Most physicists foolhardy enough to write a paper claiming that “there are no black holes” — at least not in the sense we usually imagine — would probably be dismissed as cranks. But when the call to redefine these cosmic crunchers comes from Stephen Hawking, it’s worth taking notice. In a paper posted online, the physicist, based at the University of Cambridge, UK, and one of the creators of modern black-hole theory, does away with the notion of an event horizon, the invisible boundary thought to shroud every black hole, beyond which nothing, not even light, can escape.
In its stead, Hawking’s radical proposal is a much more benign “apparent horizon”, which only temporarily holds matter and energy prisoner before eventually releasing them, albeit in a more garbled form.
“There is no escape from a black hole in classical theory,” Hawking told Nature. Quantum theory, however, “enables energy and information to escape from a black hole”. A full explanation of the process, the physicist admits, would require a theory that successfully merges gravity with the other fundamental forces of nature. But that is a goal that has eluded physicists for nearly a century. “The correct treatment,” Hawking says, “remains a mystery...”
Would be interesting to see how Susskind reacts to this...Information Preservation and Weather Forecasting for Black Holes
S. W. Hawking
(Submitted on 22 Jan 2014)
It has been suggested [1] that the resolution of the information paradox for evaporating black holes is that the holes are surrounded by firewalls, bolts of outgoing radiation that would destroy any infalling observer. Such firewalls would break the CPT invariance of quantum gravity and seem to be ruled out on other grounds. A different resolution of the paradox is proposed, namely that gravitational collapse produces apparent horizons but no event horizons behind which information is lost. This proposal is supported by ADS-CFT and is the only resolution of the paradox compatible with CPT. The collapse to form a black hole will in general be chaotic and the dual CFT on the boundary of ADS will be turbulent. Thus, like weather forecasting on Earth, information will effectively be lost, although there would be no loss of unitarity.
Does "interpolated" mean something like topologically stitching together a spacetime consisting of the exterior region of a standard evaporating black hole with a new inner region where the spacetime is a "periodically identified anti deSitter metric" rather than the usual black hole interior spacetime?I take this as indicating that the topologically trivial periodically identified anti deSitter metric is the metric that interpolates between collapse to a black hole and evaporation.
I don't know enough to understand any of the details of this, but this part on p. 28, combined with the diagram on p. 32, sounds like it could be related to what Hawking was saying in the statement I quoted earlier:atyy said:Also, one prominent proposal against an event horizon is the fuzzball proposal, which has been around long before Hawking's latest paper.
http://arxiv.org/abs/hep-th/0502050
The fuzzball proposal for black holes: an elementary review
Samir D. Mathur
Also this on p. 35:Thus the full geometry is flat at infinity, has a ‘throat’ type region at smaller r where it approximates the naive geometry (4.104), and then instead of a singularity at r = 0 it ends in a smooth ‘cap’. This particular geometry, corresponding to the profile (5.105), was derived earlier in [14, 15] by taking limits of general rotating black hole solutions found in [16]. We have now obtained it by starting with the particular NS1-P profile (5.105), and thus we note that it is only one member of the complete family parametrized by F⃗. It can be shown that all the metrics of this family have the same qualitative structure as the particular metric that we studied; in particular they have no horizons, and they end in smooth ‘caps’ near r = 0. We will review the argument for this smoothness below
Is this "smooth cap" the same as (or related to) the "periodically identified anti deSitter metric" that Hawking says "interpolates between collapse to a black hole and evaporation"?There is a naive geometry that has a horizon, and we have ‘actual’ geometries that have no horizons but that differ from each other inside a region of the size of the horizon.
If there is matter being temporarily held behind the apparent horizon, would not this matter be packed even denser than neutrons? Perhaps the matter, once behind the apparent horizon, has been reduced to quarks due to temperature and pressure.If Hawking is correct, there could even be no singularity at the core of the black hole. Instead, matter would be only temporarily held behind the apparent horizon, which would gradually move inward owing to the pull of the black hole, but would never quite crunch down to the centre. Information about this matter would not destroyed, but would be highly scrambled so that, as it is released through Hawking radiation, it would be in a vastly different form, making it almost impossible to work out what the swallowed objects once were.
I'm not sure "held" is the right term. I believe the picture that Hawking is presenting is essentially that black holes are regions of space-time with extreme turbulence. This turbulence both makes it take a very long time for matter to re-emerge, and also effectively randomized said matter.|Glitch| said:If there is matter being temporarily held behind the apparent horizon, would not this matter be packed even denser than neutrons? Perhaps the matter, once behind the apparent horizon, has been reduced to quarks due to temperature and pressure.
Black holes do radiate.|Glitch| said:Also, if the event horizon becomes smaller than the apparent horizon, would not the black hole "radiate" light?
I don't think Hawking was suggesting deterministic chaos alone could explain it without also invoking some theory of quantum gravity--I thought the Penrose-Hawking singularity theorems showed definitively that in classical general relativity, if a mass collapses beyond a certain point there's no way to avoid the collapsing mass getting crushed into a singularity.Ken G said:Looking at all the recent flap about the black hole information paradox, and the more recent "firewall paradox", it is hard not to conclude that we have a tempest in a teacup the size of an event horizon! Look at Hawking's own admission: "The correct treatment remains a mystery." So what we have is, theorists arguing about the right way to do calculations in the overlap of two theories that are known to be inconsistent with each other. What we don't have is people saying "solution X leads to observable effect Y, so let's look for Y," or "we already have observation Y, so we know we should be looking for X in the correct solution." So this is a pretty sorry state of affairs. What makes it really clear how bad the problem is, is that Hawking's recent paper is essentially concluding that the resolution of the black hole information paradox is something completely mundane: deterministic chaos. This means it is the same resolution as the resolution of the "information paradox" in understanding the movements of the air around us right now, despite all the tours de force invoking AdS/CFT duality and holographic speculation. That should be giving us some pause, I would say.
You need quantum mechanics to have Hawking radiation in the first place, but if you accept that you have that, then Hawking's resolution of the information paradox is just like a resolution of the information paradox in a Newtonian description of weather. Newtonian mechanics respects CPT too, and seems to have all the essential elements Hawking quotes in his paradox resolution, as long as we stipulate the existence of Hawking radiation. Indeed, if chaos precludes the creation of real event horizons, then the Penrose-Hawking theorems are invalid anyway, even without Hawking radiation or quantum mechanics.JesseM said:I don't think Hawking was suggesting deterministic chaos alone could explain it without also invoking some theory of quantum gravity--I thought the Penrose-Hawking singularity theorems showed definitively that in classical general relativity, if a mass collapses beyond a certain point there's no way to avoid the collapsing mass getting crushed into a singularity.
It's not clear that's true--Hawking radiation would ultimately be derived from a theory of quantum gravity but the current derivation uses semiclassical gravity, which from what I've read still involves a classical spacetime geometry obeying the rules of general relativity, rather than a superposition of different geometries or something. So I don't think any of the conclusions about when event horizons and singularities become inevitable would be different in semiclassical gravity than they are in general relativity with classical matter fields. However, as I mention below the singularity theorems do depend on certain energy conditions, and quantum fields can violate them in certain cases, so this might be a way out.Ken G said:You need quantum mechanics to have Hawking radiation in the first place, but if you accept that you have that, then Hawking's resolution of the information paradox is just like a resolution of the information paradox in a Newtonian description of weather.
What are the essential elements you're referring to? It seems to me that Hawking is invoking ideas beyond just classical chaos + Hawking radiation + CPT invariance--for example look at the section from his paper I was asking about earlier, where he said "I take this as indicating that the topologically trivial periodically identified anti deSitter metric is the metric that interpolates between collapse to a black hole and evaporation." Also, the Nature article here suggests Hawking is just trying to sketch how he thinks things would work in a future theory of quantum gravity: 'A full explanation of the process, the physicist admits, would require a theory that successfully merges gravity with the other fundamental forces of nature. But that is a goal that has eluded physicists for nearly a century. “The correct treatment,” Hawking says, “remains a mystery.”'Ken G said:Newtonian mechanics respects CPT too, and seems to have all the essential elements Hawking quotes in his paradox resolution, as long as we stipulate the existence of Hawking radiation.
I don't think it's true that the Penrose-Hawking singularity theorems depend on the assumption that there's a "real event horizon". The wikipedia article does at one point describe the singularity theorem in terms of event horizons--"The singularity theorems prove that this cannot happen, and that a singularity will always form once an event horizon forms"--but there is no citation for this claim and I suspect it's incorrect, because Hawking's theorem dealt with the Big Bang singularity which wouldn't have an event horizon, and anyway I thought event horizons were defined in terms of the boundary between points where all lightlike worldlines hit a singularity and points where some can escape to infinity. The review on singularity theorems at http://arxiv.org/pdf/physics/0605007.pdf gives an outline on pages 7-8 of what conditions are used to derive the conclusion that a singularity forms, and event horizons aren't among them:Ken G said:Indeed, if chaos precludes the creation of real event horizons, then the Penrose-Hawking theorems are invalid anyway, even without Hawking radiation or quantum mechanics.
As explained on p. 8, #1 is satisfied as long as the matter field doesn't violate certain energy conditions like the strong energy condition, and p. 5 of this presentation by Matt Visser mentions that the Penrose singularity theorem which is relevant to black holes (as opposed to the Big Bang, which Hawking's dealt with) requires the weak energy condition. Also note that p. 6 of Visser's presentation mentions that the averaged null energy condition (ANEC) is used in the "generalized Penrose singularity theorem" by Roman (which seems to be this paper, which says "we show that Penrose’s singularity theorem will still hold if the weak energy condition is replaced by a weaker (nonlocal) energy condition and if the null generic condition holds"), and that "ANEC is the weakest averaged energy condition in common use." So although quantum fields like those involved in Hawking radiation can violate various energy conditions, it sounds like the conclusion of an inevitable singularity would still apply provided Hawking radiation didn't violate ANEC--I'm not sure if current theory says anything definite about this one way or the other.The culmination was the celebrated Hawking-Penrose theorem (Hawking and Penrose, 1970), which since then is the singularity theorem par excellence. However,
all of the singularity theorems share a well-defined skeleton, the very same pattern. This is, succintly, as follows (Senovilla, 1998a)
Theorem 1 (Pattern Singularity Theorem) If a space-time of sufficient differentiability satisfies
1. a condition on the curvature
2. a causality condition
3. and an appropriate initial and/or boundary condition
then there are null or time-like inextensible incomplete geodesics.
Well, you need quantum mechanics to get any kind of radiation, classical physics has the ultraviolet catastrophe. But I don't know if these details are that important, what matters is that even the issue of whether or not one expects Hawking-Penrose singularities in the first place is a matter of debate, and if Hawking thinks that simple chaos provides a way out, like it allows information to be lost in the practice of weather prediction, then what it means to me is that we have yet another example of physicists taking their physical theories too seriously, essentially because they can.JesseM said:It's not clear that's true--Hawking radiation would ultimately be derived from a theory of quantum gravity but the current derivation uses semiclassical gravity, which from what I've read still involves a classical spacetime geometry obeying the rules of general relativity, rather than a superposition of different geometries or something. So I don't think any of the conclusions about when event horizons and singularities become inevitable would be different in semiclassical gravity than they are in general relativity with classical matter fields. However, as I mention below the singularity theorems do depend on certain energy conditions, and quantum fields can violate them in certain cases, so this might be a way out.
My point is, the details of the local theory being used are really not terribly important if chaos is being invoked as its solution. We have all the same issues with a Newtonian description of weather. This is the subtext of Hawking's own analogy, that the resolution is similar to unpredictability of weather. It is nice to have an analogy that everyday people can find meaning in, but in this case, the analogy is very much a double-edged sword: to whatever extent it is a valid analogy, that is the extent to which the heavy-duty theory being invoked to support the conclusion is not terribly relevant. In other words, if you believe you need that heavy-duty theory to get a believable outcome, but the believable outcome ends up looking just like weather prediction, then the actual answer to the paradox transcends the heavy-duty theory, it is a resolution that is available in a broad range of local theories, including ones we haven't even come up with yet. It is a perfectly mundane resolution, the one we should probably have expected all along: that formal physics theories do not always accurately reflect the behavior of the usable information that physicists need to manipulate in order to make testable predictions, of the kind you need to even decide if you will regard that formal theory as accurate. This is not exactly a new discovery!What are the essential elements you're referring to? It seems to me that Hawking is invoking ideas beyond just classical chaos + Hawking radiation + CPT invariance--for example look at the section from his paper I was asking about earlier, where he said "I take this as indicating that the topologically trivial periodically identified anti deSitter metric is the metric that interpolates between collapse to a black hole and evaporation."
Right, that's exactly my point. In my view, Hawking is basically saying that since we don't know the right theory to use, the best we can do is hit it with the biggest sledgehammer we have around, but even when we do that, we end up staring a very simple and old problem right in the face: the formal theory doesn't work in practice, because of simple chaos. That is a conclusion that is quite separate from the formal theory chosen, so if we expect that black hole formation is chaotic, we don't need a formal theory at all, to reach the conclusion that Hawking reaches.Also, the Nature article here suggests Hawking is just trying to sketch how he thinks things would work in a future theory of quantum gravity: 'A full explanation of the process, the physicist admits, would require a theory that successfully merges gravity with the other fundamental forces of nature. But that is a goal that has eluded physicists for nearly a century. “The correct treatment,” Hawking says, “remains a mystery.”'
I'm not saying an event horizon is needed to be assumed, I'm saying that whatever you need to assume to get a singularity is going to produce an event horizon. A "real" event horizon, as opposed to an "apparent" event horizon, seems to be the difference between an eternal black hole, and a temporary one. I believe the idea is that in pure GR, you expect black holes to be eternal if they create real event horizons, which creates a paradox (which is essentially the black hole information paradox) if you tack on Hawking radiation. In other words, the paradox is, how do you marry the quantum mechanical expectation of Hawking radiation with the Hawking-Penrose singularity theorems, which seems related to the problem of marrying unitarity with no-hair theorems.I don't think it's true that the Penrose-Hawking singularity theorems depend on the assumption that there's a "real event horizon". The wikipedia article does at one point describe the singularity theorem in terms of event horizons--"The singularity theorems prove that this cannot happen, and that a singularity will always form once an event horizon forms"--but there is no citation for this claim and I suspect it's incorrect, because Hawking's theorem dealt with the Big Bang singularity which wouldn't have an event horizon, and anyway I thought event horizons were defined in terms of the boundary between points where all lightlike worldlines hit a singularity and points where some can escape to infinity. The review on singularity theorems at http://arxiv.org/pdf/physics/0605007.pdf gives an outline on pages 7-8 of what conditions are used to derive the conclusion that a singularity forms, and event horizons aren't among them:
I'd say the entire landscape of what assumptions lead to what expectations is somewhat overshadowed by the deeper problem of not really having any idea what assumptions are even valid in the first place-- in the absence of better observational constraints.If I'm understand the above summary of the Penrose-Hawking singularity theorems correctly, it shouldn't be possible in general relativity or semiclassical gravity to have such a trapped surface and to avoid a singularity, at least not unless the spacetime contains closed timelike curves or violates ANEC. It might be that current knowledge doesn't rule out the idea that Hawking radiation violates ANEC and that this means semiclassical gravity alone can give a model where there are no singularities and no true event horizons, but I doubt Hawking was trying to argue for such a purely semiclassical explanation, since he doesn't even mention energy conditions in his paper.
I was talking about the fact that Hawking radiation has been derived using semiclassical gravity, which involves modeling quantum fields on the curved spacetime of classical general relativity.Ken G said:Well, you need quantum mechanics to get any kind of radiation, classical physics has the ultraviolet catastrophe.
Debate about what exactly? The Penrose-Hawking singularity theorems only claim anything about what happens in general relativity, they don't claim that general relativity's predictions will perfectly model our actual physical reality. I don't think there's any debate that the theorems are correct about what must happen in general relativity. Since semiclassical gravity still uses the rules of general relativity to model spacetime--it just treats the "matter field" in a quantum way--the singularity theorems should still apply to it. And as far as I can tell the only relevant loophole in the singularity theorems is that if the averaged null energy condition isn't satisfied in the neighborhood of a black hole the singularity theorems might not apply, but Hawking doesn't even mention energy conditions so it doesn't seem that this is that argument he is making.Ken G said:But I don't know if these details are that important, what matters is that even the issue of whether or not one expects Hawking-Penrose singularities in the first place is a matter of debate
That is precisely the thing I am disputing! I don't think that Hawking in fact says anything like "simple chaos provides a way out" in the paper, I think you are misreading it. And my point, again, is that the phrase "simple chaos provides a way out" seems to imply that if we take chaos into consideration, even our existing theoretical model which we use to analyze black hole formation and Hawking radiation (semiclassical gravity) might allow for the possibility that no event horizon or singularity would form. Is that what you're arguing? If so, again my point is that the singularity theorems still apply to semiclassical gravity, they have been proved mathematically so there's no chance that "simple chaos" would negate the conclusion that if certain conditions are satisfied (conditions 1-3 that I quoted), a singularity is inevitable.Ken G said:and if Hawking thinks that simple chaos provides a way out
When you say "biggest sledgehammer we have around" and "the formal theory", what theory are you referring to? General relativity with quantum fields on the spacetime, i.e. semiclassical gravity? That's the only approach I know of to dealing with quantum effects in curved spacetime that doesn't get into speculations about quantum gravity, and we don't have a completed theory of quantum gravity yet (which I assume is what you meant when you said "we don't know the right theory to use").Ken G said:Right, that's exactly my point. In my view, Hawking is basically saying that since we don't know the right theory to use, the best we can do is hit it with the biggest sledgehammer we have around, but even when we do that, we end up staring a very simple and old problem right in the face: the formal theory doesn't work in practice, because of simple chaos.
Again, are you claiming that even if the "formal theory chosen" is general relativity (including semiclassical gravity), chaos can negate the conclusion of the Penrose-Hawking singularity theorems that if certain conditions are satisfied, a singularity is inevitable?Ken G said:That is a conclusion that is quite separate from the formal theory chosen, so if we expect that black hole formation is chaotic, we don't need a formal theory at all, to reach the conclusion that Hawking reaches.
That's all still true, especially since you said "effectively." Hawking isn't saying there aren't things that act like we expect astrophysical black holes to act, he is drawing a formal distinction between things that act for all practical purposes like black holes, and "actual" black holes that could present an information paradox.|Glitch| said:I am confused, which I admit is not an uncommon occurrence. I have always been under the impression that anything smaller than the Schwarzschild radius was considered a black hole (at least for the uncharged, non-rotating variety), and that the surface at this radius was effectively the event horizon.
But that's a pretty big "except"! That was the cause of the paradox.I do not understand how a black hole could not have an event horizon. Is not an event horizon the point were nothing can escape the gravitational pull of the black hole (except for Hawking radiation)?
It seems to be a very technical point, perhaps involving whether the black hole has a true singularity inside there, or some kind of turbulent super-high density but that somehow escapes the singularity theorems. Others who know the details better might chime in.Also, could someone explain the difference between the event horizon and the apparent horizon?
Whether or not there "really are" black holes.JesseM said:Debate about what exactly?
I agree, but you wouldn't think the theorists believe that, the way they talk about the ramifications of their calculations. Let's start with Hawking radiation itself!The Penrose-Hawking singularity theorems only claim anything about what happens in general relativity, they don't claim that general relativity's predictions will perfectly model our actual physical reality.
I don't know if he equates the presence of singularities with the presence of real event horizons, but that seems like a natural connection to make. If so, then he is indeed saying the singularity theorems must not apply, and he agrees that GR says they should, so he is looking for a way that reality can avoid obeying GR. His conclusion seems to be that reality avoids obeying GR much like the way air avoids obeying Newtonian theorems of time-reversible motion, which has to do with the difference between a result of a formal theory, and a result of how information actually works in practice. But I don't see how that directly relates to the singularity theorems, he seems to only connect the issue directly with the information paradox.I don't think there's any debate that the theorems are correct about what must happen in general relativity. Since semiclassical gravity still uses the rules of general relativity to model spacetime--it just treats the "matter field" in a quantum way--the singularity theorems should still apply to it. And as far as I can tell the only relevant loophole in the singularity theorems is that if the averaged null energy condition isn't satisfied in the neighborhood of a black hole the singularity theorems might not apply, but Hawking doesn't even mention energy conditions so it doesn't seem that this is that argument he is making.
I think that all comes under the heading of trying to find the appropriate solution. But my point is, who cares what is the appropriate solution if one is going to invoke chaos anyway-- chaos is an extremely generic, and rather mundane at that, form of dynamics! Certainly not something that requires the sledgehammers that Hawking invokes elsewhere in his paper, such as the one you mention.Then there is also the point, which you didn't respond to, that Hawking seems to be talking about something more than just chaos when he says "I take this as indicating that the topologically trivial periodically identified anti deSitter metric is the metric that interpolates between collapse to a black hole and evaporation."
Then cite the place where Hawking justifies his appeal to chaos, and show me why it wouldn't apply to simple Newtonian physics just as well-- bearing in mind that the "weather" analogy is Hawking's, not mine!That is precisely the thing I am disputing! I don't think that Hawking in fact says anything like "simple chaos provides a way out" in the paper, I think you are misreading it.
It's not what I'm arguing, it's what Hawking is arguing. He is arguing that somewhere along the line, the proper physics will go chaotic. He does not cite any special attributes of the proper physics that will do that, he mostly uses the proper physics to defeat the firewall idea (which Hossenfelder probably already defeated anyway). Indeed, he makes an analogy with a completely different set of physics to say what chaos is doing here, I think that puts it pretty clearly into perspective that the proper physics is rather incidental to that aspect of his argument.And my point, again, is that the phrase "simple chaos provides a way out" seems to imply that if we take chaos into consideration, even our existing theoretical model which we use to analyze black hole formation and Hawking radiation (semiclassical gravity) might allow for the possibility that no event horizon or singularity would form. Is that what you're arguing?
More-- also AdS/CFT duality, the holographic principle, and even CPT symmetry.When you say "biggest sledgehammer we have around" and "the formal theory", what theory are you referring to? General relativity with quantum fields on the spacetime, i.e. semiclassical gravity?
Exactly, Hawking admits that freely-- we don't really know the proper theory to use. That's yet another reason why Hawking's argument has a very "generic" flavor, underneath all the formal references to advanced theories and speculations.That's the only approach I know of to dealing with quantum effects in curved spacetime that doesn't get into speculations about quantum gravity, and we don't have a completed theory of quantum gravity yet (which I assume is what you meant when you said "we don't know the right theory to use").
Again, that is Hawking's argument. He feels that certain aspects of the "right theory" will supercede the GR theorems. My point is, if that opens the door for chaos, as he claims, then the information paradox resolution is just as mundane as the resolution of the "weather information paradox."Again, are you claiming that even if the "formal theory chosen" is general relativity (including semiclassical gravity), chaos can negate the conclusion of the Penrose-Hawking singularity theorems that if certain conditions are satisfied, a singularity is inevitable?
1. Nobody has "seen" a black hole. There are many indirect measurements that demonstrate that objects near to the density of black holes must exist.atombuster said:I think that if black holes where not real than how did the Hubble space telescope see them or how did some astrounats see them? Really now I just think that Hawking is not right about it. You can also read about it on discover magazine online.
Though depending upon the scale of the deviations from a true event horizon, part of me wonders if this sort of experiment might be able to put Hawking's idea to the test:Ken G said:It should also be mentioned that in astronomy, the only kinds of black holes we currently have evidence of are so large that AFAIK they produce no observable effects that have anything to do with Hawking radiation, and no observable effects that have anything to do with the difference between the kinds of event horizons Hawking is talking about, i.e., real vs. apparent event horizons. So in regard to actual astrophysical black holes, none of these issues have any demonstrable relevance whatsoever, and certainly there is no practical difference between a "real" black hole and an "apparent" black hole. Hence, there is no reason to think about them any differently. We really are going to have to find evidence for black holes with much much smaller masses than any we've seen evidence of, before any of this debate is going to have the slightest physical importance, unless the Big Bang itself can be framed as some type of Hawking radiation (and I'm not sure what is the status of that kind of thinking).
Of course the reason people are interested in this is because it seems like a place where quantum mechanics and general relativity merge, so it seems relevant for unification, but I would argue it isn't the right place for that at all, for the simple reason that we do not seem to have any useful observations related to these issues in the context of astrophysical black holes. We should look somewhere else.
