Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Fire wallow favors unibabes

  1. Aug 24, 2013 #1

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    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
     
    Last edited: Aug 24, 2013
  2. jcsd
  3. Aug 24, 2013 #2

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Pointed use of the term reductio ad absurdum where you draw out the consequences of some position until they are seen to be absurd:

    http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159253
    ==quote Lee==
    ...
    The lesson in my opinion is that the key issue in quantum black holes and the information problem is not at the horizon, it is at the singularity. It is unreasonable to expect any new physics at horizons where the curvatures are small, but necessary to find new physics at the approach to singularities. The focus on the firewall problem is in my view a consequence of insufficient appreciation of this point. It can be seen as a reductio for the assumption that the problem can be resolved without investigating how quantum gravity effects eliminate the singularity and taking on board the consequences of the resulting evolution to the future of where the classical singularity would have been.
    ==endquote==
     
    Last edited: Aug 24, 2013
  4. Aug 24, 2013 #3

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Here's the current KITP firewall kerfluffle:
    http://online.kitp.ucsb.edu/online/fuzzorfire_m13/

    Don Marolf explained at the start of his talk that in spite of the fact that his talk was titled
    "The Case for Firewalls" the firewall picture was not something he believed.

    It is instead something to study and think about, a way in fact to find out if we are making unjustified assumptions, to realize that we don't yet have a good understanding of BH information problem.

    He explained that his assignment was to kick off the discussion the first day of the workshop, thus the title.

    The talk by Bill Unruh struck me as definitely to be recommended. Enjoyed it very much.
    http://online.kitp.ucsb.edu/online/fuzzorfire_m13/unruh/
    Ted Jacobson played the game Marolf proposed very seriously, avoiding any suggestion that info might leak out via whatever process occurred instead of a singularity. Good--although quite a bit more technical and less intuitive than Unruh's.
    http://online.kitp.ucsb.edu/online/fuzzorfire_m13/jacobson/
     
    Last edited: Aug 24, 2013
  5. Aug 24, 2013 #4

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    I just came back from this Fuzz or Fire workshop, I thought the talks were very interesting. Bill Unruh definitely took a traditional GR point of view, as did Bob Wald.

    However, in regards to the Aaronson / Smolin exchange in the OP: If black holes evaporate, then all the information does have to radiate out to infinity. Or I guess, I am pretty sure this is what we mean by "evaporate".

    If we are to suppose that black holes toss all the incoming information into a baby universe which is inaccessible to the outside observer, I don't think we've solved any problems. Philosophically speaking, this is a bit of an empty statement. First of all, it is inherently untestable, because the baby universe is inaccessible to observers in this universe. Second, it is tautological that a mixed state can be purified by coupling to an appropriate external system, so saying "The time evolution is unitary if you throw extra universes into the Hilbert space so as to make it unitary" doesn't really get you anything.
     
  6. Aug 24, 2013 #5

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    By the way, if you're watching the videos, I'm the guy sitting to the left of the projector with a laptop. And Iosif Bena mentions me at the beginning of his talk.
     
  7. Aug 24, 2013 #6

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    I'll look for you! I think I just saw your laptop near the start of Jacobson's talk.

    Jacobson at the start of his talk:"...almost nobody believes that the firewall is plausible...
    so we can hope to learn something about quantum gravity and ads/cft just by reconciling..."

    Jacobson says that Don Marolf's recent argument (e.g. in the paper "Holography without

    Strings") that it simply arises from diffeomorphism invariance had persuaded him to take boundary unitarity seriously.
     
  8. Aug 24, 2013 #7

    atyy

    User Avatar
    Science Advisor

    During Marolf's talk or Bena's talk? Left, looking from the audience? Second row?
     
    Last edited: Aug 24, 2013
  9. Aug 24, 2013 #8

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    Yes, that's my laptop near the beginning of Jacobson's talk. I'm sitting behind Bob Wald and next to Lenny Susskind (who doesn't appear in that video, I think).

    Atyy, in the main conference room I took the same chair every time. I think I'm in all of the first week's videos. I won't be there for the second week, though.

    I wasn't aware my bald spot was that bad already. I think that's the main thing I learned from this workshop.
     
  10. Aug 24, 2013 #9

    Haelfix

    User Avatar
    Science Advisor

    I don't understand Smolin's argument at all, it doesn't make sense. As far as I can see, the argument for firewalls, also applies along the Rindler wedge. You draw the exact same picture, and run the exact same arguments.

    If it doesn't apply on the Rindler wedge (and I am aware there are papers that say that it doesn't) it is for very subtle reasons and has almost nothing to do with any statement regarding the singularity (which everyone agrees logically dissappears in a full theory of quantum gravity), the case for firewalls is very strictly confined to horizon, near horizon and stretched horizon degrees of freedom, that are presumably entangled with late time observers.
     
  11. Aug 24, 2013 #10

    atyy

    User Avatar
    Science Advisor

    Ben Niehoff, seems you needed the holographic version of the workshop to learn that fact.
     
  12. Aug 24, 2013 #11

    atyy

    User Avatar
    Science Advisor

    So it's not generally accepted that the firewall arguments don't apply to the Rindler wedge? I'd assumed they can't if the firewall argument is to make sense at all, since we're pretty sure there's no firewall in Minkowski spacetime.
     
  13. Aug 24, 2013 #12

    Haelfix

    User Avatar
    Science Advisor

    Most people expect that firewalls don't exist physically (even if they don't apply in flat space, we might already be past an astrophysical horizon yet observed no excited modes), its just very hard to evade the argument. Logically the argument seems to also hold (the same cartoon) in Rindler space, hence its an important toy model to use and understand (its actually central to the ER-EPR picture). At this time, it seems like we have to throw out one of four cherished principles or find a flaw in the argument.

    I see that Lee likes to give up unitarity for the outside observer by postulating a baby universe. That actually works, b/c it does throw out one of the assumptions of the AMPS paper, but it is not without its own problems.
     
  14. Aug 24, 2013 #13

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Seems straightforward enough to me. What could those "own problems" be? It keeps small scale unitarity. You just have to put up with mixed states vis à vis an astrophysical black hole.

    See quote from Lee in the OP
     
  15. Aug 24, 2013 #14

    Haelfix

    User Avatar
    Science Advisor

    Having a baby universe is essentially identical to having information lost. It means that you can store an arbitrary large amount of entropy in an arbitrarily small region (and this plays havok with effective field theories), its completely unclear why you also wouldn't generically violate unitarity for any theory order by order in perturbation theory by simply pair-creating virtual blackhole loops, and these will generically and badly diverge in the infrared.

    It also implies that you have to give up conservation laws (if information is lost in the observable physical world, then you actually run into problems with the quantum version of Noethers theorems)

    All of this also typically runs afoul of the superposition principle as well.

    The point is, giving up observer unitarity runs very afoul of quantum mechanics, and is really very radical (it requires very detailed conspiracies in the microphysics to make all these violations small and unobservable)
     
  16. Aug 24, 2013 #15

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Thanks Haelfix,
    Your objections here remind me of those offered by "wolfgang". :smile:

    ==quote==
    wolfgang says:
    August 24, 2013 at 9:20 am
    @Lee
    ...
    One problem of simply allowing a non-unitary evolution in the exterior is that
    particle physics becomes non-unitary as soon as you (have to) include virtual black holes.

    Lee Smolin says:
    August 24, 2013 at 10:01 am
    Wolfgang,

    That is not a convincing argument and it is partly addressed in the paper I mentioned. The basic point is that there is no reason one has to include contributions from “virtual black holes.” When one looks at it carefully it becomes not at all clear what would be meant by that in a well defined background independent formulation of quantum gravity. The intuition that any process should have large or even divergent contributions from “virtual black holes” is based on an incorrect use of effective field theory, as discussed in section 4 of the paper with Hossenfelder I mentioned above.

    Another reason is that there is no reason to think that horizons make sufficient sense in terms of quantum geometry at the Planck scale to give meaning to the semiclassical intuition of a virtual or Plank scale black hole. If quantum geometry is discrete at Planck scales then there are no horizons, curvatures or singularities at those scales and no way to give meaning to a Planck scale black hole. There is no contradiction in believing that quantum gravity is simply unitary at small scales while real astrophysical black holes create baby universes.

    Thanks,

    Lee
    ==endquote==
    http://www.math.columbia.edu/~woit/wordpress/?p=6208&cpage=1#comment-159265
    The paper Lee cited is arXiv:0901.3156
     
  17. Aug 24, 2013 #16

    jimgraber

    User Avatar
    Gold Member

    To reduce it to a slogan: Give up Black Holes, Save Unitarity
    Baby Universes were mentioned, but I think the idea of replacing black holes deserves much more serious study than it has received. The problems discussed at the recent “Black Holes Complementarity Fuzz or Fire?” workshop emphasize only some of the serious problems with the black hole concept. I realize the workshop was predicated on acceptance of the black hole concept as well as the general accuracy of the Hawking radiation paradigm, but neither of these ideas has a strong experimental support. Gravitationally collapsed objects similar but not identical to black holes could alleviate many issues if they were unitary and did not involve either horizons or singularities.
     
  18. Aug 24, 2013 #17

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Hi Jim,
    The part from "I realize the workshop was predicated..." onwards was quoted by Scott Aaronson earlier, as if you posted in the Woit thread. But I could not find your post there, so it may have somehow gotten erased.

    Lee and Sabine Hossenfelder present what they are discussing in much more conservative terms
    http://arxiv.org/abs/0901.3156
    Don't get rid of black holes. KEEP black holes but think about what could be happening at the singularity.
    IOW make the least possible disruption in existing theory that will resolve the info puzzle.
    Conservative solutions to the black hole information problem
    Sabine Hossenfelder, Lee Smolin
    (Submitted on 20 Jan 2009)
    We review the different options for resolution of the black hole loss of information problem. We classify them first into radical options, which require a quantum theory of gravity which has large deviations from semi-classical physics on macroscopic scales, such as non-locality or endowing horizons with special properties not seen in the semi-classical approximation, and conservative options, which do not need such help. Among the conservative options, we argue that restoring unitary evolution relies on elimination of singularities. We argue that this should hold also in the AdS/CFT correspondence.
    25 pages, 7 figures

    They examine a number of alternatives for a conservative resolution of this type, not just what I was calling "unibabes" (baby universes). Argue pros and cons and so on.
     
  19. Aug 24, 2013 #18

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Jim,
    I see it now, right before Scott's.
     
  20. Aug 24, 2013 #19

    Drakkith

    User Avatar

    Staff: Mentor

    Good lord I thought this was spam when I saw the title, Marcus!
     
  21. Aug 24, 2013 #20

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    For shame, Drakkith! You were just letting your imagination run wild :biggrin:
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Fire wallow favors unibabes
  1. Graviton is out of favor (Replies: 14)

Loading...