Slashdot Mirror
Intergalactic Race Shows That Einstein Still Rules
Ponca City, We love you writes "The NY Times reports that after a journey of 7.3 billion light-years, a race between gamma rays ranging from 31 billion electron volts to 10,000 electron volts, a factor of more than a million, in a burst from an exploding star, have arrived within nine-tenths of a second of each other. A detector on NASA’s Fermi Gamma-Ray Space Telescope confirmed Einstein’s proclamation in his 1905 theory of relativity that the speed of light is constant and independent of its color, energy, direction or how you yourself are moving. Some theorists had suggested that space on very small scales has a granular structure that would speed some light waves faster than others — in short, that relativity could break down on the smallest scales. Until now such quantum gravity theories have been untestable because ordinarily you would have to see details as small as the so-called Planck length, which is vastly smaller than an atom — to test these theories in order to discern the bumpiness of space."
The spread in travel time of 0.9 second between the highest- and lowest-energy gamma rays, if attributed to quantum effects rather than the dynamics of the explosion itself, suggested that any quantum effects in which the slowing of light is proportional to its energy do not show up until you get down to sizes about eight-tenths of the Planck length, according to the Nature paper, whose lead author was Sylvain Guiriec of the University of Alabama.
Granted they say it would have to be proven much smaller than a planck length for most people to accept this as empirical proof, it is empirical data backing Einstein. The 9/10s could be due to the explosion or a physical effect but the latter is now more unlikely given the many light year distance.
My work here is dung.
9/10th of a second is only about 3/4th the distance from earth to the moon. I don't know for sure but I think that's a rather small difference and could be accounted for just by the size of the star that exploded. Our own sun is about 4 seconds across isn't it?
It all starts at 0
What a time we live in where a "slow new day" consists of a 7.8 million year race being recorded (regardless of the results), a fusion reactor is being developed, and a real time speech translator was released.
The event was approximately 2.2 seconds long. Thus it is plausible that these two photons left .9 seconds apart.
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
Actually, it really indicates nothing, except that any "bumpiness" of space doesn't have a profound effect on the speed of light within the wavelength range tested. It's good data. However, this neither proves nor disprove there was an effect, just proves that the effect (if it exists) is very insignificant at the tested wavelengths.
Insignificant != Nonexistent
Tested Wavelengths != All Wavelengths
In order to prove or disprove the theory that light changes speed based on wavelength or other factors, you'd need to be sure that both pulses started the race at the exact same moment, that the two pulses travel through the same space without interfering with each other,and that they complete the race at the exact same moment (ie, within the margin of error of your testing equipment). The margin was almost one second, which is terribly insignificant when compared to 7 billion years, of course, but demonstrates clearly one of the following three things:
1. The pulses left about a second from each other, which we can neither prove nor disprove.
2. The test equipment was flawed and they really did arrive at the exact same time (which leads to #1, maybe they left at different times and just happened to arrive at the exact same moment).
3. The speed of the various wavelengths WAS affected by "space potholes", but it took 7 billion years to accumulate less than one second of variance.
If #3 is possible, which it still is even after this test, then the theory of bumpiness of space has not been disproven, it just appears that evidence points toward the bumps being really, really small or somehow only marginally effective at affecting the speed of light.
Plus, the original article goes on to explain that the tested wavelengths were relatively large, and that much smaller wavelengths might be more susceptible to the "bumpiness" of space depending on the size of the bumps. If the bumps are really tiny, then they might have just tested wavelengths that were too large to be affected by them. If we can measure some really high-frequency (low-wavelength) pulses against the ones we think are nearly identical, that would be much more compelling data.
"This post contains words, known to the State of California to cause thought. Wash brain thoroughly after reading."
9/10th of a second is only about 3/4th the distance from earth to the moon. I don't know for sure but I think that's a rather small difference and could be accounted for just by the size of the star that exploded. Our own sun is about 4 seconds across isn't it?
The speed of light = 299 792 458 m / s (commonly 3.0 * 10^8 m/s)
The average centre-to-centre distance from the Earth to the Moon is 384,403 kilometres
(384 403 kilometers) / the speed of light = 1.28223039 seconds
The diameter of the sun = 1391000 kilometers
(1392000 kilometers) / the speed of light = 4.64321221 seconds
Either you do good math in your head, or I respect your physics teacher for teaching you such interesting facts.
Not true, if the theory requires that they would be separated by say 900 seconds, they left within 2.2 seconds of one another maximum, and we observe them at 0.9 seconds apart, then the theory is proved wrong.
Not true. If we know that the event that generated the rays lasted only 2.2 seconds and we have a theory that would delay one of the rays by more than 3.1 seconds (2.2 + 0.9) relative to the other, we can invalidate that theory. From my understanding, that is exactly the case we're dealing with. You are correct though that this cannot completely validate any specific theory - All it can do is reinforce the assumption that our current theory is more accurate than some others proposed and eliminate some competing ideas.
He's getting rather old, but he's a good mouse.
It doesn't prove that the speed of light is constant, but it does reasonably prove that the speed of light is independent of wavelength, since they left from the same source at the same time.
This is actually just the latest in a series of measurements of this type. Since the Nature paper isn't free online, people may want to look at this similar paper from earlier this year that is available.
The article talks about testing "some theories" of quantum gravity. AFAIK the only theory of quantum gravity that makes anything like a prediction that could be tested in this way is loop quantum gravity (LQG). The two leading contenders for a theory of quantum gravity are LQG and string theory. String theory essentially assumes a background of flat spacetime (plus an xtra 6 rolled-up dimensions), so I don't think it's capable of addressing the issue of whether spacetime is frothy at the Planck scale. LQG doesn't assume a background of flat spacetime, and in fact one of the main research programs in LQG is focused on showing that flat spacetime can emerge as a solution to LQG in the appropriate limit. LQG unambiguously predicts that the vacuum is dispersive, i.e., that the speed of light depends on the energy of the photon. However, LQG does not unambiguously predict the exact form of the energy-dependence. The possible form that is usually assumed in order to evaluate observational tests is |v/c-1|~(E/E_P)^n, where v is the speed of the photon, c is the speed of cause and effect in relativity (often referred to as the speed of light), E is the energy of the photon, E_P is the Planck energy, and n=1 or 2. Previous observations, such as the one in the arxiv paper I linked to above, have pretty much ruled out n=1, so if LQG is right, we'd presumably have to have n=2. Some people have been saying that LQG is ruled out by these measurements, but I don't think that's really correct, it's just constrained by them. Here is a paper by LQG researchers discussing the empirical tests, and they don't seem to be saying "OK, we give up." It's actually very exciting for people in quantum gravity to have observations that even have some chance of disproving a theory (or some version of a theory); the whole field is a dead end if it can never be tested by experiment.
In a broader sense, the holographic principle gives strong, model-independent reasons for believing that spacetime is probably discrete, not continuous, at the Planck scale. Otherwise it's hard to imagine how there could be an upper bound on the information content of a given region of space. And any theory in which spacetime is discrete at the Planck scale will naturally give a dispersive vacuum. Therefore I'd say that either (a) we should eventually observe dispersion of the vacuum once the observations get sensitive enough, or (b) the holographic principle is telling us something that we don't yet understand.
Two good popular-level books that get into this kind of thing are Three Roads to Quantum Gravity by Smolen, and The Black Hole War by Susskind. Because Smolen and Susskind represent very different points of view on quantum gravity, anything that both books agree on is probably correct.
Find free books.
I don't know for sure but I think that's a rather small difference and could be accounted for just by the size of the star that exploded.
You're confused because the summary, and the press release on which it is based, are misleading and wrong.
This is a gamma ray burst (GRB), which originate from neutron stars, not a super-nova (which is the only reasonable meaning one can give "exploding star".) Neutron stars are small, resulting in much finer burst timescales.
The paper discusses the time-structure of GRB's, which has been extensively studied. The fundamental result they get is from a single high-energy gamma ray at the end of the last spike in the burst, which comes 0.9 s after the onset of that spike (seen in the lower-energy photon flux). They do a lot of analysis to argue that the most plausible explanation of that single photon is that it is a member of that spike rather than a random cosmic ray. Anyone familiar with modern statistical techniques will see that this is straightforward, albeit non-trivial.
This is the way science works: we squeeze limited and imperfect experimental evidence as hard as we can using established theory and other, supporting, observations. All the "yeah, well, it could be something else" kind of commentary we see so much of on /. is irrelevant to the scientific process, because it is doing nothing but repeating what everyone already knows: sometimes the most plausible explanation turns out to be wrong.
The exciting thing about this measurement is that they have shown it is possible to put quantum gravity to a rather good test using entirely conventional gamma-ray spectroscopy techniques, and repeating this kind of measurement over the next few years or decades on different bursts will rapidly push down the limits on potential planck-scale effects, because eventually we'll see bursts where there are a few high-energy photons closer to the onset, or we will see bursts from objects at larger (known) distances.
The present authors argue, rightly, that their observation makes theories that have a linear dependence of light velocity on wavelength less plausible. At some point in the next few years it is likely that those theories will be dead, and there's really nothing so beautiful as a theory killed by a fact.
Blasphemy is a human right. Blasphemophobia kills.