![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinThere's a news item making the rounds about a claim by theorist George Chapline that black holes cannot exist, combined with the strange claim that they are really "dark energy stars". The conference proceeding is very short and refers to papers I haven't studied, but it's the sort of claim that strikes me as almost certainly wrong, and now I discover that Jacques Distler pretty effectively ripped it apart in March after hearing about it in a colloquium talk.
This notion that black holes are so bizarre that general relativity absolutely has to break down at the event horizon seems to come up over and over among people who don't do a lot of work with general relativity. The fact is that in general relativity, nothing particularly unusual happens at the event horizon of a black hole; it's defined by the future behavior of outgoing paths of things moving at the speed of light (including photons themselves; it's simply the surface they will never get out of). At the event horizon of a sufficiently large black hole the local structure of space-time approaches that at any point in empty space. The only way you could in general tell you were at the event horizon, in all possible cases, would be to know about the future, as Distler said. Locally, there's no reason the structure of space-time would prevent quantum mechanics experiments from coming out normally.
Of course, if general relativity is really wrong, then all bets are off—a priori, I suppose anything goes. But Chapline doesn't make a very convincing case that it has to break down right at the event horizon.
His argument has to do with the disturbing singularity in time dilation that appears at the event horizon when you use the traditional Schwarzschild coordinates to describe a black hole (often described, more or less incorrectly, with the claim that "time stops there"). This disturbs Chapline because he insists that quantum mechanics must have an absolute notion of time. If he were right, then QM would be inconsistent even with special relativity, since simultaneity is relative in special relativity and depends on your frame of reference!
But this is an invalid argument. Even ignoring general relativity, in special-relativistic quantum field theory, observable operators are constructed precisely such that the choice of surfaces of simultaneity does not have any observable consequences (technically speaking, they "commute at spacelike separation"). A million science-fiction novels notwithstanding, you can't signal faster than light or tell what the universal time coordinate is using nonlocal correlations, weird as they may be. There appears to be an absolute time variable if you write down the time evolution using a non-covariant Schrödinger or Heisenberg formalism, but workers in QFT devised ways long ago to keep the relativistic symmetry explicit. The absolute time only appears to be crucial if you're using mathematical formalisms borrowed from nonrelativistic quantum mechanics.
Furthermore, suppose he were right and for some weird quantum or sub-quantum reason (Bohm pilot wave theory?) you did have to choose an absolute notion of time. Who says your surfaces of equal time have to correspond to the Schwarzschild t coordinate, which blows up at the event horizon? Well-behaved coordinates for black holes are an extensively studied subject. The universal surfaces of equal time could just as easily be based on something else (such as, say, the Kruskal-Szekeres timelike coordinate, or some variant thereof) that is perfectly well-behaved inside the black hole, at least until you hit the singularity. They could be some set of spacelike surfaces that warp like the pages of a book that fell in the bathtub, closer in some places and further apart in others! (I wrote a half-baked science-fiction story based on this idea once. I'm going to fully bake it someday.)
The spacelike singularity that is the ultimate fate of matter inside the black hole is a real problem for quantum mechanics, as Chapline says; I'd bet that general relativity does break down there. But that's well-known already and mountains of papers have been written about it. The event horizon is an entirely different kettle of fish; the only singularity there is in some ill-chosen coordinate systems.
This isn't just Chapline. Papers insisting that black holes cannot exist come up every so often and essentially all of them make this same mistake, assuming that there is something physically alarming in things that are just properties of Schwarzschild coordinates, such as the "divergent time dilation". They're just coordinates; I could choose a set of coordinates that explode just as spectacularly in my living room tomorrow afternoon.
Finally, the statements about "dark energy stars" are just bizarre—what field theory is he imagining here to posit all that localized vacuum energy? I suppose I should be kind and assume he's got one, since I haven't followed all the references. I know that in Alan Guth's models, localized blobs of dark energy like that tended to inflate off into new baby universes, not sit around as dark energy stars. But, then, Guth wasn't assuming that general relativity is totally wrong in there.
This notion that black holes are so bizarre that general relativity absolutely has to break down at the event horizon seems to come up over and over among people who don't do a lot of work with general relativity. The fact is that in general relativity, nothing particularly unusual happens at the event horizon of a black hole; it's defined by the future behavior of outgoing paths of things moving at the speed of light (including photons themselves; it's simply the surface they will never get out of). At the event horizon of a sufficiently large black hole the local structure of space-time approaches that at any point in empty space. The only way you could in general tell you were at the event horizon, in all possible cases, would be to know about the future, as Distler said. Locally, there's no reason the structure of space-time would prevent quantum mechanics experiments from coming out normally.
Of course, if general relativity is really wrong, then all bets are off—a priori, I suppose anything goes. But Chapline doesn't make a very convincing case that it has to break down right at the event horizon.
His argument has to do with the disturbing singularity in time dilation that appears at the event horizon when you use the traditional Schwarzschild coordinates to describe a black hole (often described, more or less incorrectly, with the claim that "time stops there"). This disturbs Chapline because he insists that quantum mechanics must have an absolute notion of time. If he were right, then QM would be inconsistent even with special relativity, since simultaneity is relative in special relativity and depends on your frame of reference!
But this is an invalid argument. Even ignoring general relativity, in special-relativistic quantum field theory, observable operators are constructed precisely such that the choice of surfaces of simultaneity does not have any observable consequences (technically speaking, they "commute at spacelike separation"). A million science-fiction novels notwithstanding, you can't signal faster than light or tell what the universal time coordinate is using nonlocal correlations, weird as they may be. There appears to be an absolute time variable if you write down the time evolution using a non-covariant Schrödinger or Heisenberg formalism, but workers in QFT devised ways long ago to keep the relativistic symmetry explicit. The absolute time only appears to be crucial if you're using mathematical formalisms borrowed from nonrelativistic quantum mechanics.
Furthermore, suppose he were right and for some weird quantum or sub-quantum reason (Bohm pilot wave theory?) you did have to choose an absolute notion of time. Who says your surfaces of equal time have to correspond to the Schwarzschild t coordinate, which blows up at the event horizon? Well-behaved coordinates for black holes are an extensively studied subject. The universal surfaces of equal time could just as easily be based on something else (such as, say, the Kruskal-Szekeres timelike coordinate, or some variant thereof) that is perfectly well-behaved inside the black hole, at least until you hit the singularity. They could be some set of spacelike surfaces that warp like the pages of a book that fell in the bathtub, closer in some places and further apart in others! (I wrote a half-baked science-fiction story based on this idea once. I'm going to fully bake it someday.)
The spacelike singularity that is the ultimate fate of matter inside the black hole is a real problem for quantum mechanics, as Chapline says; I'd bet that general relativity does break down there. But that's well-known already and mountains of papers have been written about it. The event horizon is an entirely different kettle of fish; the only singularity there is in some ill-chosen coordinate systems.
This isn't just Chapline. Papers insisting that black holes cannot exist come up every so often and essentially all of them make this same mistake, assuming that there is something physically alarming in things that are just properties of Schwarzschild coordinates, such as the "divergent time dilation". They're just coordinates; I could choose a set of coordinates that explode just as spectacularly in my living room tomorrow afternoon.
Finally, the statements about "dark energy stars" are just bizarre—what field theory is he imagining here to posit all that localized vacuum energy? I suppose I should be kind and assume he's got one, since I haven't followed all the references. I know that in Alan Guth's models, localized blobs of dark energy like that tended to inflate off into new baby universes, not sit around as dark energy stars. But, then, Guth wasn't assuming that general relativity is totally wrong in there.