Freidel talk on proposed new physical principle (relative locality)

[marcus]

Freidel will be giving a talk in the online seminar ILQGS, about Relative Locality---the subject of this paper:

http://arxiv.org/abs/1101.0931
The principle of relative locality
Giovanni Amelino-Camelia, Laurent Freidel, Jerzy Kowalski-Glikman, Lee Smolin
12 pages, 3 figures
(Submitted on 5 Jan 2011)
"We propose a deepening of the relativity principle according to which the invariant arena for non-quantum physics is a phase space rather than spacetime. Descriptions of particles propagating and interacting in spacetimes are constructed by observers, but different observers, separated from each other by translations, construct different spacetime projections from the invariant phase space. Nonetheless, all observers agree that interactions are local in the spacetime coordinates constructed by observers local to them.
This framework, in which absolute locality is replaced by relative locality, results from deforming momentum space, just as the passage from absolute to relative simultaneity results from deforming the linear addition of velocities. Different aspects of momentum space geometry, such as its curvature, torsion and non-metricity, are reflected in different kinds of deformations of the energy-momentum conservation laws. These are in principle all measurable by appropriate experiments. We also discuss a natural set of physical hypotheses which singles out the cases of momentum space with a metric compatible connection and constant curvature."

The audio and slides of the ILQGS talks are made available online, there is often some lively discussion involving people at Marseille, PennState, Perimeter, Nottingham, Potsdam, Warsaw... When this happens it seems to provide additional information (about how the ideas are being received) beyond what one gets by reading the article.

I am curious to know how this proposed Relative Locality principle will be received.

Here is the ILQGS website:
http://relativity.phys.lsu.edu/ilqgs/
This lists the seminars which have been given on LQG topics over the past several years, giving links to audio and slides. As you listen to the audio, the speaker will say which slide number, and when to advance to next slide.

Freidel's is the fourth talk in this spring's series, scheduled for 1 March.


Jan 18 Inflationary observables and observational constraints in LQC Gianluca Calcagni Albert Einstein Institute
Feb 1st Spinfoam cosmology Francesca Vidotto Marseille
Feb 15th Semiclassicality of the loop quantum universe Tomasz Pawlowski University of New Brunswick
Mar 1st The principle of relative locality Laurent Freidel Perimeter Institute
Mar 15th Quantum deformation of 4d spin foam models Winston Fairbairn Hamburg University
...
...

 

[marcus]

"...So look around. You see colors and angles, i.e. you are seeing into phase space. The idea that underlying it is an energy independent, invariant spacetime geometry could be an approximation, reliable only to the extent that we measure the geometry with quanta small compared to the Planck energy and we neglect phenomena of order of |z||Γ|k|/h. Whether this is correct or not is for experimental physics to decide. If it turns out to be correct, then a new arena opens up for experimental physics and astronomy, which is the measurement of the geometry of momentum space."

concluding paragraph of http://arxiv.org/pdf/1101.0931

 

[Fra]

For a starter I skimmed for interesting reasoning and following from their paper sounds like great keys to my liking:

1) "We local observers do not directly observe any events macroscopically displaced from our measuring instruments."

2) "The idea that we live in a spacetime is constructed by inference from our measurements of momenta and energy."

3) "why should we presume that we construct the same spacetime from our observations as observers a cosmological distance from us?"

4) "In this paper we show that absolute locality, which postulates that all observers live in the same space time, is equivalent to the assumption that momentum space is a linear manifold. This corresponds to an idealization in which we throw away the information about the energy of the quanta we use to probe spacetime and it can be transcended in a simple and powerful generalization of special relativistic physics which is motivated by general considerations of the unification of gravity with quantum physics."

As always, things can be read in different way but as I see it this tangents to some of may favourite ideas:

The conceptual point I pull from that (and which conincides with my opinon anyway) seems to be that "locality" in space is something that needs to pass through the "inference system" of a particular observer, and this is of course - observer dependent. So there is no absolute locality.

That's is what I consider the main point.

Next, their "solution" to the argued problem, they argue for a construction that "relativize" this observer dependence, like is common practice in physics. So as to restore locality in a difference space.

This is is what I in the general case considers throwing away information about hte observers "inference system". Ie. in current models it's simply ASSUMED that all inference systems are equivalent.

It's even close at hand to connect this to the renormalization game; one of the main lessons is that INFERENCE of the THEORY (the coupling constants - which is the parameterization of theory) is again dependent on the inference systems (energy and momentum scaled used for interactions).

Regardless of where all details in the paper ends up perfect these basic key ideas seems like a step in the right direction, so I like the paper. (Of course my own idea is that the relativity of observer inferences dependon on much more than ONLY momentum and energy scales! but that's taking this further that they suggest)

Thanks for the link Marcus!

Edit: I think part of the implications is of course that the often resulting apparent non-locality of quantum theories is the result of comparing "inferences" from incompatible inference systems (ie. observers). This is conceptually analogous to what we had pre-relativity where it was found that special connections or parallell transports are needed to compare say vectors in different tangent spaces on the manifold. The story is similar here but far more generalized: as I extrapolate this: we need to find the corresponding transformations rules so that "inferences" can be translated between two inference systms (observers). This clearly tangents to all things in physics, the logic behind relativity, the logic behind renormalisation, the logic of how how inferences are made (QM).

One can even argue that what we seek is a generalization of the parallelltransport transformations as we generalize what constitutes the set of observers. Like has been done lready in SR -> GR. That corresponds to considering larger and larger sets of observers, and thus requiring more complex transformations.

/Fredrik

 

[MTd2]

Wow, finally something that Fra actually likes!

 

[Fra]

Wow, finally something that Fra actually likes!

That's funny

I see that sometimes it may appear like I object to everything, since the objections apply to apparently all programs. It's just because I think it's better to direct the efforts where the constructions are the weakest, even if it means reconstructing a lot of things.

I find that even though in a modest and constrained form, to acknowledge that some structures simply are the result of the observers inference, and that we need to understand how inferences scale, is a step in the right direction worth hailing.

I see physics progressing at two fronts. Those working with experimental physics, or applications obviously can't sit and question the basic foundations as they would get nothing done. These people need to work from what we think we konw and try to improve and collect new.

From a theorist point of view however, I think some current open issues calls for rethinking that is far more radical that what the big programs allow for. This is also why most papers are moderatly interesting to someone looking for fresh thinking.

/Fredrik

 

[marcus]

compare http://arxiv.org/abs/quant-ph/0604064
Relational EPR
Matteo Smerlak, Carlo Rovelli
Foundations of Physics 37:427-445 (2007)
(Submitted on 10 Apr 2006)
"We study the EPR-type correlations from the perspective of the relational interpretation of quantum mechanics. We argue that these correlations do not entail any form of 'non-locality', when viewed in the context of this interpretation. The abandonment of strict Einstein realism implied by the relational stance permits to reconcile quantum mechanics, completeness, (operationally defined) separability, and locality."

 

[Fra]

compare http://arxiv.org/abs/quant-ph/0604064
Relational EPR

Yes the logic of reasoning in that paper is indeed identical and equally excellent.

I commented on Rovelli's RQM in past threads and part of what he says is IMO just perfectly on spot. There are long sections of that paper (basically section 2) where I pretty much agree on every word!

The part of Rovelli's reasoning implicit in PARTS of his RQM is IMHO excellent. This I never questioned. One can tell that Rovelli isn't arguing for someone who shares these basic things already, he argues for those that are still stuch in too much realism.

They write in the end:

"Needless to say, the weakening of realism implied by RQM may be considered too high a price to pay by some."

This is not even by far an issue for me. The weaking of realism is not a price or aproblem, it's a core insight. I rather think Rovelli does not take this excellent reasoning ever further! The reason why I disagree with rovelli's take on QG (the way he applies QM to the equivalence class of observers) is touching subtle details that's not metioned in this paper.

But this radical thinking, is maybe "inhibited" by Rovelli's conservatism in not wanting to modify QM. He is trying to be both bold and conservative at the same time.

Indeed to question more, it gets considerably more difficult. But I can't help myself to think that it's nevertheless the only reasonable thing to do. It will take whatever time needed.

If we for a second release ourselves from thinking that QM is perfect, GR is perfect and SM is perfect and that we can in fact change everything if we want... and try to just rethink wht Rovelli and Smerlak says about the meaning of the state vector etc, subjective encoding histories of interactions etc and try to merge that with difference observers...then it seems clear that there are more things we don't understand.

I think the basic spirit of RQM is great, we just need to be even more crazy to wrap it up with gravity and unification.

Of course it's hard to discuss this, if some people still are stuck back at: "Needless to say, the weakening of realism implied by RQM may be considered too high a price to pay by some.". Those who think this, probably can't even imagine how it can get worse.

/Fredrik

 

[marcus]

EDIT: Just saw your comments after I posted this, Fra. Thanks for the encouraging and insightful post! This is what I had written:

As I recall from reading the 2006 Smerlak Rovelli, reality is viewed as a consistent set of observers, being able to agree on events to the extent that they can communicate about them.

Each observer has his/her own Hilbertspace of quantum states which encodes the information that this observer has acquired*.

There is no one big comprehensive Hilbertspace---and no one official view of reality. This is the "full Einstein realism" which it is proposed we give up.

Instead of believing in the existence of one Hilbertspace, one quantum state, it is suggested that we believe in a process of dialog and discussion among observers which can and does lead to agreement about specific events.

====================

I could be misremembering, but as I recall the spirit of the 2006 Smerlak Rovelli was that we are all in the same universe and that disagreements about what has occurred can be ironed out by communication, which however takes time.

Seeming paradoxes or logical contradictions only arise if one ignores the slow pace at which real observers can actually communicate. Or if one pretends to be an instantaneous super-observer overlooking the whole show.

====================

I'll have to look back at the 2006 paper. In the meantime it seems that with this 2011 Freidel et al paper, something analogous is happening. Each observer has his/her own phase space. One does not assume that one can put everybody's picture together into one official composite spacetime modeling the universe.

The 2011 paper and the 2006 paper may not be saying the same thing but I gather that they are comparable, or on the same wavelength. If you see them saying radically different things, please tell me and explain how they differ!

=====================

* I picture each Smerlak Rovelli observer as having a Hilbertspace which looks like a laptop. As the observer gets new information he/she types it into the laptop. That changes the observer's quantum state---that observer's own personal wavefunction or quantum state of information. What is given up is the idea that there is one big laptop, containing one big quantum state of information about the whole works.
There is no "manyworlds" stuff here, no "branching". Just a bunch of observers sharing the same reality to the extent that they can communicate and compare notes.

 

[Fra]

The 2011 paper and the 2006 paper may not be saying the same thing but I gather that they are comparable, or on the same wavelength. If you see them saying radically different things, please tell me and explain how they differ!

Yes they aren't saying exactly hte same of course but I think tehy are certainly making use of the same logic of reasoning. Perhaps one can try to pick on details and find disagreements, but I think the similarities where more interesting.

As I recall from reading the 2006 Smerlak Rovelli, reality is viewed as a consistent set of observers, being able to agree on events to the extent that they can communicate about them.

disagreements about what has occurred can be ironed out by communication, which however takes time.

Seeming paradoxes or logical contradictions only arise if one ignores the slow pace at which real observers can actually communicate. Or if one pretends to be an instantaneous super-observer overlooking the whole show.

To the extent that they can be communicated and negotiated is what I think is a key point.

This is why I think it's more accurate in this spirit to declare the GR lesson as DEMOCRACY between observers. Equivalence somehow implicitly contains an assumption that there perfect consistency is attained. It does ignore the time and equilibration process you mentions. It's in these ponderings my objections to Rovelli and his LQG thinking enters.

Even worse does this get, if you take the "encoded information" that psi means seriously. Then you need to consider how matter encodes the historical data in some forms, and how this is constrained by information capacity bounds, and how this furthermore associate to mass which is sourcing gravity; here the coding structure of psi, are sourcing gravity, and somehow Rovelli seeks equivalence classes of observers, to apply QM. This is where I think something is misapplied. At least I don't see the solution. But maybe I'm too stupid.

I personally think it's because Rovelli tries to be consevative. Even if he wanted to overcome these things, how would you do it? It's not easy that's for sure. But this type of reasoning that they are already doing may be the way of analysis that leads the way.

/Fredrik

 

[marcus]

I've been reading the 2011 Freidel et al paper. It is beautiful. It constructs a connection between observers----between the local phase spaces of two different observers who are separated spatially. What I want to say is that 1101.0931 goes way beyond analytically.

They show how proven and familiar analytical tools of physics, instead of being applied to something that may not exist (a unique monolithic spacetime colored with matter fields) can still be applied to what is more likely to exist: the separate colored phase-spaces of all the individual observers---and one even has a connection to move from one observer's perspective to another.

This in principle allows one to MEASURE (though torsion and curvature of this connection) by how much spacetime fails to be monolithic! So it becomes an empirical proposition to say that reality is not coherent in a universally consistent way. One can test by how much it deviates. This is what Freidel et al gives us in 1101.0931, and will give us in the three papers they say are "in preparation" that they cite.

And if, when tested, we learn that the deviation is zero? Well then a monolithic picture of reality is possible after all, one NOT relative to the observer! So then we can go back to the old way. The point is that their analytical tools make the matter measurable and empirical, at least in principle.

This is why I think 1101.0931 deserves study. It is quite a fine paper. IMHO.

I'm glad you like it, Fra. Also I will give the link in case anyone else is reading the thread and wants to take a look:
http://arxiv.org/abs/1101.0931

 

[Fra]

They show how proven and familiar analytical tools of physics, instead of being applied to something that may not exist (a unique monolithic spacetime colored with matter fields) can still be applied to what is more likely to exist: the separate colored phase-spaces of all the individual observers---and one even has a connection to move from one observer's perspective to another.

As one get into the proposed solutions to this lack of absoluteness, I start to see things differently.

The disagreement is subtle and I think it's easy to mix it up. I wouldn't say that spacetime doesn't exist at all just be because it's observer dependent. Similarly I would not assume that the observer dependence is always restored by objective transformations laws. I think even these laws are subject to the same inference by observer that they acknowledge.

When people use the notion of "relativity" it almost always contains the assuption that subjectivity is removed by considering equivalence classes of these things, that are furthermore objective and comes with connections. I think that this is possible, and if possible, makes things easier. But I think it's going too fast, as the same logic that suggests that structures are a result of observer inferences, suggests that even such grand ideas are also subject to this.

The final conclusion I personally make is that two observers in fact can't even agree completely on the theory space, except in special cases which probably corresponds to equilibirum points that takes a lot of time. But the fixed superstructure misses the intermediate physics. So a picture which truly involves evolving law is the conclusion I make.

With that disclaimer I full share the idea that observer depedent notions, encode interactions when the observers communicate. But this interaction needs to be observer from yet a third observer.

If you like, we can similarly consider that any law of physics are "relative" to the obsever, because the observer-inference of couplings depends on the observer. RG and the experience from particle physics partially supports this. It's not just about perturbation theory, it really says thay the "observable" laws of physics are observer dependent. This IMPLES when you consider one observer, observing two other observers that interact by communication and "information updates" that the inferred regularities in this interaction depends on the third observers perspective.

The reason this modle worked os nice for particle physics is becaues the laboratory frame is orders of magnitures in domination wrt encoding and ocmputing power as compared to the "observers" participating in our experiments, ie. electrons, quarks etc.

It seems to me that not many people seems to share this objection

Here I use the word observer just like Rovelli (ie meaning any subsystem, say a proton, quark or molecule).

If we have three observers, one atom, one proton and one human. The scale assymmetry here is what allows us to get away with these sins, but if we consider cosmological models, or try to see unification, I think these sins blow up into problems. I haven't see Rovelli mention these things anywhere... except as notes where he "doesn't want to discuss meaning of probability", and he mentions "meaning of observer". So I think he should be aware of this, anything else would be surprising.

My point is just that, the first excellent stances that things are observer dependent, can't be solved just be considering the "set of all observers" and then find transformations rules that restore objectivity. Because that supertheory is subject to the same constriants, so all we get is a perturbative expansion in some unkonwn theory space, where the space itself is also expanded. At some point, whene the observer assymmetries no longer hold the complexity of the constriction is too large to be encoded.

All I want, is that these problems aren't ignored. I don't expect instant solutions.

Edit: Though, the above things does relate to the points Marcus raise about "equilibration times" etc... Marcus, do you have any refefences where Rovelli (or any of this similar thinkers) discussed these details? Ie. equilibration times etc... I would love to read that and see if he does indeed address any of this, or just mentions them as problems to be solved for the future! I would love to see that he did address this. So far i haven't see it, but then I'm not near as updated on rovelli's all papers as you are.

/Fredrik

 

[marcus]

...
Edit: Though, the above things does relate to the points Marcus raise about "equilibration times" etc... Marcus, do you have any refefences where Rovelli (or any of this similar thinkers) discussed these details? Ie. equilibration times etc... I would love to read that and see if he does indeed address any of this, or just mentions them as problems to be solved for the future! I would love to see that he did address this. So far i haven't see it, but then I'm not near as updated on rovelli's all papers as you are.
...

Fra, in this particular department I think you know the literature at least as well as I do.
My impressions are based almost entirely on the 2006 Smerlak Rovelli paper, and my (imperfect) memory of it.

I think I understand what you mean by "equilibration times" but I did not use that terminology, which sounds a bit technical, like thermodynamics. I just remember, as you must too, from the 2006 paper that part of way the EPR paradox was resolved was by pointing out that the two observers cannot communicate instantaneously. The passage of time somehow buffers them and prevents a head-to-head conflict about reality. This is unfortunately quite vague. My memory of the paper is not clear. I should review it.

Or perhaps you will review the 2006 paper and correct my vague impression with a clear summary.

But for now I just want to read the Freidel et al 2011 paper and get used to their message. It is radical, but on the other hand they have invented ways to compute stuff. This connection gamma is very interesting. Shuttling from one observer's phase space to another's.
I should get the Arivero greek alphabet so that we can say Gamma in good style and comfort ;-D
αβγδεζηθικλμνξοπρσςτυφχψωΓΔΘΛΞΠΣΦΨΩ∏∑∫∂√±←↓→↑↔~≈≠≡ ≤≥½∞(⇐⇑⇒⇓⇔∴∃ℝℤℕℂ⋅)

 

[atyy]

What an interesting paper! This is a very careful consideration of where DSR might go after Hossenfelder's objections.

Since Freidel and Smolin are coauthors, I hope it will also affect eventually how classical spacetime is interpreted in the spinfoam/GFT formalism.

 

[atyy]

Hmmm, does this really address Hossenfelder's objection?

Could they possibly be claiming that present experiments do not place as tight a bound on locality as she claims?

The spacetime that emerges still seems very conservative, not much different from the old DSR picture.

 

[atyy]

Alice and Bob again! They are right up there with Michelson and Morley

 

[MTd2]

I really do not see any resemblance from DSR picture. What I see it is that the propose that myopia is a principal feature of nature.

 

[atyy]

For comparison:

DSR review, the last few slides are extremely interesting, they acknowledge Hossenfelder's work, and point to a possible solution, which at that point in time could not be made consistent.
http://pirsa.org/10090070

Hossenfelder objection to a particular interpretation of DSR
http://arxiv.org/abs/1004.0418
http://arxiv.org/abs/0912.0090

Current paper
http://arxiv.org/abs/1101.0931

Edit: Adding a quote from the current paper (p10) that shows it is related to DSR
"Last but not least, momentum spaces of constant curvature find its natural application as a model of Doubly (or deformed) Special Relativity (DSR) [28–31], whose second observer independent scale is naturally associated with the curvature of the momentum space [30]... For a long time it has been suspected that these were different ways of formulating the same theories, the developments of this paper show how an energy dependent metric and non-commutative spacetime coordinates are different ways of expressing a deeper idea, which is relative locality. ... The issue of possible macroscopic non-local effects in DSR was raised several times [33–37]. Attempts to address these issues led to some partial anticipations of the ideas proposed here [38–40]. From the present perspective we can appreciate that the concern was well justified and there is a pot of gold under the rainbow of apparent non-locality."

 

[marcus]

...
The spacetime that emerges still seems very conservative, not much different from the old DSR picture.

The spacetime of DSR was flat, so could never be more than a crude simplification, like Euclidean R⁴ or like Minkowski space of special relativity.

DSR did not, as far as I know, provide something that competed with General Relativity, where you can have different neighborhood geometry around different observers. DSR made no attempt to be realistic in that sense.

So there must be substantial difference between this new Rel-Loc setup and DSR.

There is this similarity though: I recall one way of formulating DSR had a curved momentum space, with necessarily unusual rules for combining momenta.

In Rel-Loc world, each observer has his own local momentum space geometry. And in Rel-Loc world there can be a connection, to allow us correspondence between My local phase space (my momentum x position space) and Your neighborhood of phase space. Just my take on it. corrections welcome.

I'm very glad you are reading the Rel-Loc paper, Atyy. Something came up here and I have not been able to study it so far today, but I hope to get back to it this evening.
The connection on phase space (which they write down) may itself have curvature and torsion. A long distance transport may mix the phase space coordinates---mixing position and momentum coordinates. The possibility was mentioned in the paper. It might be possible to constrain some of these things by observation---and maybe find out that a single consistent overall geometry exists after all. But this is not a given.

 

[MTd2]

For comparison:

DSR review, the last few slides are extremely interesting, they acknowledge Hossenfelder's work, and point to a possible solution, which at that point in time could not be made consistent.
http://pirsa.org/10090070

Hossenfelder objection to a particular interpretation of DSR
http://arxiv.org/abs/1004.0418
http://arxiv.org/abs/0912.0090

Current paper
http://arxiv.org/abs/1101.0931

Those criticism do not apply to this paper since there is no dependence on frequency for the speed of light, although there is for far away events. But it doesn't really matter in the former case because the farther the even the farther it is blured.

 

[marcus]

The connection on phase space, that I mentioned at the end of post #19, strikes me as perhaps the most interesting thing in the paper.
They say that "spacetimes" are just "observer-dependent slices" of phase space.
There is no one official "spacetime". Physics happens in phase space, in their view, not in some one observer's version of spacetime.

Or at least this is an empirically testable possibility. They argue there are experimental/observational ways of constraining the connection. And that it is the curvature and torsion of this connection which prevent having one unique official observer-independent version of spacetime. If I misinterpret here please offer correction

Reference [2] is to a paper in prep by Laurent Freidel. I believe he goes into detail about the algebra of combining momenta, and how the connection on phase space derives from this algebra, and then discusses things like the curvature and torsion of that connection.

I hope the paper [2] is available before 1 March, when the seminar talk is scheduled. It will be good to have looked at it prior to the talk.

 

[Max™]

Very interesting paper, I love how they produced their results.

I've long pondered variations of realism as a means to resolve EPR, should be interesting to see where this line of research goes.

 

[marcus]

Very interesting paper, I love how they produced their results.

I've long pondered variations of realism as a means to resolve EPR, should be interesting to see where this line of research goes.

I agree, the results seem quite novel and interesting, with new insight shown their way of obtaining them. As I re-read the Freidel et al Rel-Loc paper I began to think that I was mistaken to compare it with the 2006 Smerlak Rovelli "Relational EPR" paper.

That older paper does indeed explore a "variation of realism as a means to resolve EPR" and would probably be of considerable interest to you, especially if you haven't yet looked at it.

But about the new Freidel et al paper, I'm feeling the need to backtrack and reconsider.
How would you describe what they do?

I now see it as what you referred to as a variation of realism, but not a relaxation (as in the 2006 paper explicit departure from "strict" Einstein realism.)

Correct me if you think I'm wrong, but I see Freidel et al as having a strict full monolithic realism, but just a different kind. There is a single comprehensive phase space, I suppose 8 dimensional, coding position (displacement from an observer) and momentum. It really exists and physics happens in it. At least in the classical version of the story, in the quantum version maybe what exists is Hilbertspace and we live in that and physics happens there. I don't know what they would say.

They don't have to introduce any ad hoc philosophy about different observers having slightly different reality because of something lucky that happens in the mathematics. If the momentum space is curved, then the connection that transports from one observer to another just automatically turns out to be non-trivial. So they get a kind of fragmentation of reality without asking for it.

A kind of apparent atomization of reality just happens out of the math. Each observer will establish a coordinate patch around himself and he will SPLIT the surrounding phase space, as seems natural to him, into 4D position and 4D momentumspace.

Now what Freidel et al refer to as the "pot of gold" they find under the (mirage?) "rainbow" of DSR is the discovery that the connection is non-trivial and that there is no universal projection down onto any absolute spacetime. The bundle is not a simple cartesian product. the connection MIXES THE COORDINATES as it transports from one observer to another. Because momentum space is curved. That is the great thing about this paper.

So reality is real, but, happily enough, it is screwed up. It gets screwier and screwier as you get farther and farther from home. The world has the illusion of being atomized, the way we feel intuitively it should be. The frog of paradox has been kissed and has become a gallant prince or gracious princess.

So Max, a propos your reference to EPR, this is the story that I got from Freidel et al last night. It is different from the 2006 Smerlak Rovelli EPR paper. There might still be some subterranean passage between them, but this is how it looked on the surface.

 

[MTd2]

I found some contradictions on the conclusion the paper.

First p.9-10, they sugest that quantum mechanics comes from phase space

"one can suggest that the fundamental description is one in which there
is dynamical curvature in phase space, so that the fundamental constants are GNewton and mp. This implies that Planck’s constant could be a derived quantity, h = G^2, suggesting that quantum mechanics is emergent from a dynamics of phase space geometry."

Later, they sugest that phase space comes from quantum mechanics:

p.10

"We have seen here that the notion of curved momentum space has generic and vivid consequences for our understanding of basic physics. We do not live in spacetime. We live in Hilbert space, and the classical approximation to that is that we live in phase space."

So, what comes first, phase space or quantum mechanics?

 

[marcus]

So, what comes first, phase space or quantum mechanics?

What comes first (their paper suggests to me) is whatever underlies both conventional QM and phase space dynamics.

They gave a reason why conventional QM MIGHT be emergent from whatever mathematical structure underlies phase space dynamics. They did not speculate as to the mathematical form of such. They only treated the classical phase space dynamics.

they also said, at the end, "we don't really live in phase space, we live in Hilbertspace". This is a loose carefree way of suggesting that we live in whatever underlies both conventional QM and phasespace dynamics. they only wanted to say that there must be something deeper than the classical picture they offer.

You will probably have to wait until you can read Freidel's in prep called "The Geometry of Phase Space".

 

[Max™]

I'll have to check over the other paper later, kinda noisy here right now, too much so to concentrate.

 

[atyy]

From footnote 11 on p10 of the current paper "While finishing this paper we learned that our construction is related to the mathematical theory of loops developed, among others, by M. Kikkawa and L.V. Sabinin, [45]. We thank F. Girelli and E. Livine for pointing this out to us."

http://arxiv.org/abs/1009.4762!

 

[marcus]

From footnote 11 on p10 of the current paper "While finishing this paper we learned that our construction is related to the mathematical theory of loops developed, among others, by M. Kikkawa and L.V. Sabinin, [45]. We thank F. Girelli and E. Livine for pointing this out to us."

http://arxiv.org/abs/1009.4762!

Thanks for catching that footnote, Atyy! It's interesting to know of related mathematical constructions. There seem to be two papers involved, one by Kikkawa(1975) and one by Sabinin(1999).

I could find neither of those two on line, so I had a look at your Girelli find:
http://arxiv.org/abs/1009.4762 (!)

 

[atyy]

This paper in the OP and the Girelli one above are trying to follow this line of thought:

The famous Freidel and Livine http://arxiv.org/abs/hep-th/0512113

Which Girelli and Livine followed up in
http://arxiv.org/abs/0903.3475
http://arxiv.org/abs/1001.2919
http://arxiv.org/abs/1004.0621 (which explicitly mentions curved momentum space)

Also of interest is their earlier work http://arxiv.org/abs/gr-qc/0406100 , which also mentions curved momentum space.

 

[MTd2]

Thanks for catching that footnote, Atyy! It's interesting to know of related mathematical constructions. There seem to be two papers involved, one by Kikkawa(1975) and one by Sabinin(1999).

I could find neither of those two on line, so I had a look at your Girelli find:
http://arxiv.org/abs/1009.4762 (!)

Kikkawa(1975)

http://www.projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.hmj/1206136626

The other one, which is a book, is not available. But there is a review article of the subject by the author:

http://www-sbras.nsc.ru/EMIS/journals/CMUC/pdf/cmuc0002/sabinin.pdf

 

[Fra]

What comes first (their paper suggests to me) is whatever underlies both conventional QM and phase space dynamics.

They gave a reason why conventional QM MIGHT be emergent from whatever mathematical structure underlies phase space dynamics. They did not speculate as to the mathematical form of such. They only treated the classical phase space dynamics.

they also said, at the end, "we don't really live in phase space, we live in Hilbertspace". This is a loose carefree way of suggesting that we live in whatever underlies both conventional QM and phasespace dynamics. they only wanted to say that there must be something deeper than the classical picture they offer.

I have been busy doing carpentry work at home and will be the next week at least and haven't had time to spent more time analysing their papers but to just connect MTd2's question with my general reflections early on:

To speak for my own expectations of a possible extenstion of this logic (without knowing that the authors views are) I definitely think that a possible extension of this thinking can yield QM as result. I commented on this before but in a different context but one possible idea is that one it's established that there exists information coded in non-commutative structures (Q and P for example) then computation of expectations containing conditions suchs as A and B or A or B where they aren't commutative, can be computet first using a sort of "parallelltransport" Add that with limited information capacity and the result becomes "lossy" transports. Somehow the connections are lossy transformations and the result is possible something like quantum logic. So I think it's highly realistic scenario that this can help explain QM. But in that view, the logic can be applied to ANY memory record consisting of non-commuting variables. The real question is which sets that have special meaning in nature.

I see many interesting possible developments here. I realize that there is a need to make the above explicit in a paper. I'm not aware of anyone that has, and I've postpone anything since I want to make far more progress before writing anything myself as bits an pieces are more likely to just be misunderstood without a little bit more compelte picture.

/Fredrik

 

[marcus]

Slides PDF available now at the ILQGS (International Lqg Seminar) http://relativity.phys.lsu.edu/ilqgs/
for Laurent Freidel's talk
on The principle of relative locality
http://relativity.phys.lsu.edu/ilqgs/freidel030111.pdf

Audio files are not available as yet. The talk was scheduled for Tuesday 1 March.
If things went as planned, audio will be posted online within a day or two.

 

[Fra]

I almost forgot about this thread since the last post.

Don't remember if I wrote this already but it seems to me that while there are irresistable ingredient in this, I have the some objection as I have to rovelli's RQM.

Is the the spacetime manifold directly measured by an inside observer? or is it just infered, in an observer dependent way? I think it's hard to disagree with Freidel here. I think they are right, but I think more right than they seem to think themselves.

But the very same argument applies to his phase-space, does it not? This is almost exactly analogous to when Rovelli simply assumes that observer-observer communication obey the known QM structure? Ie. Rovelli assumes that the observer relative inferences does fit together in a bigger invariant structure - this may be neat, but it seems unwarranted and unphysical, and in particular it does not seem to be an inference. How come?

This is what is incoherent about this.

Seriously, does anything this this objection is unfair or irrelevant, then can you explain why.

Note: What I am trying to say here is not that their argument is wrong, it's just that just like with rovelli, they aren't radical enough. If their general argument is correct, I simply don't see why it's applied only once, and in a particular way that looks like it's constructed? Again, no too unlike rovelli, who has constructed an interpretation to suit the LQG picture (or so is my impression).

Since some of the arguments are so good, it's annoying that it seems to not go all the way. Why? I don't get it. It's almost like they don't quite take the inference argument serious themselves??

/Fredrik

 

[marcus]

The audio for Freidel's talk is now available.
http://relativity.phys.lsu.edu/ilqgs/

It is a good talk. The ILQG audience was smaller than usual (many of the regulars were at the QG school) and only two people (Alex Corichi was the first one) asked questions at the end. For me the talk added a lot to the paper that we saw, making the ideas of Rel. Loc. easier to understand.

At this point, Fra, it would seem that the approach is phenomenological. They have a plausible approximate model of curved-momentum-space physics with hbar and NewtonG both zero.

and the squareroot of their ratio (ie. Planck mass) is non-zero which is what makes momentumspace curved. So then they have many predictions from the model, that one can test.

Indeed the predictions seem to me strange and outlandish---so that the Rel. Loc. model would seem not only falsifiable but quite possibly wrong! But I don't want to pre-judge. The problem now is not to make the theory better but to do the experiments to see if momentum space is actually curved or, as we used to assume, flat.

 

[Fra]

The problem now is not to make the theory better but to do the experiments to see if momentum space is actually curved or, as we used to assume, flat.

That is a reasonabler way of seeing it for those betting on the specific implementation of the theory. ie. it's plausible enough to motivate invest in experimental feedback before revising the idea more. Or put differently: It's investing in putting the theory to test; has higher odds of gain than investing in polishing it further before putting it to test.

But since I'm not personally going to do any experiments or investments, I only judge the plausability of the approach and reasoning.

They have a plausible approximate model of curved-momentum-space physics with hbar and NewtonG both zero.

I think this is what I tried to question.

I mean, how plausible is it really? Their original spirit is really plausible, but I don't find it plausible how they twist it in the end. So I suppose my personal opinon is that it's better to invest in developing the idea more (answering to some of the critiques I suggested) before trying to kill it. I think it won't live long enough. If we can smell this righ now, what is the rationale in ignoring that?

/Fredrik

 

[atyy]

http://pirsa.org/11020116

 

[John86]

This new Freidel paper (lLocal Relativity) is just posted yesterday

Gamma ray burst delay times probe the geometry of momentum space
Authors: Laurent Freidel, Lee Smolin
(Submitted on 29 Mar 2011)

Abstract: We study the application of the recently proposed framework of relative locality to the problem of energy dependent delays of arrival times of photons that are produced simultaneously in distant events such as gamma ray bursts. Within this framework, possible modifications of special relativity are coded in the geometry of momentum space. The metric of momentum space codes modifications in the energy momentum relation, while the connection on momentum space describes possible non-linear modifications in the laws of conservation of energy and momentum. In this paper, we study effects of first order in the inverse Planck scale, which are coded in the torsion and non-metricity of momentum space. We find that time delays of order Distance * Energies/m_p are coded in the non-metricity of momentum space. Current experimental bounds on such time delays hence bound the components of this tensor of order 1/m_p. We also find a new effect, whereby photons from distant sources can appear to arrive from angles slightly off the direction to the sources, which we call gravitational lensing. This is found to be coded into the torsion of momentum space.

 

[John86]

I really enjoyed this paper by Freidel and Smolin. But not being a physicist really makes you feel running into conceptual walls. Can anybody help tme to make any sense out of the concept of "Phase Space" that is often mentioned in this new paper ?...

Wiki really didn,t help me in a positive way !..

 

[marcus]

I really enjoyed this paper by Freidel and Smolin. But not being a physicist really makes you feel running into conceptual walls. Can anybody help tme to make any sense out of the concept of "Phase Space" that is often mentioned in this new paper ?...

Wiki really didn,t help me in a positive way !..

This really should be answered, and we have company coming (I'm already an hour late)
Someone else should reply.

Phase space is a classical idea, probably 19th century.

It's simple, for one particle you take the position and momentum, conceptually R³xR³. For N particles you take an N-fold cartesian product of that. So you are keeping track of all the particles position and momentum, in the observer's immedciate neighborhood.

The nontrivial angle to this is the possibility that MOMENTUMSPACE ACTS LIKE IT IS CURVED instead of flat. That gives a whole new interest to the idea of phase space. Momenta don;t add in a simple vectorspace way any more.

the tendency in Freidel's papers is to focus on just one copy of momentum space and just one particle, because that is where the interest is. And so it all boils down to that. You forget about N particles and N-fold cartesian products for the time being.

If somebody doesn't like this simplified answer, please give a better!

Did you try Wippy?
http://en.wikipedia.org/wiki/Phase_space

 

[atyy]

When the dynamics can be cast in Hamiltonian form, phase space is the space of the fundamental degrees of freedom. Ashtekar's variables were a change of variables in general relativity so that the theory had Hamiltonian form. The ADM formulation of general relativity is yet another Hamiltonian formulation of GR.

http://www.scholarpedia.org/article/Hamiltonian_systems
http://arxiv.org/abs/1007.0402
http://grwiki.physics.ncsu.edu/wiki/Hamiltonian_(ADM)

I guess the rough idea of relative locality is that the phase space variables are usually [position, momentum]. But here the variables are [x,momentum] where x is not position. Then position is reconstructed from [x,momentum]. The reason is that some LQG related formalisms give [x,momentum] with momentum curved. But Hossenfelder showed that such a theory is already likely falsified if x is position. So the question was, can we interpret x as something else, and still recover spacetime, at least locally.

 

[atyy]

The last reference in the relative locality paper http://arxiv.org/abs/1101.0931 is, rather amusingly, http://arxiv.org/abs/1007.2200 . LQG's destiny is condensed matter, just like string theory

 

[John86]

Thank you both

 

[John86]

A new paper by Smolin

http://arxiv.org/abs/1104.2822

A real ensemble interpretation of quantum mechanics
Lee Smolin
(Submitted on 14 Apr 2011)
A new ensemble interpretation of quantum mechanics is proposed according to which the ensemble associated to a quantum state really exists: it is the ensemble of all the systems in the same quantum state in the universe. Individual systems within the ensemble have microscopic states, described by beables. The probabilities of quantum theory turn out to be just ordinary relative frequencies probabilities in these ensembles. Laws for the evolution of the beables of individual systems are given such that their ensemble relative frequencies evolve in a way that reproduces the predictions of quantum mechanics. These laws are highly non-local and involve a new kind of interaction between the members of an ensemble that define a quantum state. These include a stochastic process by which individual systems copy the beables of other systems in the ensembles of which they are a member. The probabilities for these copy processes do not depend on where the systems are in space, but do depend on the distribution of beables in the ensemble. Macroscopic systems then are distinguished by being large and complex enough that they have no copies in the universe. They then cannot evolve by the copy law, and hence do not evolve stochastically according to quantum dynamics. This implies novel departures from quantum mechanics for systems in quantum states that can be expected to have few copies in the universe. At the same time, we are able to argue that the centre of masses of large macroscopic systems do satisfy Newton's laws.

 

[marcus]

Jeff Morton (Baez PhD 2007) has a blog. He recently commented on the Principle of Relative Locality, and a presentation of it by Laurent Freidel.
http://theoreticalatlas.wordpress.com/2011/03/

I found this helpful. Morton has acquired the knack of explaining things in a relaxed gradual way without unnecessary technicality. (Math jargon used selectively and intelligently---avoided when some other description will do as well.)
Maybe he is not as good at this as his thesis advisor Baez, but I didn't see that Baez covered Freidel's recent papers and talk about Rel Loc. So I'll take Morton's discussion of it.

Morton will be participating at the mid-June Zurich conference "Quantum theory and Gravitation"
http://www.conferences.itp.phys.ethz.ch/doku.php?id=qg11:participants
He may have some blog goodies on that, so I'll try to remember to look.

Here's a self-introduction page at Morton's site:
http://www.theoreticalatlas.net/

 

[atyy]

marcus just posted https://www.physicsforums.com/showpost.php?p=3320858&postcount=1486 in his bibliography.

Now, we know Freidel and company are interested in relative locality because of the non-commutative fields which seem to come out of some spin foams. But doesn't the philosophy behind a different (?) sort of non-commutativity seem very similar to to relative locality's? Compare http://arxiv.org/abs/1008.0985 and http://arxiv.org/abs/1101.0931

However, relative locality is not fundamental (assuming it exists), since Freidel et al explain that it is only a certain limit, and the underlying theory is presumably string theory or whatever you favourite one is. OTOH, I guess Chamseddine and Connes are thinking that the spectral action is fundamental (assuming it can be quantized).

 

[marcus]

...
However, relative locality is not fundamental...

?
This sounds strange.
I don't recall Freidel et al ever saying that the Rel Loc principle is "not fundamental".

You can implement the principle in a flat space limit in order to have something to test. There would be various possible theories satisfying/implementing it. But that doesn't prevent Rel Loc being proposed as a fundamental physical principle.

I suspect, in fact, that Rel Loc IS being proposed as a fundamental physical principle. That would make sense AFAICS. I can't say definitely--haven't read all that much about it.

 

[atyy]

?
This sounds strange.
I don't recall Freidel et al ever saying that the Rel Loc principle is "not fundamental".

You can implement the principle in a flat space limit in order to have something to test. There would be various possible theories satisfying/implementing it. But that doesn't prevent Rel Loc being proposed as a fundamental physical principle.

I suspect, in fact, that Rel Loc IS being proposed as a fundamental physical principle. That would make sense AFAICS. I can't say definitely--haven't read all that much about it.

Last paragraph of p1 and the first paragraph of p2. http://arxiv.org/abs/1101.0931

 

[marcus]

I may not be remembering Rel Loc correctly, it came out some weeks back and I've been concerned with other things in the meanwhile.
As I remember, it says there is no space time shared by all observers.

A given observer can construct for himself a model of space time and say what worldlines crossed and what events occurred. But a distant observer may construct a different model that will not coincide-- with different events.

The curvature of momentum space that this incompatibility depends on is testable. It can be falsified. It might be a good thing if it were. The Rel Loc idea is quite radical and disturbing. It essentially says that no collective space time exists, that we can all share.

I think Rel Loc would be bad news for AdS/CFT and everything else we are used to. No space time. No bulk. No boundary. Many physical theories would be disrupted.

In GR different observers have different time and see simultaneity differently, but at least they all share the same space time----they just slice it up differently into spatial slices. The Rel Loc principle goes beyond that. They no longer share the same model of space time. Their spacetimes are inconsistent. Rel Loc is worse than GR---in the sense of unpalatable and disruptive. IMHO.

Maybe there is some more general phase space that all observers can share, that we can say exists, and that can somehow be "sliced" differently by different observers in order to realize their separate spacetime models. Maybe you can find some mathematical construct of reality which is shared in common. I do not recall seeing such a thing when I examined the papers.

 

[marcus]

Last paragraph of p1 and the first paragraph of p2. http://arxiv.org/abs/1101.0931

I know that! That is my point. They implement the principle in an approximation, as they say. That gives a testable realization. I think I talked about that earlier.

Surely the Rel Loc principle is not confined to that approximate picture!

 

[atyy]

I know that! That is my point. They implement the principle in an approximation, as they say. That gives a testable realization. I think I talked about that earlier.

Surely the Rel Loc principle is not confined to that approximate picture!

It should be. The more fundamental picture is the spin foam picture, where we may not have locality, or smooth geometry even in momentum space.

Or have we given up LQG?

 

[John86]

"new scientist" talks about phase space proposal and principal of relative locality !..

http://www.newscientist.com/article/mg21128241.700-beyond-spacetime-welcome-to-phase-space.html