Holographic Dualities and the Potential Misdirection of String Theory

[tom.stoer]

Nice turn!

This is indeed how many people today think about holography. The volume degrees of freedom of a certain system are represented by a boundary Hilbert space living on the surface. These boundaries need not be physical; they can be introduced rather artificially, e.g. as boundary between a "quantum system" and an "observer". Then the Hilbert space structure encodes naturally what we (= the observer) can know about the system, as we (the observer) have defined what this system really _is_ - namely in drawing the boundary.

 

[inflector]

I don't think so. Up to now there is reason to believe that any fundamental principle is wrong. It's just that QG is currently neither verifiable nor falsifiable in the QG domain.

As long as there are no hints that something goes wrong experimentally, I can't see any reason why we shouldn't use these well-established principles.

I understand your position, and most scientists probably hold the same view that you do.

I see hints all over the place that "something is wrong experimentally.":

• Galaxy rotation curves and the need for dark matter which we can't yet detect
  
• The large-scale structure of the universe
  
• The pioneer anomaly
  
• The flyby anomaly
  
• Dark energy
  
• Ultra high energy cosmic rays (OMG particles)
  
• The recent detection of changes in radioactive emission tied to the sun and solar flares

These unexplained phenomena tell me that some part of current theory is wrong or incomplete.

Changing some basic principle doesn't make QG falsifiable / experimentally testable!

Not by itself it doesn't you are right. Just changing widely-held assumptions doesn't guarantee success or a testable theory. Nevertheless, a theory that changes one of our unproven but still widely held assumptions might have features that are unexpected but testable.

All of the above list of "hints" could potentially be explained by some sort of bizarre gravity scenario and this idea would be testable. For example, consider the pioneer anomaly, if we have observed effects using probes we have sent already using early 1970s technology, it stands to reason that we'd see the same effects if we sent new probes specifically to test a given theory if the new bizarre theory is valid.

Suppose this new theory of gravity predicts anisotropies in gravitational mass and that gravity is a statistical phenomena on the large scale but anisotropic on the level of an individual particle. Suppose, again that it is posited that this new theory can explain the pioneer anomaly, ultra-high-energy-cosmic rays, and the existence of dark matter. The ability to tie the predictions of this new theory to specific observational anomalies also would imply the ability to test it further through observation and experiment. Anisotropy in gravity is something we could test using current technology if that is what is behind the acceleration anomalies I listed above.

This is just one very hypothetical example of how discarding a widely held assumption, in this case gravitational isotropy, could lead to a testable theory.

This is admittedly, not the path most would take. Nevertheless, I do think it represents one approach that should be developed.

 

[Fra]

> Nice turn!

I agree :) now we're closing in.

> In fact, it should be possible to find areas where one could make a prediction which could be confirmed by experiment, one would think.

This is my conviction as well, but before we get there, the exact formulation and meaning is still unclear.

For example

> Then the Hilbert space structure encodes naturally what we (= the observer) can know about the system

One may ask, what DIFFERENCE does it make, what the observer knows? ;-) Unless we just thinkg of knowledg or "information" in a realist sense, we need to have some ideas to make sense of this.

The idea I have is the idea of rational action, which means that this has implications for how the observer ACTS onto it's environment. This can (with some work) possibly give plenty of predictions, that does not need black holes produced in labs.

/Fredirk

 

[negru]

If black hole entropy is not proportional to area, then QFT is wrong as well. You can do the computation when the curvature is not very big, and that's what you get. Higher order corrections are not going to change area to volume.

You are expecting way too many things to be wrong just in order to accommodate nothing in return.

 

[inflector]

If black hole entropy is not proportional to area, then QFT is wrong as well. You can do the computation when the curvature is not very big, and that's what you get. Higher order corrections are not going to change area to volume.

You are expecting way too many things to be wrong just in order to accommodate nothing in return.

Who and what specific assertion are you addressing?

 

[negru]

This

Finally, it could be true that:
Our theory of black holes is not correct, they don't exist as we currently conceive them
Entropy for the black hole replacement in the new theory scales as r^3

and also why AS can't be right.

 

[inflector]

Well, it is only the last of four items. I don't consider it the most likely that's why it is last.

It remains nevertheless a possibility. If someone were to propose a specific theory at some time in the future that had this trait then it would be incumbent upon the authors of that theory to address the issues with respect to QFT that you have shown.

As Tom pointed out earlier:

In GR a black hole is a curvature singularity with an event horizon hiding the singularity from the outside oberver. One can calculate its temperature and entropy using classical GR with a semiclassical approx. for QFT. In LQC the curvature singularity is replaced by a mathematically well-behaved region w/o any singularity, whereas in full LQG the event horizon enclosing the singularity seems to stay intact. So the interior which no longer contains a singularity is still hidden from the outside observer.

So we have different theories predicting different ideas about what a black hole is. My earlier point that this means that we can't necessarily rule out behaviors which are predicted by GR stands. Let any given theory of quantum gravity deal with black holes as it may. As long as it is internally consistent, deals with the correspondence principle and can explain more than GR and quantum mechanics alone, it will find some success.

As to AS? I don't know enough about it specifically to offer even a modest assessment.

The issue of whether or not gravity is emergent or fundamental is orthogonal to the issue of black hole entropy being a function of area. That was the real focus of this thread. Black hole entropy is why holographic dualities are of interest because they solve the problem in one plausible way.

There are other ways that can address black hole entropy without relying on holographic dualities and some of these ways include the idea that gravity is fundamental.

 

[tom.stoer]

So we have different theories predicting different ideas about what a black hole is.

No, not really.

We know tat GR breaks down at or near the singularity and has to be replaced by something which is mathematically consistent.

LQG seems to do the job, unfortunately it is too complicated to solve all equations and one has to use approximations. LQC is (in a certain sense - please don't mess up my argument with minor details) such an aproximation. It predicts the elimination of the singularity but is unfortunately not able to talk about entropy of the gravitational degrees of freedom because it reduces the theory to only a finite number of them - which means that it has nothjing to say about entropy.It's a kind of toy model for full LQG - but a rather interesting one as it discusses exactly the same limiting cases as the spherically symmetric Schwarzsschild black hole and the spherically symmetric spherically symmetric scenarios.

Full LQG is too complicated to discuss the full dynamics, but one is already able to derive some quite interesting results, namely entropy via nearly exact counting of states, and the quantum geometry of the horizon.

So we do not have three different theories, but theories which are related to each other by taking the classical limit (GR from LQG) and by symmetry reduction (LQC from LQG).

Please note that in LQG the horizon degrees of freedom emerge automatically and need not be introduced by hand. Therefore the holoraphic principle is somehow emergent as soon as one studies a boundary between too regions of spacetime.

 

[inflector]

Thanks for the clarification tom.

 

[atyy]

This


and also why AS can't be right.

I understand your argument about the scaling of entropy, and agree that it is unclear how AS will solve it. Nonetheless, it is not clear that AS cannot solve it.

What do you make of this statement http://golem.ph.utexas.edu/~distler/blog/archives/001585.html ?"In any case, the existence of a “quantum” conformal symmetry in quantum gravity is compatible with there being a nontrivial dimensionful scale in the theory, so I don’t see a-priori why it’s incompatible with black holes."

A second question is, is it clear in AdS/CFT, that if one runs the renormalization flow from high energies to low, that there isn't an IR fixed point in the full theory that would be a UV fixed point in gravity?

 

[tom.stoer]

For me the question is if AdS/CFT can be translated to the real world which is not AdS.