![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinSo there's this paper that got a lot of recent publicity by Tanmay Vachaspati, Dejan Stojkovic and Lawrence Krauss that proposed to solve the black hole information loss paradox by claiming that black holes never actually form in the first place. They got a suggestive result for the model of a collapsing sphere of domain wall in a field theory that supports such a thing (kind of like a cosmic string, only two-dimensional like a sheet). They say that the domain wall radiates away all its energy as "pre-Hawking radiation" before the event horizon forms.
This is reminiscent of an objection to black holes that I came up with independently and got into arguments about as a college kid, and that lots of college kids come up with independently. If time slows to a stop at the event horizon, and if Hawking radiation comes out at a finite rate, won't the black hole evaporate before anything falls into it? If you actually fell in, wouldn't you just see the black hole evaporate faster and faster until it disappeared, just before you got there? The conventional answer is no, as I explained in the sci.physics FAQ many years ago—this is just a confusion arising from ascribing too much importance to Schwarzschild coordinates; but as I said there, that conventional story about what happens involves some conjecture.
Vachaspati, Stojkovic and Krauss, as I understand it, are saying that this objection is actually correct. Maybe Hawking radiation as described by Hawking doesn't do the trick, but there's this "pre-Hawking radiation" that does. Anything that is going to collapse into a black hole will radiate away as pre-Hawking radiation before it happens.
Maybe it works for the particular setup and field theory they consider, but the more I think about it, the more I think that this can't possibly be a general solution to the problem without some very strange modifications to known physics. The basic problem is the same one as Jacques Distler's objection to George Chapline's objection (though their paper makes much more sense than Chapline's). An event horizon is a global property of spacetime; the local physics there can in general be as uninteresting as you please. What is going to make generic matter radiate away when it gets there?
They discuss the case of an infalling observer who visits after the fact, but we can remove a step. Suppose you've got nothing but an enormous hollow sphere of freely-falling observers of finite mass, many, many light-years across, say. They fall toward each other, observing as they go. At some point, the sphere gets small enough that the observers are now within the Schwarzschild radius for their mass. If there are enough of them, they can be as widely separated at this point as you like, and tidal forces will be low (as the paper actually acknowledges). They won't even know they're now within an event horizon without doing some calculations—and assuming things they can't know directly about the observers on the other side of the sphere. Local measurements won't do it. But they're now inexorably doomed to collapse together and form a singularity, some years in the future.
That's what conventional physics says, at least. But if Vachaspati, Stojkovic and Krauss really have the general solution to the black hole information loss paradox, then it seems to me that sometime before the observers get within that radius, they should all spontaneously explode.
It's got to be some sort of extraordinarily efficient energy conversion, so that their mass can radiate away. How does that happen? Is baryon number not conserved? There aren't any convenient horizons to hide it behind, or singularities so that they can fall off the edge of conventional physics. What if the observers on the other side of the sphere decide at the last minute to rocket away in the opposite direction—do the observers on this side still explode? If not, why not?
Oh, yes, and: In Distler's case of the collapsing shell of light, their analysis would seem to imply that the shell should bounce off its would-be Schwarzschild radius, which is equally bizarre.
There was also a lively discussion about this at Bad Astronomy.
This is reminiscent of an objection to black holes that I came up with independently and got into arguments about as a college kid, and that lots of college kids come up with independently. If time slows to a stop at the event horizon, and if Hawking radiation comes out at a finite rate, won't the black hole evaporate before anything falls into it? If you actually fell in, wouldn't you just see the black hole evaporate faster and faster until it disappeared, just before you got there? The conventional answer is no, as I explained in the sci.physics FAQ many years ago—this is just a confusion arising from ascribing too much importance to Schwarzschild coordinates; but as I said there, that conventional story about what happens involves some conjecture.
Vachaspati, Stojkovic and Krauss, as I understand it, are saying that this objection is actually correct. Maybe Hawking radiation as described by Hawking doesn't do the trick, but there's this "pre-Hawking radiation" that does. Anything that is going to collapse into a black hole will radiate away as pre-Hawking radiation before it happens.
Maybe it works for the particular setup and field theory they consider, but the more I think about it, the more I think that this can't possibly be a general solution to the problem without some very strange modifications to known physics. The basic problem is the same one as Jacques Distler's objection to George Chapline's objection (though their paper makes much more sense than Chapline's). An event horizon is a global property of spacetime; the local physics there can in general be as uninteresting as you please. What is going to make generic matter radiate away when it gets there?
They discuss the case of an infalling observer who visits after the fact, but we can remove a step. Suppose you've got nothing but an enormous hollow sphere of freely-falling observers of finite mass, many, many light-years across, say. They fall toward each other, observing as they go. At some point, the sphere gets small enough that the observers are now within the Schwarzschild radius for their mass. If there are enough of them, they can be as widely separated at this point as you like, and tidal forces will be low (as the paper actually acknowledges). They won't even know they're now within an event horizon without doing some calculations—and assuming things they can't know directly about the observers on the other side of the sphere. Local measurements won't do it. But they're now inexorably doomed to collapse together and form a singularity, some years in the future.
That's what conventional physics says, at least. But if Vachaspati, Stojkovic and Krauss really have the general solution to the black hole information loss paradox, then it seems to me that sometime before the observers get within that radius, they should all spontaneously explode.
It's got to be some sort of extraordinarily efficient energy conversion, so that their mass can radiate away. How does that happen? Is baryon number not conserved? There aren't any convenient horizons to hide it behind, or singularities so that they can fall off the edge of conventional physics. What if the observers on the other side of the sphere decide at the last minute to rocket away in the opposite direction—do the observers on this side still explode? If not, why not?
Oh, yes, and: In Distler's case of the collapsing shell of light, their analysis would seem to imply that the shell should bounce off its would-be Schwarzschild radius, which is equally bizarre.
There was also a lively discussion about this at Bad Astronomy.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2007-07-04 01:08 pm (UTC)no subject
Date: 2007-07-04 02:01 pm (UTC)no subject
Date: 2007-07-04 02:02 pm (UTC)no subject
Date: 2007-07-04 02:05 pm (UTC)no subject
Date: 2007-07-04 04:59 pm (UTC)Do I have that right? It seems pretty counterintuitive, not that intuition is necessarily all that trustworthy.
no subject
Date: 2007-07-05 12:05 am (UTC)In ordinary treatments of black holes there's no sharp distinction made between the stuff that initially collapses to make the hole and stuff that falls in later. With these strange not-quite-holes it seems as if there is a sharp distinction, and that bothers me too. My sphere of massive observers is an intentional short circuit of that.
no subject
Date: 2007-07-08 01:43 pm (UTC)no subject
Date: 2007-07-12 03:29 am (UTC)