Monday, May 30, 2011

Great day in arxiv

Today there are in arxiv two articles that look really great.

The firs (in the order that arxiv gives to them) is from Samir D. Mathur: Effective information loss outside the horizon.

It argues that there is no loss of information inside a black hole because he information simply doesn't go inside the black hole. The abstract explains it more carefully:

If a system falls through a black hole horizon, then its information is lost to an observer at infinity. But we argue that the {\it accessible} information is lost {\it before} the horizon is crossed. The temperature of the hole limits information carrying signals from a system that has fallen too close to the horizon. Extremal holes have T=0, but there is a minimum energy required to emit a quantum in the short proper time left before the horizon is crossed. If we attempt to bring the system back to infinity for observation, then acceleration radiation destroys the information. All three considerations give a critical distance from the horizon $d\sim \sqrt{r_H\over \Delta E}$, where $r_H$ is the horizon radius and $\Delta E$ is the energy scale characterizing the system. For systems in string theory where we pack information as densely as possible, this acceleration constraint is found to have a geometric interpretation. These estimates suggest that in theories of gravity we should measure information not as a quantity contained inside a given system, but in terms of how much of that information can be reliably accessed by another observer.

The other article is written by Maldacena: Einstein Gravity from Conformal Gravity.

The abstract is:

We show that that four dimensional conformal gravity plus a simple Neumann boundary condition can be used to get the semiclassical (or tree level) wavefunction of the universe of four dimensional asymptotically de-Sitter or Euclidean anti-de Sitter spacetimes. This simple Neumann boundary condition selects the Einstein solution out of the more numerous solutions of conformal gravity. It thus removes the ghosts of conformal gravity from this computation. In the case of a five dimensional pure gravity theory with a positive cosmological constant we show that the late time superhorizon tree level probability measure, $|\Psi [ g ]|^2$, for its four dimensional spatial slices is given by the action of Euclidean four dimensional conformal gravity.">We show that that four dimensional conformal gravity plus a simple Neumann boundary condition can be used to get the semiclassical (or tree level) wavefunction of the universe of four dimensional asymptotically de-Sitter or Euclidean anti-de Sitter spacetimes. This simple Neumann boundary condition selects the Einstein solution out of the more numerous solutions of conformal gravity. It thus removes the ghosts of conformal gravity from this computation.
In the case of a five dimensional pure gravity theory with a positive cosmological constant we show that the late time superhorizon tree level probability measure, $|\Psi [ g ]|^2$, for its four dimensional spatial slices is given by the action of Euclidean four dimensional conformal gravity.


Unfortunately until the next Friday I am going to be very busy and I badly will have time to read them carefully those days so I can't say too much more about them. I suppose that (at least) Lubos will talk about them so I will read its report before I can read them myself. I write this entry partially to recommend the articles to however could be interested and also to keep a link to them so I could later have a quick access to them from wherever I want.

Update: Well, at last I had no patient and read the first article (after all is a brief one, only 7 pages). I have a mixed filling about it. The author computes a few things related to the fall of a body towards an event horizon. Firstly he does for an Schwarschild one.

There he considers two cases. The first in the free fall. In that case the last light (containing the info about the object) is emitted, because of the red-shift at a frequency bellow the Hawking temperature and so it can't be differentiated from this and he concludes that we actually don't have the information about that object.

The second case is when an observer at infinity holds the infalling object until the last time. In that case it is the unrhu radiation associated to the acceleration of an object at rest respect to a gravitational field which is responsible for a dissipation of the information of the object when it finally is released and cross the horizont.

Later he calculates similar things for a Reissner-Nordstöm like black hole and he finds that somewhat different mechanism operate in order to get similar qualitative and quantitative results.

In the last part he does calculations using string theory and the fuzzball paradigm for black holes (where the notion of event horizon is replaced by an stringy construction). Still he finds equivalent results.

Certainly the fact that many different calculations lead to a similar result is appealing. But still I don't see clear the whole subject. I think that at best he would be saying that the lost of information happens before the horizon (or its fuzzball "equivalent") so the problem of lost of unitarity remains (and even we could say that is getting worst because it happens in a region causally connected with the outsider observer). But the whole thing is that one could think that a priory we could think that if the outside of the black hole is clean of other infalling matter (other that the actual object under study) we could argue that if we know the state of the object at infinity we can apply the laws of quantum mechanics to know t's state when it is falling (eve it we can't actually do a measure to be sure that nothing has perturbed our object). That contrast the case of the object that falls behind the horizon when we have no idea of which it's final state would be because we don't know the laws of quantum gravity near the singularity. Well, I am ware that this last objection is somewhat wrong because the key point of the lost of information is the horizon and not the singularity but I have no more time just now to see what point I am missing. I'll realize it for sure later, but I don't promise to write it here soon. But keep calm, for sure Lubos will write about it sooner or later and will clarify the relevant points ;).

1 comment:

Mitchell said...

I wrote about these same two papers - and a third - here!

I believe the paper by Maldacena is the most important one.