# Fermi (GLAST) almost kills all Lorentz violating theories.

1. Feb 22, 2009

### MTd2

http://www.symmetrymagazine.org/breaking/2009/02/19/most-extreme-gamma-ray-blast-also-probes-quantum-gravity/Fermi%20Observations%20of%20High-Energy%20Gamma-Ray%20Emission%20from%20GRB [Broken]
http://www.nasa.gov/mission_pages/GLAST/news/high_grb.html
http://sciencemag.org/cgi/content/abstract/1169101v1

The measurement sets the best bounds on Lorentz violation we have so far. In the parametrization popular among the Lorentz-violating self-described "quantum gravity theorists", the typical scale where the Lorentz violation starts to be big must be above 1/10 of the Planck mass. Everything below this gigantic mass scale has been proven to respect the Lorentz symmetry, if you wish.

People who have been promoting a complete denial of special relativity - in the form of various loop quantum gravities, doubly or deformed special relativities, dynamical triangulations, and all this stuff have been idiots, just like I have always emphasized."

http://motls.blogspot.com/2009/02/fermi-glast-almost-kills-all-lorentz.html" [Broken]

Last edited by a moderator: May 4, 2017
2. Feb 22, 2009

### atyy

Isn't CDT Lorentz invariant? There is a discretization of space for the computation, but the action for the computation is not the continuum action, so CDT is more like a computational form of asymptotic safety, which shares the effective field theory framework with string theory, just differing in the guess as to what the "true" degrees of freedom are.

3. Feb 22, 2009

### MTd2

Yes, that's what I was thinking. CDT was invented to help with strings computation...

4. Feb 22, 2009

### jostpuur

Some confusion probably rose from the fact, that the link in the final period, indicating that your post contained a copy paste quotation from Lubos' blog, is quite difficult to detect

Last edited by a moderator: May 4, 2017
5. Feb 22, 2009

### marcus

As it happens, the main form of LQG that has been researched in the past couple of years is Lorentz invariant.

6. Feb 22, 2009

### marcus

Not exactly, it fixes a particular time-slicing. I believe requiring Lorentz invariance would tend to favor fully covariant LQG over CDT.

Don't misunderstand me: I like the CDT work a lot.
And also I don't think Lorentz violation at Planck scale has been ruled out.
My only point is that besides the fact that the comments quoted in the original post do not make much sense, CDT (or string thinking for that matter) is not especially favored by the Fermi results.

7. Feb 22, 2009

### confinement

Can anyone explain this:

8. Feb 22, 2009

### BenTheMan

I thought this was one of the ways to "test LQG"---is that true, even in some limited sense?

9. Feb 22, 2009

### marcus

There's a certain amount of unreliable and misleading material on the web which is better just to ignore, but I can explain the standard dispersion calculation which for several years has motivated people to look for Lorentz violation "suppressed" by the planck energy in GRB and flares which have traveled distances on the order of a billion lightyears.

A good source on this is a 2007 article by several string theorists (John Ellis, Nick Mavromatos...) together with a team of astrophysicists. I'll get the link later.

Unfortunately not. The mainstream LQG development (e.g. Rovelli and co-workers) has never predicted Lorentz violation. Socalled DSR was investigated a lot up to around 2006 but people have lost interest. After 2006 you don't hear much about that from LQG people. Before that considerable effort was made trying to prove that some form of LQG implied DSR, but it failed! (It worked in 3D, results by Freidel and Livine, so there was hope it would extend to 4D, but it didn't.)

In fact, as you may remember, in 2007 when results from a imaging air cherenkov telescope (IACT) suggested a few minutes delay of high energy photons in a gammaray flare, it was some string theorists who rushed into print about it----claiming a variant of string theory predicted it .
The LQG community said nothing. Remarkably. None of the LQG folks said "I told you so."

Last edited: Feb 22, 2009
10. Feb 22, 2009

### Haelfix

I'd say its more problematic for things like Doubly special relativity and the host of modified dispersion relations, Lorentz breaking theories on the market that were always plagued with conceptual problems. Absent a workaround, it doesn't kill them, but it makes them much less relevant for Quantum Gravity.

11. Feb 22, 2009

### marcus

Yes, it doesn't concern the main LQG research program. That was just innuendo of questionable sincerity.

Here, I found the article by the Cherenkov telescope team and the string theorists Ellis, Mavromatos, and Nanopoulos:
http://arxiv.org/abs/0708.2889
Probing quantum gravity using photons from a flare of the active galactic nucleus Markarian 501 observed by the MAGIC telescope
J. Albert et al. (for the MAGIC Collaboration), John Ellis, N.E. Mavromatos, D.V. Nanopoulos, A.S. Sakharov, E.K.G. Sarkisyan
12 pages, 3 figures, Phys. Lett. B
(Submitted on 21 Aug 2007)
"We analyze the timing of photons observed by the MAGIC telescope during a flare of the active galactic nucleus Mkn 501 for a possible correlation with energy, as suggested by some models of quantum gravity (QG), which predict a vacuum refractive index$$\simeq 1 + (E/M_{QGn})^n, n = 1,2$$. Parametrizing the delay between gamma-rays of different energies as$$\Delta t =\pm\tau_l E$$ or$$\Delta t =\pm\tau_q E^2$$, we find $$\tau_l=(0.030\pm0.012) s/GeV$$ at the 2.5-sigma level, and $$\tau_q=(3.71\pm2.57)x10^{-6} s/GeV^2$$, respectively. We use these results to establish lower limits $$M_{QG1} > 0.21x10^{18} GeV$$ and $$M_{QG2} > 0.26x10^{11} GeV$$ at the 95% C.L. Monte Carlo studies confirm the MAGIC sensitivity to propagation effects at these levels. Thermal plasma effects in the source are negligible, but we cannot exclude the importance of some other source effect."

Somebody asked about the math. Basically the hypothesis is that the higher energy photons are slowed by a tiny percentage proportional to
E/Eplanck. The fractional decrease in speed is proportional to DeltaE/Eplanck by a proportionality constant of order one.

The Cherenkov telescope was able to observe TeV photons from a brief flare at Markarian501, it could see their tracks as they entered the atmosphere and estimate the energy of individual photons. So Delta E was on the order of a TeV. You can divide a TeV by the Planck energy and see how large the expected fractional slowdown would be.

Now this tiny fractional slowdown, acting over a distance like a billion lightyears, translates into a delay on the order of minutes. So it is observable.
I don't have time to give more explanation now but if it isn't clear, ask. I'll get back to it.

Last edited: Feb 22, 2009
12. Feb 22, 2009

### BenTheMan

Sure, somewhere in the multiverse you can find Lorentz violations, probably.

I was under the impression that Lorentz violations was one of the only ways to test the current LQG theories, but apparently that's wrong? That's fine---are there other ways to test it?

13. Feb 22, 2009

### MTd2

Yes, by seeing if there is a kind of background that would radomly disturbe otherwise straight lines.

14. Feb 22, 2009

### BenTheMan

How is that an experimental test of LQG?

15. Feb 22, 2009

### atyy

Yes, the time slicing means Lorentz invariance may not be there. But it doesn't mean it isn't there, since GR (or at least most physically meaningful bits of it) can be formulated in ADM or 3+1 form. Benedetti et al seem to think of CDT as computational AS in http://arxiv.org/abs/0901.2984 (see what they say about reference 17). If that's the case (is it?), then presumably CDT would have Lorentz invariance from AS?

16. Feb 22, 2009

### marcus

The topic of this thread is observational constraints on Lorentz violation.
The thread is not about ways to test LQG. (or string theories, for that matter.)
If anyone wants they can start a special thread about ideas for testing LQG.

There are quite a few papers on this, just in the past couple of years.

Last edited: Feb 22, 2009
17. Feb 22, 2009

### BenTheMan

Does this mean you can't summarize, in a single post, an answer to my question?

18. Feb 23, 2009

### xylbs

Since Lorentz symmetry is violated in noncommutative spacetime,
does this result mean the noncommutative spacetime theories also be ruled out?

19. Feb 23, 2009

### marcus

Lorentz violation has not been ruled out.

20. Feb 23, 2009

### marcus

I couldn't get one of the links in MTd2's top post to work. I think it is this one:
http://www.symmetrymagazine.org/bre...-gamma-ray-blast-also-probes-quantum-gravity/

So far I have not seen the actual article by Abdo et al, that all this fuss is about. It was published in a recent issue of Science
Here's the link to the abstract, given in the top post of this thread:
http://sciencemag.org/cgi/content/abstract/1169101v1

It sounds, from the summary and paraphrase Ive seen, as if they came within a factor of 10 of ruling out something Smolin wrote about as a possibility, but did not establish as a LQG prediction, namely an order one proportion between the fractional change in speed of light and this fraction
E/MPlanck

Judging from the paraphrase, they didn't actually rule that out, they ruled out violations 10 times more extreme than what was conjectured, namely what you get if you replace the Planck energy by 1/10 of the Planck energy.

Then the fractional change in speed of photon is related to the photon energy E by order one proportion to
E/(0.1 MPlanck)
which is the same as 10 E/MPlanck
so the fractional change in photon speed is much larger.

But as far as I remember, nobody (not even the string theorists Ellis, Mavromatos,...) was talking about that.

So if I can believe, from the top post paraphrase, that this is all Abdo et al reported, then we have an exaggerated fuss. They have showed that they can get within a factor of ten---so the technology is promising and hopefully they can eventually succeed in testing Lorentz violation at the interesting order-one level.
And in any case nobody ever succeeded in deriving a prediction of Lorentz violation from the main version of LQG, although they tried a lot around 2005 and 2006. I don't know any version of 4D LQG that anybody managed to make predict Lorentz violation, but maybe someone can dig something up if you go back a few years.
However that may be, Ellis and Mavromatos say they have a version of non-critical string that predicts just the same thing that was being talked about in LQG circles. And this type of violation has not been ruled out by Abdo et al. If we can believe the lead post here, they only came within a factor of 10.

Maybe someone who has seen the full article in Science magazine will correct me on this, or say more precisely what the result was.

Last edited: Feb 23, 2009