Is Black Hole Complementarity the Key to Understanding Quantum Mechanics?

[Finbar]

Hi,

I was wondering if anyone who really understands black hole complementarity can answer some questions for me.

I understand what happens from the view point of the observer outside a very large black hole.
But I'm concerned with the view point of the in falling observer. In particular does the black hole evaporate according to her? If not this suggests that the black holes mass will never decrease.

In the many worlds interpretation of QM it seems to suggest that by crossing the horizon she actually moves into a different spacetime history than the one observed by the observer outside the black hole? This seems to be the only viable option once you take the backreaction of the Hawking radiation into account?


This all suggests that when is comes to the quantum mechanics of black holes we can preform an experiment with two different observers and the expected result for each observer will be very different.

So do we need to modify the principles of standard QM in any way to accommodate for
black holes??

Thanks,

Finbar.

 

[mitchell porter]

I think the idea is that the interior of the black hole has a dual (holographic) description in terms of states on the horizon; a lot like AdS/CFT, with the horizon being the boundary to the interior. So when someone crosses the horizon from outside, there's a description which involves them continuing to fall inwards, until they are torn apart by tidal forces and their degrees of freedom redistributed among the black hole's degrees of freedom, all of which will later leak away via Hawking radiation; but there's another description in which, when you arrive at the horizon, your degrees of freedom get holographically smeared across it, once again mingling with all the black hole's prior degrees of freedom (also located on the horizon), which all eventually leak away as Hawking radiation.

Scott Aaronson has a vivid word-picture of http://www.scottaaronson.com/democritus/lec14.html" :

"it seems as if you have all these bits that are near the event horizon of the black hole... If you're standing outside a black hole, you never see someone pass through the event horizon. Then, if you want to preserve unitarity, and not have pure states evolve into mixed states when something gets dropped into a black hole, you say that when the black hole evaporates via Hawking radiation, then the bits get peeled off like scales, and go flying out into space."

Samir Mathur's fuzzball picture of a black hole suggests that a black hole is actually a stringy tangle that extends all the way out to the event horizon, rather than being packed into a singularity; and Matrix theory tells us that in a black hole, nonlocal "off-diagonal" degrees of freedom - strings stretched between branes - are just as important as spatially localized, "diagonal" degrees of freedom - the brane constituents of the black hole. If you combine those pictures, then the horizon is a threshold in space-time between a region dominated by spatially local degrees of freedom (the region outside the horizon) and a region dominated by a form of spatial nonlocality specifically due to quantum gravity (the region inside the horizon).

Unlike the holographic principle, to my knowledge, black hole complementarity is not an idea which has an exact realization yet (whereas AdS/CFT is an exact realization of an otherwise fuzzy idea about quantum gravity, the holographic principle, as proposed by 't Hooft and Susskind a few years earlier).

 

[Fra]

we can preform an experiment with two different observers and the expected result for each observer will be very different.

So do we need to modify the principles of standard QM in any way to accommodate for
black holes??

I certainly think QM needs revision.

That different obsevers make different inferences is something we already have in SR. What recovers objectivity, is then the transformations that relates the observations, which usually also implies new interaction terms between the observers making the difference inferences.

The difference is if we consider the comparasion between observers as an "external inference" (which usually means the inference is made by an observer at infinity or the boundary, observing small observers inside interacting) or if we need to consider this as an inside inference. In the latter case QM as it stands fails as I see it.

In particular with gravity where the concept of inertia is esssential and the gravitational interaction between two observers really depends on their relative inertia, then I think the understanding of this interaction becomes destroyed if you always just consider asymptotical observables like scattering matrixes, because these make use of IMO an observer at infinity with unlimited inertia - destroying the whole mechanics behind gravity. Ie. the mechanisms get lost once we take this limit, and we end up in an ad hoc way of comparing infinites in a way that doesn't seem rational.

So I'm convinced we need to rework measurement theory, to account for these things in order to better understand how to compare intrinsice observables. In SR/QFT it's of course easier, we get away with a principally flawed reasoning since we can always rely on an asymptotically flat background that is observer independent.

/Fredrik

 

[Finbar]

I think the idea is that the interior of the black hole has a dual (holographic) description in terms of states on the horizon; a lot like AdS/CFT, with the horizon being the boundary to the interior. So when someone crosses the horizon from outside, there's a description which involves them continuing to fall inwards, until they are torn apart by tidal forces and their degrees of freedom redistributed among the black hole's degrees of freedom, all of which will later leak away via Hawking radiation; but there's another description in which, when you arrive at the horizon, your degrees of freedom get holographically smeared across it, once again mingling with all the black hole's prior degrees of freedom (also located on the horizon), which all eventually leak away as Hawking radiation.

But in the description of the in falling observer there is no way for the information to get out. Only in the description of the where the degrees of freedom are smeared across the horizon is this possible.

I think actually the crucial point in understanding all this is Maldacena's/Hawking's insights
from considering the full sum over geometries.

http://arxiv.org/abs/hep-th/0106112

http://arxiv.org/abs/hep-th/0507171


What I want to understand is how to combine these insights with those of black hole complementarity.

 

[Naty1]

FINBAR:

This seems to be the only viable option once you take the backreaction of the Hawking radiation into account?...

What is "backreaction"??

But in the description of the in falling observer there is no way for the information to get out.

Get out of what: there is no horizon for such an observer...Only for an accelerating observer, stationary outside the horizon, ARE there horizon effects.

If you are interested in stringy explanations and things like ADS explanations, try Leonard Susskind's THE BLACK HOLE WAR his decades long arguments with Stephen Hawking over the "true" nature of black holes.

Porters post above comports with my understanding, but there are other views:
Susskind says (pg 407):

Unlike ordinary black holes, the ADS variety doesn't evaporate. The horizon is an infinitely hot surface, which continually emits photons. ..Instead of evaportaing into space, they fall back into the black hole.

Without having studied all the detailed math myself, I've come to the conclusion from reading expert's interpretations that "observing" a black hole is akin to observing things in relativity: what you "see" depends on your frame of reference...there is no universal "right" answer anymore than there is a "right" frame of cosmologial reference in GR.

 

[Naty1]

I just realized no one mentioned th Unruh Effect:

Near a black hole quantum and thermal jitters get mixed up and appear as nothing unusual to the free falling observer (who sees no horizon) but become exceedingly dangerous thermal fluctuations to a stationary (accelerating) outside observer (who is confronted by an horizon).

This IS the Unruh effect...Wikipedia explains how different observers perceive the vacuum differently..and how the mathematics of the Hamiltonian and Rindler coordinates, for example, reflect those differences.

 

[Finbar]

Naty,

Back reaction is the coupling of the Hawking radiation back on to the spacetime which isn't taken into account normally.

You can define the event horizon globally without having to specify any observer.
What black hole complementarity suggests is over and beyond different observers observing events in different orders as in standard general relativity. The two descriptions are not related by a coordinate transformation or a diffeomorphism. As far as I can tell the spacetime geometries appear to be different.

I'm not looking for a pop science account. I need someone that really understands the subject.

 

[mitchell porter]

What I want to understand is how to combine these insights with those of black hole complementarity.

I recommend http://arxiv.org/abs/1108.0302" (inventor of the fuzzball hypothesis): "the phase space of quantum gravity is so large that the measure in the path integral can compete with the classical action for macroscopic objects undergoing gravitational collapse". (He's sounding a bit like Erik Verlinde there, who speaks of "the hidden phase space of our universe".)

But I think the idea of black hole complementarity is being taken down some wrong paths - specifically, the idea that there is no conflict between the observer outside and the observer inside, just because they measure noncommuting variables. It's a bit like Wigner's friend, as a paradox for people who believe consciousness collapses the wavefunction: if there's an "observer" in Schrodinger's catbox, was the wavefunction collapsed when you opened the box and looked inside, or did the observer already in the catbox collapse it first? The whole premise (consciousness collapses the wavefunction) is silly, and it's also silly to think that the similar problem of two observers, one who sees a horizon and one who doesn't, can be dealt with by a sleight of hand involving noncommuting observables. The destruction of an observer is not going to be basis-relative. This is why I prefer Mathur's approach, since it involves a detailed physical picture of the black hole.

He does see some relationship between his approach and Maldacena's, by the way. But here is how he summed it up once: "the important point of the fuzzball proposal is that the size of the bound state is always large (horizon size) and not Planck size... The throats of extremal holes end in "fuzzy caps" rather than in a horizon, and this solves the information puzzle."

For more ,see Lubos http://motls.blogspot.com/2005/10/samir-mathurs-black-holes.html" .

 

[Fra]

What black hole complementarity suggests is over and beyond different observers observing events in different orders as in standard general relativity. The two descriptions are not related by a coordinate transformation or a diffeomorphism.

Yes, this is because the diffeomorphism invariance is an external inference; in classical GR you don't need an "observer" comparing and thus leaving bias to equivalence classes of observations.

I think the argument that since no observer can compare the two contradicting pictures, it is per see no contradiction is correct.

Classically we can get away with this since the original observers aren't distorted by submitting or sharing information to a "superobserver". This is why this somewhat not quite right reasoning slips through in classical GR. Ie. it remains a philosophical objection that no one takes seriously.

But in a measurement theory including gravity this won't do. I think that instead if you look at the physics of the inference, comparing the inside vs outside view, one needs to understand the general mechanics for how bh radiation encodes information. Then one also realizes that in order to compare the two views, the BH needs to be fully evaporated (and decoded by the externa observer). But you can't both have the black hole and decode it at the same time, without actually consuming it.

And I think this overall picture, can't be described in an observer independent way. And to make sense out of that I think we need a new understanding of what a theory means if scientific theories are suddently observer dependent.

IMHO, this has absolutely nothing to do with "consciussness", it has to do with the complexity(mass) of the system making the inferneces (observation, encoding and retention of information). The external asymptotic observer contains more information. And the conclusion of thermal character is possibly made relative to the inside observer (which is less massive). Thus the information content of radiation is generally observer dependent, and a light observer are physically inable to ever crack the code even in principle.

This is something that is completely ignored in QFT. We just talk about measurements, but we never care about ow the information about measurments is coded an retained. Because it's in this encoded information that these presumed "inconsistencies" must appear. BH complementarity suggests that these inconsistencies don't appear, because the inconsistency is not a valid physical inference.

So I definitely think we need to reunderstand QM.

In QFT, we mentally encode all information in asymptotic infinity. That makes perfect sense for many cases, but not for "inside observers" in gravity.

/Fredrik

 

[Fra]

The analogy with wigners friend is not perfect. QM is verified for small subsystems where we can rely on observers at infinity. Then this problem stays almost at a philosophical objection (if you don't count understanding unification, we still lack a GUT).

if there's an "observer" in Schrodinger's catbox, was the wavefunction collapsed when you opened the box and looked inside, or did the observer already in the catbox collapse it first?

The point is that there is no such thing as THE wavefunction of the cat; the wavefunction is encoded in the observing system. so there are as many wavefunctions as there are observers.

Then the problem becomes that of defing the equivalence classes of observers. In QFT, with asymptotic observers we can solve this for the case of observers related by poincare transformations.

With gravity we can't do it in the same way if we accept inside observers.

Actually we cant' do it with inside observers in QFT either. because we just get hte S-matrix.

Is this satisfactory?

/Fredrik