- #1
- 24,775
- 792
The fuss over "firewalls" is basically a challenge to the assumption that watching a black hole form and evaporate, the observer at infinity sees unitary evolution. If you think the BH singularity might be resolved in a bounce, you don't make that assumption (as Lee explains in the following). On the other hand if you do make that assumption you arrive at one or more unattractive or paradoxical conclusions.
===quote===
Scott Aaronson says:
August 23, 2013 at 9:32 pm
...While I’m obviously far from an expert, where I think I part ways from you and Unruh is on the following. We’re pretty sure black holes have an entropy, which goes like the area of the event horizon in Planck units. We’re pretty sure that, from an external observer’s perspective, infalling stuff gets “pancaked” on the event horizon and scrambled beyond recognition, never making it through to the interior. Finally, we’re pretty sure that the external observer ultimately sees the black hole evaporate, through Hawking radiation that emerges (appears to emerge?) from the horizon. To me, these facts would seem like an intolerable coincidence, if the black hole didn’t have microstates—”stored,” one wants to imagine, on or near the event horizon—and if the Hawking radiation didn’t carry away the information about those microstates...
...
...
Lee Smolin says:
August 24, 2013 at 7:18 am
Dear Scott,
Thanks, but either I don’t understand your argument or else it is circular. What do you suppose happens to the singularity as well as to the quantum state of the star whose collapse formed the black hole in the first place? If the singularity is eliminated then the Hilbert space in the future is a direct product of a factor spanned by observables which describe degrees of freedom to the future of where the singularity would have been and a factor spanned by observables external to the horizon. The evolution onto this product can be assumed to be unitary but (I feel silly telling you this) it cannot be when restricted to either of its factors. Hence the observer at infinity describes a density matrix gotten by tracing out the degrees of freedom in the baby universe inaccessible to them.
Isn’t this a completely reasonable option, especially because it avoids the otherwise paradoxical implications of the firewall argument?
The pancake is a non-sequitur: why does it matter what information does or doesn’t get to infinity or when, if infinity is not the only place information goes to? So to refer to it seems to assume what you are claiming to demonstrate.
Many thanks,
Lee
==endquote==
http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159261
and
http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159264
===quote===
Scott Aaronson says:
August 23, 2013 at 9:32 pm
...While I’m obviously far from an expert, where I think I part ways from you and Unruh is on the following. We’re pretty sure black holes have an entropy, which goes like the area of the event horizon in Planck units. We’re pretty sure that, from an external observer’s perspective, infalling stuff gets “pancaked” on the event horizon and scrambled beyond recognition, never making it through to the interior. Finally, we’re pretty sure that the external observer ultimately sees the black hole evaporate, through Hawking radiation that emerges (appears to emerge?) from the horizon. To me, these facts would seem like an intolerable coincidence, if the black hole didn’t have microstates—”stored,” one wants to imagine, on or near the event horizon—and if the Hawking radiation didn’t carry away the information about those microstates...
...
...
Lee Smolin says:
August 24, 2013 at 7:18 am
Dear Scott,
Thanks, but either I don’t understand your argument or else it is circular. What do you suppose happens to the singularity as well as to the quantum state of the star whose collapse formed the black hole in the first place? If the singularity is eliminated then the Hilbert space in the future is a direct product of a factor spanned by observables which describe degrees of freedom to the future of where the singularity would have been and a factor spanned by observables external to the horizon. The evolution onto this product can be assumed to be unitary but (I feel silly telling you this) it cannot be when restricted to either of its factors. Hence the observer at infinity describes a density matrix gotten by tracing out the degrees of freedom in the baby universe inaccessible to them.
Isn’t this a completely reasonable option, especially because it avoids the otherwise paradoxical implications of the firewall argument?
The pancake is a non-sequitur: why does it matter what information does or doesn’t get to infinity or when, if infinity is not the only place information goes to? So to refer to it seems to assume what you are claiming to demonstrate.
Many thanks,
Lee
==endquote==
http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159261
and
http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159264
Last edited: