Cycles of time-Penrose says his cyclic cosmology obeys thermodynamics.

[marcus]

Cycles of time--Penrose says his cyclic cosmology obeys thermodynamics.

Roger Penrose has devised a cyclic cosmology which he sees as not violating the second law of thermodynamics. There are several online videos of him lecturing about it. I'll get some links. I'm curious to know if others follow his argument.

One talk was at Princeton, another at Cambridge, several at Perimeter. Berkeley, George Mason University...

The idea is strange. All [or nearly all] matter decays (if only because it collects in black holes and they evaporate). And after ages of expansion there is [almost]nothing but uniform radiation. With nothing remaining to define scale, he says, this blank featureless world of expanding energy is indistinguishable from the next big bang!

The Perimeter videos are easiest to find online. Here it is:
http://pirsa.org/06090005/
Before the Big Bang: an Outrageous Solution to a Profound Cosmological Puzzle

http://pirsa.org/08090078
Clocks at the Big Bang? Quantum gravity is not what you think!
"It has been a common viewpoint that the process of quantization ought to replace the singularities of classical general relativity by some chaotic-looking structure at the scale of the Planck length. In this talk I shall argue that whereas this is to be expected at black-hole singularities, Nature's true picture of what goes on at the Big Bang is very different, where clocks cannot exist and the conformal geometry is completely smooth."

That is part of his "Outrageous Solution" idea. The endstate of the universe gets so uniform that it is timeless and eventless and actually comes round to reproducing the big bang conditions. But, you say, the scale is so different! Ahah! he says, scale doesn't mean anything any more. Geometry is purely conformal or scale-less.

He explains this not only in video lectures but also in a book coming out this month in UK.
http://www.amazon.com/dp/0224080369/?tag=pfamazon01-20

Here's another video where he gives a general-audience version of the talk as part of Perimeter Public Lecture Series

http://pirsa.org/08100081
Before the Big Bang: Is There Evidence For Something And If So, What?
"There is now a great deal of evidence confirming the existence of a very hot and dense early stage of the universe. ... But the information presents new puzzles for scientists. One of the most blatant examples is an apparent paradox related to the second law of thermodynamics. Although some have argued that the hypothesis of inflationary cosmology solves some of the puzzles, profound issues remain. In this talk, Professor Penrose will describe a very different proposal, one that suggests a succession of universes prior to our own. He will also present a recent analysis of the CMB data that has a profound bearing on these issues."

[I edited to allow for matter being very nearly but not precisely zero.]

 

[marcus]

The first talk he gave on this cyclic cosmology model was at Cambridge in November 2005. It was not recorded video. But the audio and slides are on line. Longtime Physicsforums member Arivero gave us the link. I'll see if I can did it up.

Yes! This is perhaps the best delivery available online. In November 2005 the topic is fresh for him.
http://www.Newton.ac.uk/webseminars/pg+ws/2005/gmr/gmrw04/1107/penrose/

It is slides and voice, with the slides automatically change as the audio proceeds. It's an effective format because you can concentrate on the slides without any distracting camera-work. You can also select a slide and hear the talk starting from that point.

 

[sheaf]

That is part of his "Outrageous Solution" idea. The endstate of the universe gets so uniform that it is timeless and eventless and actually comes round to reproducing the big bang conditions. But, you say, the scale is so different! Ahah! he says, scale doesn't mean anything any more. Geometry is purely conformal or scale-less.

In order for the physics to be conformally invariant, the universe presumably would have to be full of only zero rest mass fields, such as photons, meaning all the particles with rest mass would have decayed away. This is one bit that, as far as I understand it, relies on some physics we don't yet have ?

 

[marcus]

In order for the physics to be conformally invariant, the universe presumably would have to be full of only zero rest mass fields, such as photons, meaning all the particles with rest mass would have decayed away. This is one bit that, as far as I understand it, relies on some physics we don't yet have ?

I heard him give this talk in Berkeley, almost the same slides as the Cambridge version. This has links to both the video and the slides PDF:
http://www.msri.org/communications/vmath/VMathVideosSpecial/VideoSpecialInfo/3004/show_video
What I understood him to be saying is using only known physics we can already say everything decays to photons in the very very long run, for the following reason.

(It might decay faster than this if we discovered, say, that protons decay, but he did not speculate about such unproven physics, so his scenario is very very slow.)

We know that matter tends to clump and form clusters of objects which orbit each other.
We know that orbits decay, if only by radiating off energy as grav waves. (Binary pulsar observations.) Therefore all matter eventually merges into black holes, if nothing else happens to it first.

We accept that black holes decay to photons etc. (stuff with zero rest mass). Therefore all matter eventually decays into a soup of photons. QED.

=====================

He may have better arguments and quicker paths to tell us about in his new book. I heard him talk in March 2006. He didn't mention this, I think, but it is a textbook fact (e.g. Weinberg "Cosmology") that the expansion of the universe drains momentum from matter. This is how dark matter is able to cool and curdle into clouds. This is how neutrinos which are currently moving at relativistic speeds will eventually be slowed down.

Sheaf this may be already familiar to you, and you may like to think of it differently, say as successive doppler effects shifting coordinates from frame to frame to frame. People think of the cosmological redshift in different ways. In case anyone reading this is not familiar with it, i want to point out that the redshift, or reduction in momentum, applies to other stuff besides light, that can have nonzero rest mass like neutrinos or putative dark matter.

When dark matter falls together into higher density regions, gravitational energy gets converted to kinetic energy which needs to be canceled somehow. It may be in the form of a portion of the cloud which gets ejected with higher than escape velocity. Whatever the form, expansion can deprive stuff of momentum (relative to Background, or Hubble flow) and cancel kinetic energy. It is analogous e.g. to the redshifting of the CMB photons. Think of the fast-moving matter as "redshifted" by expansion, by being slowed down, by the same amount as photons are. CMB photons have lost about 999/1000 of their original momentum.
Or likewise their energy, their temperature.

=========================
Sheaf, here is an inconsistency. I went back and listened to the first talk he gave on this, Cambridge November 2005. And he DID say he had to assume some unknown physics. Your post is in this sense RIGHT.
I think it was around slide #22, or at least somewhere in the range #22-27.
He mentioned protons decaying. And he said he had asked some physicists about the possibility of electrons decaying to something with mass zero (!). He was, at that point, quite forthright about how speculative the idea was, and not trying to "get around" the problems.

But from later talks, and this may be a mistake of mine, I get the impression that he is saying that if you wait long enough ALL matter clumps into black holes and these then evaporate to radiation. So either I misunderstood, or the idea gradually changed somewhat after the November 2005 presentation.

 

[MTd2]

Everything will decay to photons, but as little the energy the photon might has, it will still self interact and yield all possible particles for a brief period of time. Fields will exist forever and vacuum will still generate all possible particles spontaneously all the time.

It doesn't consider that photons might clump somewhere and form chunks of matter by sheer chance.

 

[marcus]

If anyone is interested in technical details of Penrose' argument, or his cycic cosmological model, I would suggest getting his book:
http://www.amazon.com/dp/0224080369/?tag=pfamazon01-20
You can order it from UK amazon already, though official release is not scheduled until about two weeks from now.

He is argument makes a distinction between two kinds of curvature: Weyl curvature versus Ricci curvature. It's a technical point that I'd like to understand better.

One point I like very much is what he says around slide 27 and 28 of the Cambridge 2005 talk. Mass is what allows clocks. A photon can't distinguish between its beginning and its end. His vision of the very very distant future is that it is boring but there is nothing around that can experience the boredom---because photon's don't experience time.

(He thinks partly visually and partly empathically like this, it seems.)

So he says his very very distant future becomes timeless and merges (in a sense very naturally) into the next big bang moment.
=================

He also talks about the Weyl curvature being constrained by his model. For things to work, it does not necessarily have to be zero but it has to be very small, within certain bounds. Weyl curvature is volume-preserving tidal distortion. In a matterless world, residual gravity waves could still cause tidal distortion. His "Weyl curvature hypothesis" is that this is somehow restricted, I think. Constrained to be below some level.

Then he talks about predictions the model makes. Perhaps related to what I just mentioned. Features that might be observed in the CMB. In the imprint of ancient gravitational fluctuations. As I recall this is also around slides #22-27 of the Cambridge talk.

In the Q&A after the talk, Abhay Ashtekar has some questions, also Neil Turok speaks up.
I will repost the link to the Cambridge talk.
http://www.Newton.ac.uk/webseminars/pg+ws/2005/gmr/gmrw04/1107/penrose/

 

[sheaf]

He is argument makes a distinction between two kinds of curvature: Weyl curvature versus Ricci curvature. It's a technical point that I'd like to understand better.

Just in case anyone here hasn't seen it, I've always liked Baez' one-sentence Ricci/Weyl illustration !

http://math.ucr.edu/home/baez/gr/ricci.weyl.html"

 

[bcrowell]

We know that matter tends to clump and form clusters of objects which orbit each other.
We know that orbits decay, if only by radiating off energy as grav waves. (Binary pulsar observations.) Therefore all matter eventually merges into black holes, if nothing else happens to it first.

We accept that black holes decay to photons etc. (stuff with zero rest mass). Therefore all matter eventually decays into a soup of photons. QED.

I haven't viewed the videos yet, just looked through the slides online, so maybe I shouldn't comment yet, but this argument seems to me to have problems, because of the cosmological constant. If a particular hydrogen molecule is sitting in one of the voids between superclusters right now, then it seems almost certain to me that it will never participate in gravitational collapse to form a star. The rate at which cosmic expansion is accelerating is so great that the thin cloud in which that molecule lives will become diluted far too quickly to allow star formation. Therefore that molecule will never be recycled through a black hole.

In order for the physics to be conformally invariant, the universe presumably would have to be full of only zero rest mass fields, such as photons, meaning all the particles with rest mass would have decayed away. This is one bit that, as far as I understand it, relies on some physics we don't yet have ?

This slide seems to say that he considers that new physics to be a prediction of his theory: http://jessica2.msri.org/attachments/penrose1/pages/penrose-bigbang0034.htm He's predicting the existence of massless charged particles. I'm having a hard time seeing how that is possible, though. E.g., when gamma rays interact with matter, we get e+e- pairs. If there were charged particles x with lower mass than an electron, wouldn't we observe x+x- with higher probability than e+e-?

I'm ignorant of the conformal stuff, but IIRC I've seen this issue posed elsewhere in less technical language. Basically his argument is that after a certain time, there will be no more clocks or observers, because you can't make clocks or observers out of massless particles. I assume this maps somehow onto the conformal invariance idea?

 

[sheaf]

I'm ignorant of the conformal stuff, but IIRC I've seen this issue posed elsewhere in less technical language. Basically his argument is that after a certain time, there will be no more clocks or observers, because you can't make clocks or observers out of massless particles. I assume this maps somehow onto the conformal invariance idea?

Yes that's right - the proper time of a zero rest mass particle is zero so it can't be used as a clock. Zero rest mass field equations are invariant under the conformal group (all 15 parameters worth !).

 

[qsa]

I have already read some theory like that in Scientific American ,oh, maybe 10 years back. Where as the universe expands in trillion trillion trillion years each particle will become as big as our present solar system and it evaporates like a black hole and the unuverse becomes cold and featureless. That article was basically abou the fate of the universe.I will try to dig that article from the internet or my overloaded shelf.

 

[qsa]

ok, correction. the article is about 10 years old but it was on the cover of Time and not Scientific American. Here is the link to that article.(see under science)

http://www.time.com/time/magazine/0,9263,7601010625,00.html

 

[marcus]

I see what you were talking about---orbital collections stuff, not actual atoms but vaguely analogous because outer stuff going around inner stuff. Should not call his idea "atoms".
It is on page 8 of the article.

I don't know if anybody still thinks about that idea. It was just a speculation by some astrophysicist named Adams. We could look him up and see if he is still trying to promote this idea.

Maybe he has gotten some other people to publish about it! Who knows?
==quote Time magazine==
But that's not the end, according to University of Michigan astrophysicist Fred Adams. An expert on the fate of the cosmos and co-author with Greg Laughlin of The Five Ages of the Universe (Touchstone Books; 2000), Adams predicts that all this dead matter will eventually collapse into black holes. By the time the universe is 1 trillion trillion trillion trillion trillion trillion years old, the black holes themselves will disintegrate into stray particles, which will bind loosely to form individual "atoms" larger than the size of today's universe. Eventually, even these will decay, leaving a featureless, infinitely large void. And that will be that--unless, of course, whatever inconceivable event that launched the original Big Bang should recur, and the ultimate free lunch is served once more.Read more: http://www.time.com/time/magazine/article/0,9171,1000170-8,00.html#ixzz0z5EJpcTK
==endquote==

 

[bcrowell]

I viewed the version of the talk that Marcus linked to and took the following notes, which I thought I'd post here in case they'd be useful to others. At the end of the talk, he says that the model predicts a massless neutrino, and decay of the electron into a massless charged particle. I still don't understand how a massless charged particle could exist without our having detected it in, e.g., pair production by low-energy photons interacting with matter.

http://www.Newton.ac.uk/webseminars/pg+ws/2005/gmr/gmrw04/1107/penrose/
cyclical universe
  A. Friedmann 1922
    spherical universe, a(t) is cycloid
  Wheeler
    fundamental constants change each time; solves fine tuning problem
  Smolin
    CNS (cosmic natural selection)
  Steinhardt & Turok
    colliding D-branes
two problems
  thermodynamics
  geometrical: not easy to fit singularities together
thermodynamics
  picture phase space as built of "coarse-graining cells," like a foam
             of bubbles, which are
             hard to distinguish; e.g., states of the room in which all
             that differs is the arrangement of the air molecules;
             S=k log V, where V is volume; system evolves on path that wiggles around, sometimes
             crosses boundaries
  probability argument says that predicting into the future, should expect
             S to increase; problem with this argument is that it also
             retrodicts that entropy went up in the past; so there must
             really be a constraint, that initial entropy was low
  "To ensure that the entropy continues to go down in the past, we need an
             enormous constraint on the space-time geometry at the Big Bang,
             conformal curvature approximately 0.
             Why on the geometry? Because the matter itself was apparently
             in a thermal state (=maximum entropy) to get agreement with
             observation. How do we estimate the contribution of the geometry?
             How enormous is (was!) this constraint?"
  shows CMB Planck curve; paradox: blackbody curve shows maximum entropy, so how can entropy
             have been max if it's been increasing ever since?
  three possibilities, with Lambda>0: K=0, K>0, K<0; since Lambda>0, no Big Crunch, models
             only differ in spatial geometry; pictures with initial singularities, also
             black hole singularities forming later
  gas in a box equilibrates with particles becoming unclumped
  gravitating bodies equilibrate with particles becoming more clumped,
             maximum entropy is black hole
  real universe has these two competing tendencies
  since CMB is Planck curve, matter d.f. must have been equilibrated,
             but not gravitational d.f.
  example: we don't get energy from sun, we get entropy flow;
             we absorb low-entropy visible-light photons,
             emit high-entropy ones into sky; key point is that
             sun is there; if had uniform temp across sky,
             we couldn't run this heat engine; this nonuniformity
             is because sun formed by gravitational clumping
  "time-asymmetry in spacetime, singularity structure, and the
             second law of thermodynamics imply that the BB must
             have been extraordinarily special"
  "singularities in black holes...would not be restricted inthat
             way; suggests that ...quantum gravity must involve
             time-asymmetry"
  P's hypothesis: BB singularities
             constrained such that Weyl curvature in the
             neighborhood is constrained to be small or zero;
             (c.f. black hole singularities, which have diverging
             Weyl curvature)
  entropy estimates
    Bekenstein and Hawking, entropy of black hole
             S=A/4, A propto M^2 (k=c=G=hbar=1)
    suppose 10^80 baryons in observable universe
    radiation and matter at time of decoupling: S=10^88
    now, from supermassive b.h., S=10^101
    all baryons in one big black hole, S=10^123
    10^(10^123)/(10^(10^101))=10^123 (counterintuitively),
             so probability of our current universe (or
             of any low-entropy universe) is about 10^(-10^123)
geometry
  Weyl curvature
    equivalence principle
    equivalence principle is approximate, and the error in the
             approximation is the curvature
    Riemann = Weyl (volume-preserving) + Ricci (volume-reducing)
    (to be more accurate, should state volume interp in terms
              of light rays, not world-lines of matter)
    Weyl=C_abcd, Ricci=R_ab
  inflation
    "why is inflation not adequate to explain flatness and uniformity?"
      "push out horizons so that thermalization can equalize
                 temperatures in different directions"
      "thermalization increases the entropy => the state must
                 have been even *more* special, before, than
                 without thermalization"
      "makes problem worse"
      "expand out small smooth region of initial state to get almost
                flat large-scale universe"
      "only works for initial states that are not fractal-like.
                but generic initial states are expected to be
                fractal-like (consider general collapsing
                universe)"
    horizon problem
      conformal diagrams
        standard cosmology
          BB is horizontal line at t=0
          how can regions be uniform when they've never previously
                      been in causal contact
      inflationary cosmology
        BB gets pushed back to farther before decoupling of
                   light and matter
        introducing a thermalization process makes the specialness
                  of the initial BB state *more* acute
  strict conformal diagrams (Penrose 1962, Carter 1966), discussed
               in Road to Reality
    vertical lines are spherical symmetry axes
  assume that all b.h. will eventually evaporate (some people
             argue that they might leave behind remnants)
  Weyl curvature hypothesis: universe could be extended as a conformal
             space through the BB (Paul Tod)
  older mathematical trick: treat infty as something you can
             extend through (e.g., gravitational radiation,
             conformal factor, pretend bdy at infinity that
             represents the future)
  conformal diagrams with Lambda=0 have symmetry, but
              those with Lambda>0 don't
  crazy idea: remote future is conformally equivalent to
            the next big bang;  the geometry
            is not full metric geometry, but only conformal
            geometry; has light cones, but no metric
  metric has 10 d.f.; conformally, you just have one of these
            that goes wild
  only works if:
    - only massless entities in latest stages
    - Weyl curvature=0 in BB
  "in the very remote future, only massless, conformally
              invariant entities survive, so only conformal
              structure retained"
  Dyson, paper on very remote future, argue that life could
              still exist
    Dyson, "Time without end: physics and biology in an open universe,"
             RMP 51 (1979) 447 (?) (Lambda=0)
    Krauss and Starkman, "The fate of life in the
               universe," Sci Am 281 (1999) 58 (Lambda ne 0)
  can't make clocks out of photons; universe forgets passage
            of time, can be treated as a conformal manifold,
            i.e., spacetime in which you have invariance:
    under a rescaling fo the metric g -> Omega^2 g,
            C_{abc}^2 -> C_{abc}^d
            C_{abcd} -> Omega^2 C_{abcd}
    lose ticking of clocks, but still have light-cones
  depends on everything decaying into massless particles
    is it possible that electron decays into massless charged
                   particle? "in some pictures it actually
                   fits in nicely, in others not so nicely"
   "it would be nice to have a massless neutrino" (we only
                know they have mass differences)
    P's picture makes predictions about particle physics
  gravitational radiation will still leave its mark on the
                 future, but it scales away in the future
  although radiation doesn't lead to a zero Weyl curvature,
               it does leave a derivative of the Weyl curvature;
               this is another prediction of the model
Q&A
  1
    at BB, metric is rescaled by an infinite factor; phase
           space gets rescaled infinitely
    BH isn't actually maximum entropy, it's evaporated black holes
  2
    could be something like large-small dualities of string theory
    idea needs to be mathematically developed, reformulated
                somehow
  3
    Gaudi universe
    Veneziano model

 

[bcrowell]

Here is a conference proceeding in which Penrose has written up the idea: www.jacow.org/e06/PAPERS/THESPA01.PDF

 

[atyy]

Here is a conference proceeding in which Penrose has written up the idea: www.jacow.org/e06/PAPERS/THESPA01.PDF

Thanks! Some aspects of Penrose's ideas are applied in a stringy context by

http://arxiv.org/abs/0711.1656
The Arrow Of Time In The Landscape
Brett McInnes

"Our emphasis on isotropy is motivated by Penrose’s “Weyl Curvature Hypothesis” [7]. Penrose postulates that the initial geometric “specialness” takes the form of a demand that a certain piece of the spacetime curvature tensor, the Weyl tensor, should vanish at initial singularities (and only there)."

 

[bcrowell]

Hi, atyy -- Cool paper. Love to see a nice, long, hard paper with almost no equations :-)

Here's something that bothers me about Penrose's idea. He says the universe can only proceed to its rebirth if all massive particles are first converted into massless particles. But regardless of whether we're talking about a pseudo-Riemannian geometry or a conformal geometry, light-cones are still well defined, and therefore we still have a notion that different regions of spacetime can become causally disconnected because of the accelerating expansion caused by the cosmological constant. So if there were only one hydrogen molecule remaining in the entire universe, how would that hold up the reincarnation process? That molecule is only causally connected to its own little piece of spacetime. How can it have a global effect? I guess the argument gets a little complicated because although causal connection of particles is transitive, causal connection of regions of spacetime is not.

 

[atyy]

Here's something that bothers me about Penrose's idea. He says the universe can only proceed to its rebirth if all massive particles are first converted into massless particles. But regardless of whether we're talking about a pseudo-Riemannian geometry or a conformal geometry, light-cones are still well defined, and therefore we still have a notion that different regions of spacetime can become causally disconnected because of the accelerating expansion caused by the cosmological constant. So if there were only one hydrogen molecule remaining in the entire universe, how would that hold up the reincarnation process? That molecule is only causally connected to its own little piece of spacetime. How can it have a global effect? I guess the argument gets a little complicated because although causal connection of particles is transitive, causal connection of regions of spacetime is not.

Although McInnes's view is not cyclic, I think there's an analogous discussion in his section 3.2, in which he discusses mechanisms in which the mother universe will be smooth enough that "A baby which is born at this point has at least some hope of being born with a geometric entropy as low as that of the early stages of our Universe."

McInnes hopes to use something like Carroll and Chen's proposal "Carroll and Chen, however, propose another way to scatter the Local Group to the winds. They ask us to wait until essentially all of the matter in these galaxies has fallen into black holes, and then to wait for these black holes to evaporate. Presumably the resulting radiation will cease to be bound. In effect, this process contracts the “observer” from the size of the Local Group to smaller and smaller scales. Eventually, one might hope, the “observer” will become so small that the smoothing effect of cosmic acceleration can operate over the tiny length scales on which baby universes nucleate."

That is very similar to Penrose's discussion that black holes are not a concern since they evaporate "There are certain important assumptions involved in CCC, in order that only conformally invariant entities survive to eternity . One of these is that black holes will all eventually evaporate away and disappear. This evaporation is a consequence of Stephen Hawking’s quantum considerations, and these are now normally accepted. There is, however, the issue (connected with the so-called “information paradox”) of whether they would actually ultimately disappear or leave some form of “remnant”. I am here taking the more conventional view that they would indeed disappear in a final
(cosmologically very mild) explosion."

So I guess McInnes, Carroll and Chen would try to solve Penrose's electron problem by having it fall into a black hole?

 

[bcrowell]

Interesting post, atyy.

So I guess McInnes, Carroll and Chen would try to solve Penrose's electron problem by having it fall into a black hole?

Regardless of whose baby-universe/cyclic-universe theory we have in mind, I think known astrophysics leads to certain conclusions about the universe in the very far future. A lot of matter will get recycled through black holes and turned into Hawking radiation, but that process definitely will not run to completion, because even the voids between superclusters are not a perfect vacuum. They have a lot of hydrogen molecules in them, those molecules are not gravitationally bound to anything, and therefore the accelerating expansion of the universe will cause them to become causally disconnected from any other matter that could be sufficient, when collected, to form a black hole. I think Penrose realizes this, and this is why he says that a prediction of his model is some nonstandard particle physics, such as the decay of electrons.

I think the difficulty with Penrose's idea relates to this part of the McInnes paper:

Eventually, one might hope, the “observer” will become so small that the smoothing effect of cosmic acceleration can operate over the tiny length scales on which baby universes nucleate. [quote from McInnes]

The difference here is that McInnes has a length scale available, and Penrose doesn't. Penrose's process involves a universe that loses its ability to define scales of time and length, going over into conformal geometry. Once this happens, he can claim that an extremely dilute far-future universe full of zero-rest-mass particles is the *same state* as the the extremely dense universe soon after a Big Bang singularity, in which the temperature is so high that every particle's rest mass is negligible (compared to its total mass-energy). Therefore his theory can't have any length scale available. He can't say that a region of size L spontaneously produces a new universe if it remains free of massive particles for time T, because he needs to go through a conformal-geometry phase in which scales like L and T become meaningless.

I think this must be why he constructs a picture in which the recycling of the universe is global, rather than local as in baby-universe theories. It can't be local if there is no scale to define how local is local. But if it's global, then it seems to me to have serious logical problems, because a single remaining hydrogen molecule is sufficient to prevent the recycling process from proceeding in regions from which it's causally disconnected.

A possible alternative would be to say that recycling is local, and it happens whenever the universe progresses to the point where there exists some future light-cone that is empty of massive particles. Since light cones are conformally invariant, this might compatible with Penrose's ideas. The problem is that we then have to talk about recycling some finite region of space, rather than the entire universe. But causal connection between regions is not transitive. That is, if region A contains points that are causally connected to points in region B, and B contains causal connections to C, it does not necessarily follow that A is connected to C. Therefore I don't think conformal geometry, which only has light-cones, not distances, can define any criterion for how big the "mother" region should be.

 

[marcus]

...A lot of matter will get recycled through black holes and turned into Hawking radiation, but that process definitely will not run to completion, because even the voids between superclusters are not a perfect vacuum. They have a lot of hydrogen molecules in them, those molecules are not gravitationally bound to anything, and therefore the accelerating expansion of the universe will cause them to become causally disconnected from any other matter that could be sufficient, when collected, to form a black hole. I think Penrose realizes this, and this is why he says that a prediction of his model is some nonstandard particle physics, such as the decay of electrons...

As I see it, this argument is fairly persuasive for putting interest in conformal recycling on indefinite hold.

Again just from my personal perspective, there is still puzzle (or bewildered interest) surrounding what Penrose calls "The Basic
Conundrum" and "The Enormity of the Specialness". I refer to page one of this paper
http://accelconf.web.cern.ch/AccelConf/e06/PAPERS/THESPA01.PDF

Crowell, I'm glad you posted that link to Penrose' paper. It's easier to study than the slides+audio presentation.

=========================
Perhaps we can view conformal recycling cosmology as Penrose' response to "The Basic Conundrum" that he perceives and introduces at the outset. Maybe his solution (requiring gradual decay of the electron's mass to zero) is not ripe for consideration but we can still consider the motivating problem.

He puts the 2nd Law in a way convenient to his train of thought as follows: roughly speaking specialness diminishes over time. The state of a physical system evolves so as to get less and less special. So if you look back in time, its state must always have been more special than it is at present. This formulation may seem a bit imprecise but let's try to follow what he says on page 1 of the paper:

== quote Penrose page 1 ==

...Sometimes theorists have tried to find an explanation via the fact that the early universe was very “small”, this smallness perhaps allowing only a tiny number of alternative initial states, or perhaps they try to take refuge in the anthropic principle, which would be a selection principle in favour of certain special initial states that allow the eventual evolution of intelligent life. Neither of these suggested explanations gets close to resolving the issue, however.

It may be seen that, with time-symmetrical dynamical laws, the mere smallness of the early universe does not provide a restriction on its degrees of freedom. For we may contemplate a universe model in the final stages of collapse. It must do something, in accordance with its dynamical laws, and we expect it to collapse to some sort of complicated space-time singularity, a singularity encompassing as many degrees of freedom as were already present in its earlier non-singular collapsing phase. Time-reversing this situation, we see that an initial singular state could also contain as many degrees of freedom as such a collapsing one. But in our actual universe, almost all of those degrees of freedom were somehow not activated.

==endquote==

 

[bcrowell]

As I see it, this argument is fairly persuasive for putting interest in conformal recycling on indefinite hold.

I think Penrose sees the requirement of nonstandard particle physics as being fun, an indication that his idea is "crazy enough to be right," and also an indication that it is testable -- which is important, because it's not obvious what else about the idea *is* testable.

It's also possible that Penrose's idea is partly right, but needs to be mixed up with some of the other ideas that are floating around in the gene pool.

The 2004 paper by Carroll and Chen that atyy linked to ("Spontaneous Inflation and the Origin of the Arrow of Time," http://arxiv.org/abs/hep-th/0410270 ) claims that eternal inflation works as a way of explaining baby universes and the arrow of time. Penrose, however, explicitly says in his 2005 talk that he does *not* think inflation can be made to do this. Here is a comment criticizing some aspects of the Carroll-Chen paper: http://arxiv.org/abs/hep-th/0411115 As far as I can tell, the real problem with the Carroll-Chen approach is that they have to assume some nonstandard physics, which is that a given causal patch of spacetime has infinitely many degrees of freedom. I don't know if this particular objection is the one that Penrose has in mind.

 

[marcus]

Crowell, you remember the Carlip paper summarizing how several approaches to QG converge on the idea of "spontaneous dimensional reduction" with decreasing scale.

My take on this is that it is possible that degrees of freedom are killed off or numbed by crunching the scale.

I think the key point in Penrose' paper is where he makes an unjustified assumption that DoF survive unaffected by shrinking to Planck scale. His argument for a "Basic Conundrum" (motivating his cosmology idea) fails, unless this unjustified assumption is granted:

==quote Penrose page 1==

... a singularity encompassing as many degrees of freedom as were already present in its earlier non-singular collapsing phase. Time-reversing this situation, we see that an initial singular state could also contain as many degrees of freedom as such a collapsing one. But in our actual universe, almost all of those degrees of freedom were somehow not activated.
==endquote==

Carlip said, in a talk I heard him give, that in the absence of empirical evidence you could sort of superstitiously regard the convergence of very different theoretical approaches giving similar results as suggestive semi-evidence. And he provided a classical analysis showing something like (or vaguely analogous to) spontaneous dimensional reduction at a classical singularity.

My take is, if the very dimensionality of space gets suppressed or gradually turned off as you crunch, then why not DoF more generally?
And time-reversing that, why couldn't DoF start off being radically suppressed and gradually become excited, or alive, as expansion gives them breathing room?

He acts like there is a fixed number of DoF regardless of scale. If I refuse to grant him that, it seems to derail his big conundrum.
Please, let me know if this reaction strikes you as naive or foolish.

 

[marcus]

Crowell,
I'll try to say that a different way.

1. If Penrose sees a "Big Conundrum" involving how he counts degrees of freedom, that might just point to there being problems with our conception of degrees of freedom.
He might be counting wrong, based on an unstated/unquestioned assumption about behavior at Planck scale.

2. Several QG (asymsafe, causal dynamical triangulations, and others according to Steve Carlip) point to spatial dimensionality going continuously down from 3D to 1D at very small scale. If that can happen to something as basic as spatial dimension, there are likely to be conceptual problems with our conventional DoF idea.

 

[MTd2]

I will repeat the question: what about the virtual particles?

 

[marcus]

Just to make my point of view clear. We are discussing this paper:
http://accelconf.web.cern.ch/AccelConf/e06/PAPERS/THESPA01.PDF

And the only part of the paper that interests me (after what Crowell and others had to say) is section I "The Basic Conundrum".

A "conundrum" is a puzzle. Because of what Crowell said (and if I remember right some additional comments by Sheaf and/or MTd2) I think Penrose' solution to the puzzle is too speculative, too far-fetched, for me to want to study it further. His solution is the conformal cosmic recycle. For me, this thread is no longer about that.

But the puzzle itself is quite interesting. He has a one-page argument that, for instance, inflation "only makes worse" the contradictions it is expected to resolve. He argues in effect that the conventional idea of the big bang is thermodynamically invalid, or that the initial state must have been incredibly "special".

If anyone would like to clarify Penrose' argument, and improve on my paraphrase of it, please do!

 

[bcrowell]

But the puzzle itself is quite interesting. He has a one-page argument that, for instance, inflation "only makes worse" the contradictions it is expected to resolve. He argues in effect that the conventional idea of the big bang is thermodynamically invalid, or that the initial state must have been incredibly "special".

I'm not seeing such an argument in the Penrose paper, but I think this paper makes it:

Lisa Dyson, Matthew Kleban, Leonard Susskind, "Disturbing Implications of a Cosmological Constant," http://arxiv.org/abs/hep-th/0208013

This stuff is way outside my area of expertise, but the rough impression I get from reading these papers is that nobody understands for sure how to count states.

My own feeling about these fine-tuning arguments has always been that there is no physical principle that predicts the initial state of the universe, so there was never anything to be explained scientifically about the specialness of the universe's initial state. If it was incredibly fine-tuned, then it was incredibly fine-tuned. Hence no need for inflation, worries about the horizon problem, etc.

If I had already been convinced that the universe extended infinitely far into the past, then I would be more convinced there was a problem with fine-tuning, since flatness and smoothness are unstable. No amount of fine-tuning will allow a pencil to balance on its tip for an *infinite* amount of time.

But since the best evidence is that the universe doesn't extend infinitely far into the past, I don't see a problem with the idea that the Big Bang was surprisingly fine-tuned. It *is* possible to balance a pencil on its tip for any *finite* amount of time.

-Ben

 

[marcus]

Penrose' argument may be wrong, I'll grant you. But he makes it without assuming an infinite past.

I don't mean his cyclic cosmology scenario, which does indeed go back in time indefinitely. I mean the argument on page 1 of the paper where he is claiming there's something puzzling, some contradiction.

Maybe it's possible to just shrug the alleged contradiction off, or say it's just another version of the "arrow of time" problem. I don't want to get into the position of trying to defend Penrose---I'm not a believer and even if I were I'd make a mess of it

My personal view (which no one else should feel urged to share) is that he is onto something interesting. Studying the conventional big bang model, and inflation variants on it, from a thermodynamic PoV may well raise contradictions like he says, and force some kind of response (not necessarily the one he has in mind.)

I'll see if he says anything about it in Road to Reality, and I'll check out the new book Cycles of Time (which I think goes on sale later in the US.)

 

[bcrowell]

My personal view (which no one else should feel urged to share) is that he is onto something interesting. Studying the conventional big bang model, and inflation variants on it, from a thermodynamic PoV may well raise contradictions like he says, and force some kind of response (not necessarily the one he has in mind.)

I'll see if he says anything about it in Road to Reality, and I'll check out the new book Cycles of Time (which I think goes on sale later in the US.)

I find it fun to think about because (a) it's very pretty, and (b) it's purely classical, so I feel like I might be able to figure something out about it based on my understanding of classical GR, without having to resort to my (totally nonexistent) knowledge of current research in quantum gravity.

It's frustrating that he doesn't seem to have written a full exposition aimed at physicists. I suppose I'll buy the book when it comes out in the U.S., but I'll be frustrated if it doesn't do anything more than express the schematic picture he gave in the 2005 talk, padded out with a ton of preliminary explanation for non-physicists.

 

[atyy]

I don't really understand the conformal hypothesis. As bcrowell says there are still light cones. And people do use CFTs to study black hole entropy, where we think the normal second law of thermodynamics still applies.

http://arxiv.org/abs/1008.3439
Measuring Black Hole Formations by Entanglement Entropy via Coarse-Graining
Tadashi Takayanagi, Tomonori Ugajin

http://arxiv.org/abs/0811.4393
CFT Duals for Extreme Black Holes
Thomas Hartman, Keiju Murata, Tatsuma Nishioka, Andrew Strominger

http://arxiv.org/abs/1006.1902
Wilsonian Approach to Fluid/Gravity Duality
Irene Bredberg, Cynthia Keeler, Vyacheslav Lysov, Andrew Strominger

http://arxiv.org/abs/1006.3675
A holographic view on physics out of equilibrium
Veronika E. Hubeny, Mukund Rangamani

 

[oldman]

...I think the key point in Penrose' paper is where he makes an unjustified assumption that DoF survive unaffected by shrinking to Planck scale. His argument for a "Basic Conundrum" (motivating his cosmology idea) fails, unless this unjustified assumption is granted ...

My take is, if the very dimensionality of space gets suppressed or gradually turned off as you crunch, then why not DoF more generally?
And time-reversing that, why couldn't DoF start off being radically suppressed and gradually become excited, or alive, as expansion gives them breathing room?

He acts like there is a fixed number of DoF regardless of scale. If I refuse to grant him that, it seems to derail his big conundrum.
Please, let me know if this reaction strikes you as naive or foolish.

Although you are here replying to Crowell, I can't resist butting in with this comment:

The assumption that Penrose makes, namely that that the universe's abstract phase, state or Hilbert space (call it what you will) together with the 'coarse graining' and 'degrees of freedom' associated with this space, survive unaltered throughout the universe's amazingly complicated history, seems to me right off the wall. I've said so in the cosmology forum in the thread "Do changes of spacetime geometry affect entropy?" . So your reaction sounds to me like common sense, not naive or foolish at all.

 

[marcus]

I just checked to see how Penrose' book was doing in the UK (it is not yet on sale at Amazon in US).

In the Amazon.co.uk listing it was #324 amongst all books, and it was the #1 bestseller in Astronomy and in Cosmology.

Oldman it's encouraging to know we take a similar view of that. I will get a link to the "Do changes...?" thread in cosmology forum.

Here it is: Do changes in geometry affect entropy?
https://www.physicsforums.com/showthread.php?t=415350

I see that Apeiron contributed some links to writings by Charles Lineweaver that bear on this question. I will have a look at what he found. It is an interesting thread, and one that I somehow missed seeing back in July.

Even though one of Penrose' arguments could have a weak spot (just what you point to in your thread) I am delighted to see that his book is selling well.

http://www.amazon.com/dp/0224080369/?tag=pfamazon01-20

=====EDIT=====
Here is the UK amazon physics bestseller list:
http://www.amazon.com/dp/0224080369/?tag=pfamazon01-20
(Brian Cox' book is currently #1 and Penrose' is #2)

Here is the cosmology bestseller list:
http://www.amazon.com/dp/0224080369/?tag=pfamazon01-20
(Penrose' continues #1, not much competition since it is #336 among all books Amazon sells)

I'm happy to see a "before-the-bang" book achieve such popularity. It will help to break down the unjustified conception of the bang as the "beginning of time".

Another thing that is no doubt helping erode that widely held notion is the recent one-hour BBC program called "What happened before the big bang." George Jones recently posted a complete set of YouTube links for this---6 consecutive ten-minute segments. Much of the program was filmed at Perimeter, but also some at Oxford.

 

[Naty1]

Marcus:

I just checked to see how Penrose' book was doing in the UK (it is not yet on sale at Amazon in US).

yes it is: https://www.amazon.com/s/ref=nb_sb_...oger+penrose,stripbooks,199&tag=pfamazon01-20

I happened to hear part of a Penrose interview on his new book with John Batchlor [late night radio] Wednesday evening [9-26] on WABC radio. Unfortunately I fell asleep...so missed most of it... and was unable to find a recording when I just looked this morning...So I checked to see if his book was available ..for about $10 seems like fun...
The interview may have been 45 minutes or so...aimed at a non tech audience...I'll look again to see if I can find a recording...

 

[Naty1]

ok, I found the podcast for the Penrose interview:
on this page...

http://wabcradio.com/sectional.asp?id=33447


I'll listen and post my impressions.

 

[Naty1]

Penrose Interview: Cycles of Time [39 minute discussion] with John Batchlor
Conformal Cyclic Cosmology

My synopsis: Penrose believes circular anomalies in the CMBR point to signals from a prior universe, before the Big Bang. Time and scale are lost in the crossover from one eon to another…this is the ‘conformal structure’ where lightcone angles are retained. At the end of one eon with masslessness, black holes evaporate, information is lost, and we return to a low entropy initial state.


CMBR observations…irregularities/deviations may reflect signals from before the big bang. Signals allow cosmological study in great detail and confirm cosmological origins.
Concentric circles observed [not published yet] come from collision of supermassive black holes before our big bang….Penrose claims this is observational evidence.

Second law of thermodynamics: randomness [entropy] is increasing. Black holes have
largest entropy in the universe [from Hawking, Beckenstein work] . Big Bang entropy was low in terms of gravitational entropy but high in all other aspects.

[11 minutes]
As time goes on cosmological acceleration is increasing…that expansion means colder and colder and emptier. Back holes are colder, but expansion will eventually cause universe to be colder and black holes will disappear via Hawking radiation…black holes disappear in a ‘pop’ and release entropy and DO swallow information….so our notion of entropy changes….must redefine definition entropy….. [explanation not clear]..but second law still works, but definition changes due to swallowing information by black holes…very end is low entropy state…..similar to big bang….

Mass finally fades away is Penrose’s theory…..sort of an anti Higgs mechanism….

[17 minutes]

‘Crossover’ from one eon to another [from one universe to another]…involves dark matter and dark energy….dark matter evaporates at the end says Penrose, but evidence
is not strong….what is crossover mechanism from masslessness of end to masslessness of a new beginning….E = hv…..mass and frequency are equivalent…mass is a clock….
No mass, no notion of time; more freedom, no distance….time and distance [scale]is lost in crossover…..but angles are retained….this is ‘conformal structure’. Null cones define
angles….hot concentrated and cold, empty…. long and short, are equivalent….without rest mass can rescale to a new ‘eon’….scale gets lost if rest mass disappears…

No mass at moment of big bang…..any electrons have lost their mass….mostly photons
And gravitons….Penrose’ theory accommodates second law…Friedman Tolman models
Could not…black holes were unknown then…

[27 minutes]
universal constants we have this eon come from the prior eon….restricts possible values…eons propagate repeatedly and sequentially…Penrose does like like Smolin’s approach of new universes forming from individual black hole singularities….


Veneziano’s string theory model is most like Penrose CCC….but Penrose theory is not string based. Evidence of prior existence, or possibility of such, marks Penrose theory.

[32 minutes]
At end of our expansion we come to masslessness and as black holes evaporate and as information is lost, we return to a low entropy state…
How do electrons and positrons lose mass…no evidence either way….
Dark energy….confirms extra term of Einstein [cosmological constant] which Einstein used for the wrong reason….it’s nature, so far, is ‘dark’…unknown…..dark matter must disappear at the end….reappears when the Higgs field reappears….

Temperature variation rings in CMBR….
some warm some cool…black holes relatively close are cool…..warm ones MIGHT have source of signal supermassive BH collisions at the center….from one eon to another….

 

[Naty1]

Turns out the radio show interview is a simplified verson from Marucs first post:

The Penrose presentation:

The Perimeter videos are easiest to find online. Here it is:
http://pirsa.org/06090005/
Before the Big Bang: an Outrageous Solution to a Profound Cosmological Puzzle

That is a wonderful summary of cosmology and the transparencies Penrose shows in the video can be directly copied from the windows play version...

 

[Naty1]

In this current discussion BillK seem to shoot a big hole in the Penrose CCC discussed here:


https://www.physicsforums.com/showthread.php?p=4094118#post4094118
does temperature affect black holes



BillK says that as black holes get smaller, they emit more massive particles! So how the end of the universe can be simply explained by evaporating black holes via radiation doesn't seem so clear...