Fire Wallow disfavors Event Horizons

[jimgraber]

In the context of the recent Fuzz or Fire workshop,
http://online.kitp.ucsb.edu/online/fuzzorfire_m13/
I would like to make a short argument that event horizons are irredeemably in conflict with unitarity:
An argument frequently mentioned in the Fuzz or Fire workshop is that by the principle of equivalence you should not be able to tell if you are falling in through the event horizon of a large black hole. I would extend this to say that similarly, you should not even be able to detect if you are emerging out through the event horizon of a large black hole. Since this is manifestly false, by reductio, you should conclude that event horizons are irreducibly problematic for unitarity. As singularities are also agreed (almost universally) to destroy unitarity, if you want quantum gravity to be compatible with unitarity, you require a black hole like object with neither a singularity nor an event horizon.
(I have been proposing this necessity for many years)
Singularities have been unpopular forever. It is encouraging to find that horizons are also now being challenged by serious researchers.
Jim Graber

 

[marcus]

Hi Jim,
do you see something NEW in the attitude toward event horizon? As long as I remember it has been presented as not something physical.
A convenient mathematical marker that you really could not locate unless you knew the whole future history of the universe.
"boundary of the causal past of future null infinity"

future null infinity is the set of asymptotes reachable by a light ray, the reachable "points at infinity", and
being in the causal past of that set means you could send a light ray to infinity.
points inside the EH were NOT in the causal past of...etc. So EH was defined as the boundary.

I think according to Wipikeedia the black hole EH is only rigorously defined in asymptotically flat space. I recall people pointing out "there could be an event horizon right in this room, but we can't tell because we don't know the future".

That was long ago. It seems to me it has always been treated as a conceptual, not-physical thing.
That is, with some sophistication.
But maybe I'm just not remembering cases where people were talking as if the EH was a concrete physical thing.

Anyway, you were almost certainly right all along if you were,as you say, skeptical of EH physical reality. There could also be some shift in the prevailing attitude amongst the experts, but I'm missing it.
I hear them doing a reductio ad absurdum about (an oversimple?) idea of the AdS/CFT conjecture.

As several have said, "nobody believes in the firewall". It is just a symptom that one of their working assumptions must be wrong.

If the bulk contains a BH then what you THOUGHT was the boundary is not the whole boundary so it is too simple to continue relying on AdS/CFT, if you were, in that case. Or maybe I'm wrong and it is some other assumption that is causing the trouble.

 

[jimgraber]

Hi Marcus,
What I think is new is a slightly more favorable attitude toward considering configurations such as baby universes or wormholes as a serious alternative to the standard black hole picture. Some of these configurations actually have no event horizons, even if you can look all the way to infinity. I agree that the local inability to detect an event horizon is a concept that has been around and accepted for a long time.

 

[atyy]

I like best the fuzzball approach, which has long suggested no horizons. It isn't clear that any of the new arguments from AMPS are right, although there isn't clear evidence they're wrong either. OTOH, the fuzzball approach is supposed to be actual solutions. I'm not sure whether they get Hawking radiation, but http://arxiv.org/abs/1001.1444 indicates they have some ideas.