusion Fermi or not not-so-Fermi established now then? ;-)

    (Sorry couldn’t resist.)

    @ 1 Kuhnigget : I presume you forgot the ;-) emoticon again there?

  5. kuhnigget – I think the “you scientists and your dogma” argument doesn’t apply when we’re discussing a rigorous experimental result. Nice try though.

    Gib – Well, within 0.9 seconds. It’s possible they arrived at the exact same time, but we would have no way of knowing – it’s a matter of the sensitivity of our instruments. And anyway, 0.9 milliseconds of wiggle room is nada compared to a few billion years of travel time!

    Kee – So they were emitted at different times and from different distances, and somehow arrived at nearly the exact same time? That’s too big a coincidence for scientists to accept ;)

  6. dcsohl

    I’m curious to know (and maybe this is covered in the video; I can’t watch those from work, so my apologies if so) — how great a time differential were they expecting? 5 seconds instead of 0.9? 30 seconds? 2 minutes? 2 months?

  7. some quantum mechanics theories propose that space is irregular, foamy, and bumpy on incredibly small scales

    Just some? I have not come across any that did not propose that.

    and this means the speed at which photons travel may change very slightly if they are more or less energetic

    I only have a lay knowledge of quantum mechanics, and I have never come across any predictions that the speed of light would vary by frequency due to spacetime foaminess, so it doesn’t seem like a consensus idea.

    Also, how does this article square with this article? I think the popular science press is falling down again.

  8. Stonegiant

    So they were emitted at different times and from different distances, and somehow arrived at nearly the exact same time? That’s too big a coincidence for scientists to accept.

    That’s still an assumption. Yes, it is FAR more likely that you are correct, but where’s the proof?

  9. I thought the spacetime “foaminess” was supposedly a case of virtual particle-antiparticle pairs popping into and out of existence in the most brief flashes of time, on the Planck scale (~10^-35 meters). Even if a gamma ray has a wavelength of 1 picometer (10^-12 meters), isn’t that still waaaay too big to be affected by the foam? Or is it just the case where we’d reasonably expect a shorter wavelength to mean a few more bumps and bruises over the course of a 7-freakin-billion year journey?

    Guess I’m really asking the same thing as others. What amount of delay was expected due to foaminess? I’m sure it depends on which exact QM theory you use, but still, what was the median?

  10. @ Pluto and Adam Solomon:

    Yes, there was a large :) implied there. If not an even larger :P

  11. Plutonium being from Pluto

    ^ I knew that, kuhnigget – I’m just not sure that everybody else would. ;-)

    Sometimes irony /sarcasm / Poe-etry is all too easy for folks to confuse & mistake as the “real” thing, alas.

    Oh & Darwin & Einstein being “consorts” as you delicately put it – nice touch! That’d really have the RTC’s* gasping. LOL. :-D


    * RTC = “Real True Christian” – y’know the small-minded sort that are intolerant of homosexuals & unmarried mothers, hate the poor, hate the reality of evolution and the need to teach it in science class, hate .. well most people that aren’t *them*, and think Jeebus’es message was all about abortion, teh Gayzz & teh Apocalypse / Rapture. That acronym there (no idea how widely its known) comes from this blog here :


    which is another of my faves.

  12. BethK

    Comparing the NASA link with the link Naked Bunny posted makes it sound like we have one example of same time to travel and many examples of different time to travel.

    Is there a more detailed yet still accessible paper about what Fermi observed and how it relates to those other observations?

    I’m googling ‘GRB090510 quantum gravity’, and it looks like articles in the Nov 19 issue of Nature are most relevant. An ArsTechnica article also looks helpful. Off to read

  13. Stonegiant – If you want proofs, become a mathematician. There are no proofs in science, only very high degrees of certainty. Did you know that according to quantum mechanics, there is a non-zero chance that Barack Obama was born in Canada due to quantum fluctuations? Sure, it’s about a 10^-10^56 chance, but it’s non-zero, so you can’t prove it’s false. Don’t tell Fox News ;) The chances of the sort of coincidence you and Gib talk about happening are so astronomically small that they can safely be ignored.

    Kuhnigget – Oh…um…of course I knew ;)

  14. NakedBunny – Your lay knowledge of quantum mechanics (not to mention my senior physics major level knowledge of quantum mechanics) would have nothing to do with these effects – these are proposed quantum gravity effects, very advanced business ;) As for the article you linked, yes, that’s a previous result which these Fermi results disagreed with. The consensus is that that previous result, being older and done with less reliable ground-based technology, is incorrect. It’s much easier to fake a time delay when there is none than to fake none when there is one.

  15. Brett

    While these kinds of results get a big splash in the media, this rather obvious type of experiment turns out not to be a particularly good way of searching for relativity violations. There are already much more sensitive measurements of the same kinds of effects; if anything were to show up in an arrival time test like this, it would be tough to reconcile it with other existing data.

  16. Yousuf Khan

    I don’t know, light of different frequencies arriving 0.9s apart seems to indicate to me that Einstein was wrong, even just slightly.

  17. Phil, you should maybe edit to clarify that it’s a 900 ms upper limit, not a firm measurement of 900 ms?

  18. Gary Ansorge

    Ah, the foaminess of the false vacuum. Just right for my french roast,,,but is it chocolate foam???

    Interaction with “virtual” particles would obviously be greater for short wavelengths than long. So why didn’t they use gamma vs visible? Probably because visible wavelengths interact very strongly with “real” particles and over a seven billion year trip, that would be too much interference to provide good data.

    Great job, Fermi. My congrats to the research crew.

    GAry 7

  19. Torbjörn Larsson, OM

    Actually this extremely exciting observation kills two birds with one stone, not only the theories that proposed spacetime frothiness at small scales but those like Loop Quantum Theory that proposed a fundamental length scale.

    None of those made much sense anyway, at least to me.

    Spacetime is emergent like classical objects independently of gravity being an effective theory, so wouldn’t be susceptible to quantum uncertainty wholesale as an effective quantum field would. (Think ice/spacetime vs water/fields.)

    And Lorentz invariance may be broken as any other invariance. But never destroyed along the simpleminded axiomatic path of some theoreticians “fundamental length scale”.

    More interestingly, this is to my knowledge the first observation beyond Planck scale! In principle “beyond Big Bang”. (IIRC at least Planck length/1.3 or so, also a hint of possible observation to Pl/100 or so – but not this time, alas.) I wasn’t sure we could ever make one, it’s a huge thing in my view.

    @ NBwaW, Moonshark:

    The foaminess was supposed to be spacetime itself being subjected to uncertainty. In these hypotheses its geometry would go from large scale flat spacetime to a small scale dynamic froth of curvature going this way and that, with spacetime itself popping in and out of existence.

    This is quite different from uncertainty of particle fields that regular quantum field theories AFAIU predict and which I believe NBwaW describes. For one, spacetime uncertainty would make the concept of fields and particles “uncertain” too. :-D

    There has been a minor subfield trade of proposing more or less unphysical theories of “quantum gravity” above and beyond simply quantizing the effective field theory of gravity (which is simple at low energies) and get gravitons.

    LQT mentioned above is the prime example. Besides now failing its main prediction of a fundamental length scale big time, it wasn’t much of a physics theory to begin with. [Think mathematicians with delusions of physics grandeur. :-D Never mind that some of those actually contributed to physics elsewhere too.] Famously LQT could neither establish a lowest energy state nor a simple harmonic oscillator to populate it with, so there was no dynamics to be had.

  20. Torbjörn Larsson, OM

    Since I’m not a quantum physicist, I have to provide a reference to the claim of quantizing gravity “is simple”. Here is “a perfectly good effective field theory description of quantum gravity” according to physicist Jacques Distler:

    “In other words, as an effective field theory, gravity is no worse, nor better, than any other of the effective field theories we know and love. The trouble is that all hell breaks loose for ε ~ 1. Then all of these infinite number of coupling become equally important, and we lose control, both computationally and conceptually.”

    Moreover, under certain implausible but possible conditions (i.e. we would know all particles up to Planck scale) and augmented with some technicalities “that theory would be perfectly predictive within its realm of validity.”

    Btw, I note that Loop Quantum Gravity (the usual name for LQT) is discussed as having a fundamental length scale because of difficulties of fine-tuning, in Distler’s word because some people “give up on taking the continuum limt [sic]“. Well then, it’s not the impression I had gotten, but if so formally there is a way out.

  21. T.E.L.

    The supernova wasn’t a point source; it was an extended object with a finite size. The difference of 900 milliseconds translates to a spatial difference of only 270,000 km, easily within the diameter of the exploding star.

    There’s also some misconception going on here about how this relates to special relativity. Special relativity doesn’t predict that space is smooth at all scales. It postulates that the speed of light is the same within every inertial frame of reference. This is a local theory, which means that it only holds strictly on an infinitesimal scale. If the geometry of space meanders about in a complex way at some finite scale, then naturally light should behave differently in practice than in the idealized system of special relativity. If the two gamma photons, of differing wavelengths, had arrived at significantly different times, it wouldn’t have been a categorical violation of relativity.

  22. Dr Cy Coe in NL

    Is Phil’s analogy accurate?

    imagine two cars traveling at 50 versus 50.5 kph: after a few seconds you’ll hardly see
    any difference, but over an hour they’re separated by half a kilometer. So the longer the
    trip, the easier it is to measure.

    Does the tiny-scaled foaminess and bumpiness imply that photons travel the entire distance at different but constant speeds?
    It seems to me if tiny-scaled foaminess is affecting speed, than the effect on the speeds of two photons should average out over the entire distance, just as the Fermi results show.


  23. Dave

    I’m a little new to all this but since this observation infers that spacetime is not foamy, does it also imply that Loop Quantum Gravity is incorrect or?


  24. IN many ways the notion of a finite resolution to space (i.e., the Planck length) has always struck me as reminiscent of the Renormalization technique in Quantum Electrodynamics. Even Feynman himself referred to it as a kind of goofy trick, subtracting infinities from infinities to force a finite answer to values. Using a finite minimum length scale seems another dodge to avoid infinities that crop up in flawed mathematical descriptions.

    Also, we have totally forgotten our limitations of description. We speak as if looking at a trajectory, line of travel 7 billions light years long (!). Strictly speaking, particle trajectories do not exist, even for photons, since we cannot directly observe them. The straight line path assumed for photons is a result of symmetry in the Feynman integral over all possible paths, i.e., left and right deviations are symmetrically distributed in the bundle of paths so that the integral averages/cancels them out, leaving the straight line.

    In actuality, we can only observe the presence or absence of a photon at a particular position and time; we have no way of continuously “observing” the position of the particle, since this would imply continuous measurement; it would also violate Uncertainty.

    I agree that this is a great event, however, since it is has pointed out an indirect measurement below the Planck length since the absence of effect implies the absence of a finite “resolution” to spacetime. Or at least implies it is much smaller than predicted.

    The idea of space itself not knowing where it is, popping around like a particle, at first attracted me because I thought that quantum behavior could then be attributed to it, rather than to the particles. If the space an electron exists in has an uncertain location, then any particle in that space would be fuzzed also. But I abandoned it eventually because it did too much violence to the concept of space.

    I’m glad the Fermi observation result seems to uphold the null hypothesis: that space is continuous. If the sum over histories (Feynman integral) averages out large scale deviations, why not small scale deviations as well? Is there any reason to believe that the alleged quantum “foaminess” is asymmetrical (especially at the 7 million light year scale of distance)?


Discover's Newsletter

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


See More

Collapse bottom bar

Login to your Account

E-mail address:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »