Verlinde, LQG, entropy and gravity as a fundamental force vs emergent

[Fra]

Maybe it is not gravity that needs to be quantized, but geometry.

Or information? :) then perhaps all interactions can be identified as emergent from classifictions of information exchange, and the quantization is already there.

After all what is "quantization" in the first place? Either it's a set of heuristic rules, that admittedly are successful in the currently known areas, or there is a deeper picture.

I think of "quantization" as requireing that the _information_ about the system in question is to be treated as an inferrable by an actual observing system with real constraints from an actual interaction history. Ie. be measurement processes.

I think this measurement ideal, or prior to things like "making into chunks", which is rather an implication of the former ideas.

So to "quantize" gravity or geometry don't we need to understand what it actually means to make such measurements? It in this context, things I can hep but think thta ideas that consider ensenbles of universes on part with ensembles of particles colliding to be the completely wrong idea.

If we just think of "quantization" as a set of mathemataicl moves in absurdum, then I think we don't know what we are doing.

This borderlines to philosophical aspects but I think physicists (not just oure philosophers with lack of physics knowledge) ought to take this more seriously that is done.

/Fredrik

 

[tom.stoer]

Maybe it is not gravity that needs to be quantized, but geometry.

What's the difference?

========= EDIT =========

If you look at LQG, the dynamics is encoded in the Hamiltonian H, whereas the geometry is encoded in the structure of the spin network Hilbert space. The structure of the Hilbert space is forced by the quantum constraints G and D, i.e. by gauge symmetry generated by the Gauss law G and by diffeomorphism invariance. The structure of the Hilbert space is well-understood, whereas the dynamics is notoriously difficult (ordering and regularization ambiguities; off-shell constraint algebra, complex term in H multiplied by the Immirzi parameter,...) so it would be really nice to get rid of it and quantize G and D only.

But that seems to be impossible due to the construction of the constraint algebra. It is simply not allowed to drop H because it is reflects the fourth (time) dimension; w/o H the theory is incomplete.

In the LQG framework I would say that gravity is the dynamics of geometry. I see what it could mean to quantize geometry "only", but I think it's impossible to separate geometry and gravity.

 

[ensabah6]

What's the difference?

========= EDIT =========

If you look at LQG, the dynamics is encoded in the Hamiltonian H, whereas the geometry is encoded in the structure of the spin network Hilbert space. The structure of the Hilbert space is forced by the quantum constraints G and D, i.e. by gauge symmetry generated by the Gauss law G and by diffeomorphism invariance. The structure of the Hilbert space is well-understood, whereas the dynamics is notoriously difficult (ordering and regularization ambiguities; off-shell constraint algebra, complex term in H multiplied by the Immirzi parameter,...) so it would be really nice to get rid of it and quantize G and D only.

But that seems to be impossible due to the construction of the constraint algebra. It is simply not allowed to drop H because it is reflects the fourth (time) dimension; w/o H the theory is incomplete.

In the LQG framework I would say that gravity is the dynamics of geometry. I see what it could mean to quantize geometry "only", but I think it's impossible to separate geometry and gravity.

Progression on spinfoam seems to be better -- dynamics seem better understood. How do SF get around the above issues?

 

[tom.stoer]

Where do you think that spin foams are better (regarding the dynamics). I do not know if anybody has addressed the issues related to H in the spin foam aproach. I missed these topics in Thiemann's papers ...

 

[MTd2]

In the LQG framework I would say that gravity is the dynamics of geometry. I see what it could mean to quantize geometry "only", but I think it's impossible to separate geometry and gravity.

Gravity makes me think about mass. But mass is not an issue here, but entropy. Entropy is not quantized, but is a consequence of quantization. So, the only thing left here to quantize is geometry and the fields that lives on it.

 

[tom.stoer]

Gravity makes me think about mass. But mass is not an issue here, but entropy. Entropy is not quantized, but is a consequence of quantization. So, the only thing left here to quantize is geometry and the fields that lives on it.

(quasi-local) mass is a very difficult and derived concept in GR ...

 

[MTd2]

I don't get what you are saying.

 

[tom.stoer]

quasi-local mass is a derived concept in GR that is not completely understood; you don't need it for the dynamics of gravity; all you need is energy-momentum density

 

[MTd2]

But energy tensor ends up involving point masses anyway, which will end up in geometry controlled by mass terms.

 

[tom.stoer]

But energy tensor ends up involving point masses anyway, which will end up in geometry controlled by mass terms.

No, it doesn't. Mass means integrating a voluem form which is tricky in curved spacetime. Forget about mass. It's not an energy tensor, it's an energy-momentumdensity tensor.

 

[MTd2]

Density is still a mass term.

 

[tom.stoer]

Formally the Einstein equations tell you that the action of an operator containing purely geometric degrees of freedom equates to something containing other (= "matter") degrees of freedom. If you want to separate geometry and gravity you have to make sense of "quantized geometry" alone = w/o taking into account the dynamics of the matter degrees of freedom = w/o taking into account dynamics. But I don't know how to separate geometry from dynamics b/c both have the same origin; that's one lesson of GR.

 

[MTd2]

Hmm, I guess what I am arguing here is the nature of matter after all. The concept is so weird that I got confused.