Cycles of Time

Today’s Wall Street Journal has a review I wrote of Sir Roger Penrose’s new book Cycles of Time. The review is aimed at a much wider audience than this blog, and is the product of substantial editing to get its length down and make it as readable as possible for as many people as possible, so here are some supplementary remarks.

I should make it clear that I’m not at all convinced by what Penrose is proposing. He needs the distant future of the universe to be conformally invariant, and this requires all particles to be massless. As far as we know the electron is completely stable, with unchanging mass, and this will always ruin conformal invariance. Penrose himself notes the problem. For this to be overcome, whatever our ultimate understanding is of how particles get mass must change so that these masses go to zero in the future. It’s also seems to me that the conformal anomaly of QCD will always be a problem, with quantization and the renormalization group always breaking conformal invariance and giving a mass scale, indefinitely far into the future.

The other main problem is the one shared by most “pre-big-bang” ideas: how do you ever test them? Penrose and a collaborator last year created a stir by claiming to see in the CMB patterns of the sort he argues might be expected from black hole decays late in an era before the Big Bang, but it’s not clear there’s a real prediction here, and others who have redone this analysis say they see nothing.

Attempts to get a Big Bang in our future as well as our past generally strike me as motivated by a very human desire to see in the global structure of the universe the same cyclic pattern of death and rebirth that govern human existence. To me though, deeper understanding of the universe leads to unexpected structures, fascinating precisely because of how alien they are to human concerns and experience. Just because we might find a cold, empty universe an unappealing future doesn’t mean that that’s not where things are headed.

The book is in many ways an unusual document. It includes an extensive appendix working out some of the details of the mathematics of his proposal. In some sense he has managed to get a trade publisher to put out a highly technical discussion of a speculative idea inside the covers of a popular book, instead of going the usual route of publishing this in a refereed journal. The only references I can find to other places where he has written some of this up are to chapters in this book and this one, as well as this contribution to a conference proceeding. The technical idea behind this, that the hypothesis of the vanishing of the Weyl curvature in the early universe leads to possible cosmological models that can be extended past the Big Bang singularity he attributes to this paper of K.P. Tod. There’s a nice recent exposition of this by Tod here.

So, I’m not convinced by the speculation about the far future, and for an evaluation of the ideas about extending back through the big bang singularity you’ll need someone more expert about cosmology than me. These topics are very clearly labeled in the book as speculative, without support from other physicists or any experimental evidence. The bulk of the book though is other material providing a background and context for the speculation, and it is this which I think makes it most valuable as a popular book. Penrose is a wonderful, elegant and clear writer, and he covers a lot of ground about physics beautifully here. Most remarkable are the illustrations, by far the best visual representations of a range of important ideas that I know of. Physicists and mathematicians work with lots of internal pictures in their minds representing important aspects of the concepts they are investigating, but very rarely do they have the technical skill to grasp some of the essence of these pictures and get them down on paper. Even more rarely do they make it into wide distribution in print, so I’m glad to see that happen here.

This entry was posted in Book Reviews. Bookmark the permalink.

44 Responses to Cycles of Time

  1. manyoso says:

    Thanks for the review! This will definitely be going on my reading list!

  2. Rich C says:

    Penrose makes another testable prediction in the book though, doesn’t he? He argues that the correlations in the microwave background should not extend past 60 degrees, for reasons that I sort of thought I understood when reading the book but can’t articulate now. As there is a wealth of new data on the CMB coming in now and expected in the near future, wouldn’t this also potentially provide either falsification or support for his speculations?

  3. Philip Gibbs says:

    I don’t see why there would be electrons left. Negative and positive charges are perfectly balanced in the universe as far as we can tell, so if the electron and positron are the only stable charged particles then eventually they will come together and annihilate.

    The theory may have other problems, such as those you mention, but I think it is right for physicists to speculate in this way. It is only by thinking along the lines of such crazy ideas and seeing where they lead that we can figure out what are the consistent possibilities.

  4. Peter Woit says:

    Philip,

    First you have to get all your protons to decay, then, in an expanding universe I don’t see why all electrons and positrons have to find each other. And, if they’re just one electron left over, you’re not conformally invariant. Besides, the same argument could be made with neutrinos.

    To make this work, it seems to me you need that whatever gives mass to fermions does so in a fashion that goes to zero at long times. Since we don’t understand really the origin of these mass terms, in principle it’s possible, but it doesn’t seem likely.

  5. Yatima says:

    From a layperson’s perspective, this review hits the mark.

    Public Service Announcement: Well-grounded discussions about what the universe is, is doing or will be doing shall have to be postponed for an indefinite time, though it’s always fun to speculate.

  6. D R Lunsford says:

    Peter said “He needs the distant future of the universe to be conformally invariant, and this requires all particles to be massless. As far as we know the electron is completely stable, with unchanging mass, and this will always ruin conformal invariance.”

    Well I think the idea is that is gets close enough to conformal invariance that the remaining masses are comparable to fluctuations in the conformal vacuum. It’s almost a way of realizing Mach’s principle. The fluctuation could be thought of as a new Big Bang. I don’t know if this is what Penrose intends, but that was how I rationalized it to myself.

    -drl

  7. D R Lunsford says:

    The other aspect of this, is that at the time scales involved, there is not enough happening to define the scale (that’s the conformal scale) without it being disturbed by fluctuations.

    -drl

  8. the next einstein says:

    ^^^^
    “Penrose is a wonderful, elegant and clear writer, and he covers a lot of ground about physics beautifully here. Most remarkable are the illustrations, by far the best visual representations of a range of important ideas that I know of. ”

    “There is nothing worse than a sharp image of a fuzzy concept.”
    — Ansel Adams

  9. abomb says:

    How it can it be hard to buy the concept that electrons aren’t infinitely stable, .i.e they are only “meta-stable”. We’re already trying to cope with quadratic renormalization problems with theories like super symmetry. How it can be that electrons are genuinely believed to infinitely stable?

    Regardless, given infinite amounts of time, are these electrons really a problem?

  10. ZZZ says:

    I find Penrose’s idea interesting. If the lambda-CDM model holds, each remaining electron positron or neutrino would become isolated in its own private observable universe and the total volume of the universe will become exponentially dominated by voids with no matter. Since physical laws and statements are accurate to only finite number of digits, at some point the universe will become empty, i.e. can be described by models with no particles. So I don’t see the electron that would not die as the fatal flaw in this idea. Rather I don’t know whether vacuum fluctuations continue to supply a time scale. The problem with setting the clock by the vacuum is, what’s that clock for?

    Also Lunsford makes a valid point in that, if the universe is finally cold, i.e. nothing happens any more, then what defines time? If the time between particle interactions increase exponentially as things fly apart and cool down, then the total number of interactions is finite. So then the universe really does have a rather definitive end. Or is the ticking of the clock really a counter of interactions because a clock must have physical reality? Then, provided that the cooling down happens faster than the flying apart, distances would actually seem to shrink, and the universe would collapse at the final tick.

  11. Chris Oakley says:

    The lack of conformal invariance of the real world is also the reason that Penrose’s Twistors have not made serious inroads into physics. The twistor is a fundamental vector in SU(2,2), a double cover of C(3,1), the conformal group in 3+1 dimensions. Interesting mathematically – and more so than the ISO(3,1) subgroup – but the latter, the Poincare Group is at least a known symmetry of nature.

  12. “Physicists and mathematicians work with lots of internal pictures in their minds representing important aspects of the concepts they are investigating, but very rarely do they have the technical skill to grasp some of the essence of these pictures and get them down on paper.”

    Almost any paper by Louis Kauffman (btw, one of the first mathematicians exploring topological quantum computing) is amazing in this respect. See http://www.math.uic.edu/~kauffman/

    Another excellent communicator is Jean-Marie Souriau (geometric quantization and many others), see his latest book “Grammaire de la Nature” (in french) here http://www.jmsouriau.com/

  13. apostolos syropoulos says:

    Intriguing review of an intriguing book!

  14. Rael says:

    But even if electrons completely annihilated with positrons (what positrons, by the way?), as long as there is a gluon field in the vacuum, wouldn’t its dynamics (also known as QCD) dynamically set a mass scale?

    We don’t need to have protons in order to have a natural mass scale of ~1 GeV, only gluons and massless quarks are enough for that.

  15. GeorgeDorn says:

    Hi Peter. Could you comment on this perhaps?

    http://arxiv.org/abs/1105.5283

  16. cormac says:

    Nice review Peter, just the right balance.
    Here’s a morsel of trivia: Mum claims that Penrose seriously considered a permanent position at DIAS in the 1970s. He came over, took on look at Dublin traffic and decided no thanks!

  17. Anon says:

    This book perhaps says more about humans than the universe, specifically our panic, as social primates, at the desolation and loneliness of a wound-down universe.

    To avoid this prediction, we will do what we can to try to force a square peg into a round hole, which is unfortunately what this looks like to me. Something like Kepler’s attempt to relate the orbits of the planets to the platonic solids.

  18. bjm says:

    @Rael: (what positrons, by the way?)

    The ones that would come from proton decay. (See Peter’s post above).

  19. jpd, MD,PhD,MfA, says:

    its not a sphere, the electric dipole moment was found to be very
    very small. for some reason many articles describe this as “spherical electrons”

  20. Philip Gibbs says:

    Peter, Rael, even if the proton is stable at the GUT scale, it is predicted to be unstable through the non-perturbative effects of the sphaleron solution in electro-weak theory. We are talking about very long timescales so any instability is relevant. Another mechanism is the proton falling into a blackhole and any excess charge being reradiated with electrons and positrons most favoured. By whatever means protons will ultimately decay into positrons in equal number of the electrons and the vast majority (if not all) will pair off and annihilate. I think most (or all) neutrinos will be captured in black holes too.

    It is true that electrons could become isolated in regions separated from positrons due to the horizon effect of the cosmological constant. This will be rare because once the particles slow down they will tend to attract. Isolated regions will be quite large with a scale determined by the cosmological constant. In this case there will still eventually be other isolated regions of this size in which there is nothing massive, not-even a neutrino.

    Does the Penrose theory really require the whole universe to be absent of massive particles or is a causally disconnected region sufficient?

  21. Markus Maute says:

    What do you think about the following statement by Lenny Susskind:

    “One of the deepest lessons that we have learned over the past decade is that there is no fundamental difference between elementary particles and black holes. As repeatedly emphasized by ‘t Hooft, black holes are the natural extension of the elementary particle spectrum.”

    Doesn’t that point to a possible resolution of the problem of the “decay of all masses” ?

  22. Peter Woit says:

    George Dorn,

    Looks like the dozens of other bogus “We’ve found a way to test string theory!” papers chronicled here in “This Week’s Hype”. I’ll be interested to see if the media pick it up, in this case it looks like the author’s institutions at least haven’t so far issued press releases. My impression is that science journalists mostly now realize that they’ve been had far too many times by these claims and may now start just ignoring them. Perhaps that’s the best thing to do here too…

  23. Yatima says:

    >Does the Penrose theory really require the whole universe to be absent of massive particles or is a causally disconnected region sufficient?

    Well, the new C³ universe doesn’t “take off” at the exact moment everything is living on the light-cone. The new C³ universe is “just” a remapping of the extremely large, fast-expanding and empty old one. If I understand correctly, once only conformally mappable fields living on the light-cone remain the remapping makes *mathematical* sense, but in effect the current universe, right now is already the next one moving towards its Big Bang. So ok, this idea needs some work.

  24. july says:

    A quick question for anyone understanding Penrose’s idea. Why does the space is remapped but not the time? In other words, for what reason does Penrose stipulates a series of eons instead of single recycling eon?

  25. Rael says:

    bjm, phillips, OK, there are enough positrons then. I have no trouble with that.

    (I’m not sure sphalerons transitions would play a role in a very cold universe, though. They may have played a role in baryogenesis when the temperature was of the order of the electroweak scale or higher, but as the universe cools down they become irrelevant.)

    But even if there are no particles left, besides very cold photons getting even colder as the universe expands, the electron mass and the QCD scale are still present in vacuum fluctuations and they would be important at the very low energies involved in a very cold universe. In other words, QED and QCD are far from scale invariant at low energies.

  26. Jess Riedel says:

    I really don’t think it’s fair to attribute interest in cyclic universes to “a very human desire to see in the global structure of the universe the same cyclic pattern of death and rebirth that govern human existence,” any more than it’s fair to attribute interest in big-bang-big-crunch universes to, say, ‘ a desire to see in the global structure of the universe the creation and ultimate destruction of existence as suggested in the bible’. (If you have actual evidence for any non-scientific motivations of Penrose or others, please share it.) I have no particular fondness for any cyclic universe model, but in the abstract they are pleasing simply for the chance that they could remove mathematical discontinuities/singularities.

  27. lun says:

    For your next book review, I highly recommend
    “Higher Speculations: Grand Theories and Failed Revolutions in Physics and Cosmology” by Helge Kragh

    It is right up your alley, plus a historical dimension; eg, at the turn of the20th century we also had a “theory of everything” which everyone considered correct because it sort of managed to include all observed forces and objects of the world (from electromagnetism to atomic spectra) and was highly mathematically elegant (helping the development of knot theory): The theory of atoms as ether vortices.

  28. Peter Woit says:

    Jess,

    “If you have actual evidence for any non-scientific motivations of Penrose or others, please share it.”

    Actually, all the scientists I’ve ever met are human beings and have non-scientific motivations of all sorts.

    Removing singularities doesn’t have anything to do with cyclicity. If you turn the Big Bang into a Big Bounce to avoid a singularity, that doesn’t imply the universe will recontract and do it again.

    lun,

    Many thanks for the suggestion. I’d read his book on Dirac long ago, which was quite good, and look forward to reading this new one.

  29. Jess Riedel says:

    Peter,

    Thanks for the response.

    “Actually, all the scientists I’ve ever met are human beings and have non-scientific motivations of all sorts.”

    Yes, of course. What I mean is: do you have good reason to believe that cyclic-universe proponents are *significantly more* swayed by such motivations than proponents of mainstream cosmologies…other than just the fact that cyclic models aren’t the dominant paradigm?

    “Removing singularities doesn’t have anything to do with cyclicity. If you turn the Big Bang into a Big Bounce to avoid a singularity, that doesn’t imply the universe will recontract and do it again.”

    Well, fine, cyclic cosmologies are a subset of those which avoid big-bang singularities through a big bounce. How’s this for a cyclic-specific benefit: cyclic universe may remove singularities *without* producing infinite spacetime volumes (by making time periodic), thereby avoiding problems choosing measures with which to compute observation probabilities.

    Again, I’m not arguing for cyclic models. I’m just offering off-the-cuff reasons why someone might reasonably prefer cyclicity to a big bang.

  30. Peter Woit says:

    Jess,

    My comment about the aesthetic appeal of cyclic models was really nothing more than that, an observation that many people seem to find them aesthetically appealing (from what I remember, Bojowald’s recent book had some explicit discussion of this). This isn’t an argument for or against such models (including Penrose’s), or a claim that some people are subject to unscientific behavior and others aren’t.

  31. Allan Rosenberg says:

    Peter,

    I thought the way you suggested that Sir Roger has, perhaps, progressed past his use by date was very well done. Anyway, I don’t blame him for the wild speculation, I blame the microtubules in his brain.

  32. july says:

    >A quick question for anyone understanding Penrose’s idea

    It seems no one here fits the description. ^^

  33. Lee Flight says:

    Penrose covered much of the material in the book in a recent talk at the PI: http://pirsa.org/11040063/

    I saw him give much the same talk live a few weeks ago, he was very open about some of the issues with his approach, the fate of matter fields “you can take it or leave it”. There is also his willingness to lose information in black holes which seemed a bit odd to me as a main driver in his book was to reflect on the second law of thermodynamics and it seemed like the evolution of the left-hand side of the Boltzmann entropy formula (S) was of paramount importance to his thinking and yet the right-hand side (logW) not so much. Any how I thought the book worth the price just to read him on conformal diagrams.

  34. Peter Woit says:

    Allan,

    I’m sorry to hear that you interpreted my review that way, it’s not at all what I think or intended to convey. The idea put forward in Penrose’s book is quite speculative and has problems, and he says as much himself. But it’s far more lucid and sensible than the kind of speculative nonsense involving multiverses that is is now highly popular among prominent physicists less than half Penrose’s age. If I wanted to classify books about fundamental physics aimed at the public in recent years by degree of apparent brain damage of the author, Penrose’s is definitely on the low end of the neurological problem scale.

  35. Anon says:

    Just to add into this conversation, Penrose also ties his CCC ideas into his Orch-OR ideas in his latest paper on consciousness with Hameroff:

    http://journalofcosmology.com/Consciousness160.html

    From towards the end:

    “The recently proposed cosmological scheme of conformal cyclic cosmology (CCC) (Penrose 2010) also has some relevance to these issues. CCC posits that what we presently regard as the entire history of our universe, from its Big-Bang origin (but without inflation) to its indefinitely expanding future, is but one aeon in an unending succession of similar such aeons, where the infinite future of each matches to the big bang of the next via an infinite change of scale. A question arises whether the dimensionless constants of the aeon prior to ours, in the CCC scheme, are the same as those in our own aeon, and this relates to the question of whether sentient life could exist in that aeon as well as in our own. These questions are in principle answerable by observation, and again they would have a bearing on the extent or validity of the Orch OR proposal. If Orch OR turns out to be correct, in it essentials, as a physical basis for consciousness, then it opens up the possibility that many questions may become answerable, such as whether life could have come about in an aeon prior to our own, that would have previously seemed to be far beyond the reaches of science.

    Moreover, Orch OR places the phenomenon of consciousness at a very central place in the physical nature of our universe, whether or not this ‘universe’ includes aeons other than just our own. It is our belief that, quite apart from detailed aspects of the physical mechanisms that are involved in the production of consciousness in human brains, quantum mechanics is an incomplete theory. Some completion is needed, and the DP proposal for an OR scheme underlying quantum theory’s R-process would be a definite possibility. If such a scheme as this is indeed respected by Nature, then there is a fundamental additional ingredient to our presently understood laws of Nature which plays an important role at the Planck-scale level of space-time structure. The Orch OR proposal takes advantage of this, suggesting that conscious experience itself plays such a role in the operation of the laws of the universe.”

  36. Peter Woit says:

    Anon,

    Thanks, that journal looks pretty peculiar and crank-ridden. I’ve no idea what Penrose is up to with getting involved with them or what his current ideas about consciousness are, he keeps that quite separate from his cosmology work. I don’t recall seeing anything at all about any of this in either his book or the talk on the subject I attended a while back.

  37. ArchBishopofCanterbury says:

    You are the first to report about all the negative things, reports about the alleged rise of string dictatorship etc etc etc. However, when there is anything interesting going on the field like the fermi lab anomaly you dont have much concrete to say. This makes me really sad and also suspicious of you. I wonder if you really have anything to say about anything that matters.

  38. Peter Woit says:

    ABC,

    Well, I’m right in the middle of writing a new posting, and was just writing about the CDF news. Not to get anyone’s hopes up, mainly I was just pointing to the postings by Jester and Tommaso Dorigo. In general, when a topic is being well covered by other bloggers, I don’t see much point in my going on about the same thing. Very often, on the latest news about some experimental data, the two of them do an excellent job of covering the topic from both the experimentalist and particle theory phenomenologist point of view. I wish I could compete and provide something better than them, but that’s rarely the case.

  39. John Baez says:

    Jess wrote:

    I really don’t think it’s fair to attribute interest in cyclic universes to “a very human desire to see in the global structure of the universe the same cyclic pattern of death and rebirth that govern human existence,” any more than it’s fair to attribute interest in big-bang-big-crunch universes to, say, ‘ a desire to see in the global structure of the universe the creation and ultimate destruction of existence as suggested in the bible’. (If you have actual evidence for any non-scientific motivations of Penrose or others, please share it.)

    Since nobody else is doing so, I’ll come out and admit that the idea of a cyclic universe always appealed to me, just because it seems depressing to think the universe will just fizzle out in a heat death. The idea of a phoenix-like “death followed by rebirth”, and another chance at a universe with no Sarah Palin, seems a lot more pleasing.

    However, like plenty of scientists, I’m sensible enough to recognize that the universe isn’t necessarily the way we’d like. Right now the evidence seems to favor an eternal and increasingly boring and chilly future – a scenario I explore in my page on The End of the Universe. Anyone who places a lot of stock in perpetual progress, or the notion that the future makes our present actions worthwhile, has got to think about this scenario and consider recalibrating their value system.

    However, it’s also important to remember that there’s a lot we don’t know about physics, and predictions of what will happen in, say, 10^(10^26) years are incredibly un-robust: changes in our understanding, or even more careful calculations based on the physics we already know, could change these predictions dramatically.

    (10^(10^26) years is one estimate I’ve seen for how long it will take for the Earth to quantum-tunnel into becoming a black hole.)

    So, besides avoiding a value system where the worth of our actions today depends heavily on their consequences in the far future, it’s probably also good to avoid a value system where the worth of our actions depends heavily on calculations that could easily be wrong.

  40. Rael says:

    (10^(10^26) years is one estimate I’ve seen for how long it will take for the Earth to quantum-tunnel into becoming a black hole.)

    But, by then, the Earth will have long ceased existing.

    Anyone who places a lot of stock in perpetual progress, or the notion that the future makes our present actions worthwhile, has got to think about this scenario and consider recalibrating their value system.

    I think you’re mixing two completely different time scales. Our values are based on a human time scale (decades, centuries, or even a few milennia), and that’s as it must be. The notion that we should recallibrate our values because the Universe will be cold and lonely in 10^(whatever>9) years seems to me a form of millenialism.

  41. John Baez says:

    Rael wrote:

    Our values are based on a human time scale (decades, centuries, or even a few millennia), and that’s as it must be. The notion that we should recalibrate our values because the Universe will be cold and lonely in 10^(whatever>9) years seems to me a form of millennialism.

    Some people seem to think that it’s important that things keep getting better forever. I’m saying that there’s no reason to think that, so we should avoid a value system where the worth of our actions today depends heavily on their consequences in the very far future.

    Maybe I should give you an example of the people I’m talking about. To take an extreme case, there’s Frank Tipler:

    According to Tipler’s Omega Point cosmology, for the known laws of physics to be mutually consistent it is required that intelligent life take over all matter in the universe and eventually force the collapse of the universe.

  42. Rael says:

    Oh well, in that context your point seems perfectly sensible to me. But, boy, what a context!!

  43. Tony Smith says:

    John Baez said
    “… we should avoid a value system where the worth of our actions today depends heavily on their consequences in the very far future …”

    which reminds me of the “corporate finance” guy who, when evaluating projects for building civilization, asked
    “… what is the net present value of these projects? …”

    in Abstruse Goose at abstrusegoose.com/363

    Tony

  44. John Baez says:

    Hi, Tony. it’s interesting to think about the whole idea of how the future matters less and less to us as it recedes into the distance… and how what counts as the “far future” depends on who you’re talking to. For some “corporate finance guys”, the far future might be 5 years from now. But for me, if I had to pick, I’d say it begins when the human race is about twice as old as it is now. We should at least try to do things to make it likely that we’ll be around as long as we’ve been around.

    The idea of hyperbolic discounting is one attempt to inject some sanity into discussions of how much we value the future.