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

Penrose's take on QM and QG?

  1. Mar 16, 2008 #1


    User Avatar

    Let's assume I'd like to learn about Roger Penrose's view QM foundations in relation to gravity.

    It seems Rovelli refers to him as an alternative approach and I'm inclined to find out exactly how his view differs.

    Any suggestions where I start reading? His logic and line of reasoning and take on QM foundations and any speculative relations to gravity is my focus.

    Should I get his book The Road to Reality? Or is there a better thing? Any QG paper by him?

    So far I didn't really read anything specifically by Penrose.

  2. jcsd
  3. Mar 16, 2008 #2

    George Jones

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I gave a very brief summary of some of Penrose's ideas in post #4 of https://www.physicsforums.com/showthread.php?t=110472".
    Last edited by a moderator: Apr 23, 2017
  4. Mar 16, 2008 #3


    User Avatar

    Thanks George!

    I will make sure to read that thread and follow up on the references. I'll post back here whatever my impressions will be.

  5. Mar 16, 2008 #4
    I would like to very strongly recommend Penrose's "Road to Reality". He doesn't explicitly form a specific argument about the nature of QG/QM but he does share his opinion in a fairly extensive way in bits and pieces scattered throughout. Also it's just a highly enlightening book.

    There's also a long chapter on Twistors, although I'm not sure how practically useful that chapter is (it's the last chapter and I actually skipped it... the book was after all overdue at the library already at that point :P)
  6. Mar 16, 2008 #5


    User Avatar

    I need to check that in more detail. But it sounds potentially interesting.

    The connection I try to find is one which the information capacity of two systems are likely to give rise to an attraction. And I think this is to be made in a relational QM, somewhat along the lines or Rovelli, but with some additional tweaks on QM. Something that Rovelli doesn't seem to want to do.

    I also think the superposition can be understood in terms of an optimation problem under information capacity constraints and that this is related to self-observation, or observing observations as I'd like to think of it.

    I was curious if Penrose ideas would be related to this, and if he already has some tentative answer to my questions.

  7. Mar 16, 2008 #6


    User Avatar

    I might want to order tha book someday, but I understand it's a massive book(=time consuming). And I'm more interested in finding answers to my questions, and my motivation for reading anything is that it might contain hints for answers. I got Rovelli's book, and started it and I strongly like parts of his vision, but he looses me in his reasoning. So I temporarily stopped reading (but I'll get back to it later). So my scanner turned my attention to Penrose. I have a hard time to see how QM can resists reformulation, and since Penrose seems to have this Idea I am curious on what he has in mind. There isn't enough time to read everything... so I'm trying to estimate if further research of Penrose ideas are worth the time, or if I could make progress directing my attention elsewhere.

    I started fiddling with my own ideas, but thought it would be a good idea to see if something has some part answers already.

  8. Mar 17, 2008 #7


    User Avatar

    I'm considering ordering his. The hardback version was less than half the price of rovelli's book.

    "The Road to Reality: A Complete Guide to the Laws of the Universe"
    I have spotted that 2005 edition, 1136 pages?

    Does that sound like the right book? or are there any later prints?

    His bold title makes me tink of Douglas Adams, but this isn't a pop sci book is it?
    The cheap price relative to Rovelli's book made me wonder?

  9. Mar 17, 2008 #8


    User Avatar

    I checked penrose slides gives by turbo1's links (which is in line with George's summary). I like the smell of this but I need to read more. I've got a feeling Penrose tries to explain decoherence using gravity? or something thereabout? It's not what I had in mind, I had in mind trying to see how gravity can be seen as a phenomenon appearing in complex systems. But it seems like it can be the same thing seen from two angles.

    So that perhaps the ultimate prediction here is that an analysis of this may generate a dynamics indistinguishable from that of gravity? But where it may give an alternative first principle angle?


    Last edited: Mar 17, 2008
  10. Mar 17, 2008 #9
    Sorry, I'm hopping on. I have a question. I know that Penrose popularized the fact that

    [tex]SO_0 (1,3) \cong PSL(2,\mathbb{C})[/tex] and that [tex]PSL(2,\mathbb{C})[/tex] is the automorphism group of [tex]CP^1[/tex]. Can you reformulate field theory in complex coordinates with proper orthochronous transformations of the lorentz group being replaced with Mobius transformations [tex]\frac{{az + b}}{{cz + d}}[/tex]? So is there a field theory constructed to be invariant under mobius transformations on a complex manifold that is isomorphic to the standard model?
  11. Mar 17, 2008 #10
    I have seen an inkling of this in Polchinski's book vol. I chapter 6.
  12. Mar 17, 2008 #11

    I've seen the paperback *really* cheap, and the one with the blue cover has good binding... and that is the right book, yes.

    As to whether Penrose's book is a "pop sci" book, it depends on how you define that. Penrose's goal seems to have been to write a "serious science" book for a popular audience. He says early on that he's trying to write for multiple levels at once, such that the material is detailed enough that someone with heavy physics experience will still learn new things, but someone with no background will at least understand the general idea of the material presented (even if the details interesting to the experienced person sail over their head). It is a difficult tightrope to walk but he gets away with a lot of this because he actually does attempt to present *ALL* of physics-- when he says "A Complete Guide to the Laws of the Universe", he means it. Because he does this, he is able to "ramp" the material such that the concepts needed to understand the later sections would have been presented to the low-end readers in the earlier sections. (And he does not take compromises in how he does this-- literally the first half of the book is just math.) Surprisingly the "basic" sections are a worthwhile read even to someone who already knows the material well, both because of the clarity of the presentation and because he chooses to focus on occasionally unusual details of those basic topics. (For example as I remember the first chapter in the "math" section is about noneuclidean spaces and the difference between dS and AdS-- a choice illustrative of how Penrose proceeds throughout, since it's not something you'd think of as something a popular science book would include at all, much less as its first example of "basic math material", but which if you think about it actually is both fairly relevant to modern physics and fairly easy to strip down to accessible essentials). For this reason though it *is* a "massive" book. Again though it's largely written to be clear and concise so it's a somewhat quicker read in places than you'd expect.

    So, I've only seen what of this idea that Penrose included in Road to Reality. But when it showed up there the idea basically worked like this: Penrose tried to depict quantum mechanics as consisting of two "operations", a U operation and a D operation (I think it was D). The U operation is the familiar unitary evolution of the schrodringer equation; the D operation corresponds to "waveform collapse". Quantum physics, as Penrose looks at it in RtR, is all about the U operation, but the D operation does happen. This is a fairly copenhageny way of looking at things, and as far as I remember Penrose does not specifically explore the more complex picture of collapse painted by decoherence. I don't know whether this is because he would view decoherence as the D operation in little tiny steps or because he was oversimplifying for purposes of the book.

    Anyway, at some point he suggests the idea that if we ask the question of what precisely inspires the "D" type operation to happen, the answer might be gravity: since GR requires mass to be in one place rather than a cloud of probabilities, maybe what happens is that the probabilities are allowed to spread unitarily until a moment comes at which one particular state is needed in order to "solve some problem" having to do with gravity, at which point collapse occurs. So we've got kind of a realist interpretation of the copenhagen interpretation: There's a quantum apparatus and a classical observer, but the "observer" is gravity, since after all in one way of looking at things gravity is the only classical thing in physics. It was kind of an interesting idea, though I may be garbling or oversimplifying it here-- It was unclear to me just from RtR how seriously Penrose took this idea, and I don't remember if he offered any reason to think it might be true beyond "well, here's two things we don't understand, therefore maybe they're linked".


    One more thing Fra, just since you're reading Rovelli and such it occurs to me you really should read Roger Penrose's 1971 paper here which John Baez recently finally posted to the internet, since it is the paper where spin networks were invented! As far as I know Penrose eventually abandoned the ideas from that paper and they weren't a part of his later thinking-- I think he mentions the idea of a combinatoric spacetime in RtR but doesn't explore it much further-- but the ideas in that paper do seem to have been fairly influential on LQG, and there is notation I have seen used by Rovelli and Smolin but which I did not actually understand until I read that paper.
  13. Mar 17, 2008 #12
    If I understood correctly, Penrose says that in QM wave function collapse there is no undeterminism is going on, it is just a way the information disappeared in black holes showing up again. So it seems that for him the information loss is important and the theory (if there is any) is essentially nonlocal.
  14. Mar 18, 2008 #13


    User Avatar

    Thanks Coin for alll your suggestions. The overall guess considering also George's questions coming out of I assume, reading the book, is that the book only contains hints.

    But Penrose's book seems to get good words, so I will order it even if I am unlikely to read it from cover to cover upon arrival.

    My skimming one of the papers quoted by turbo on gravity induced wave collapse and loosely speaking the some kind of idea is conveyed but I find it pretty fuzzy although I find it interesting it seems they are reasoning from different starting premises. I was disturbed by that the reasoning using alot of suspect baggage such as "mass", so I read it as suggestive only. If penrose isn't satisfied with the standard decoherenec views, I am with him on that. And I also find it quite likely that foundational QM issues are related to gravity. Because analysing the measurement problems gives rise to thinks that might very well give rise to gravity like stuff. And this analysis should be done anyway, so why not do it, and see if gravity pops out all on it's own? That's what I'm looking for.

    I'll check the 1971 paper, thanks for the link!

  15. Mar 18, 2008 #14


    User Avatar

    Penroses starts with a few statments of sentiment...

    "The basic theme of these suggestions have been to try to get rid of the continuum
    and build up physical theory from discreteness..."

    The motivation for this is not because it's "easier" to calculate with discrete models or that it's nicer - ie. it's not an approximation, it's something deeper. IMHO, the motivation is a basic observation: Does the continuum have a physical, observable basis? Does our experience tell us that observable distinsguishable events naturally induces a continuum, or is the continuum rather secondary?

    Not to mention the seemingly non-physical meaning of the information content in a continuum.

    Therfore it could be suggested that the continuum is unlikely to have a _physically observable_ correspondence. That's why I don't want to see it in the model. It introduces non-physical degrees of freedom that only blurs the logic. That's how I see it at least. It relates to the supposed subjective observable resolution of any physical observer.

    Of course noone can know, but a priori the continuum idea seem to me less sensible to be fundamental than the discrete foundation.

    "My basic idea is to try and build up both space-time and quantum mechanics
    simultaneously - from combinatorial principles..."

    "The idea, instead, is to concentrate only on things which, in fact, are discrete
    in existing theory and try and use them as primary concepts|then to build
    up other things using these discrete primary concepts as the basic building
    blocks. Continuous concepts could emerge in a limit, when we take more and
    more complicated systems...."

    I have no objections to this, it's all in line with what I am looking for!

  16. Mar 18, 2008 #15


    User Avatar

    He general sentiment to construct space and gravity out of relations, is to my liking. If this works, I'd expect that one would reproduce the equivalent of GR equations in the appropriate limit, rather than putting them in manually. That would be awesome. Penrose also seems to have in mind an identification of the mass concept, this is also interesting and is in line with my idea of association of mass/energy with confidence, which in turn I see as quanfied but subjective counting of distinguishable events. Which in turn are realised in terms of hte observes microstructure. This is what I originally thought could be the realisation of rovelli's ideas, but I got susepct when reading his QM view. That a microstructure defines the observer. And perhaps abstractly a microstructure can be thought of as a spin network. And the complexity of the network somehow should be related to mass or energy (I don't know yet), I was hoping someone else know. And the intertia can be related to the resistance of a massive microstructure against revision. And there is a selection for the makeup of the network. This is my original motivation for reading up on rovelli. And it sounds like this is not too far off penrose thinking.

    But from that generic agreement to his specific attempt it seems like I need to read in detail - which means time to spend. It seems related to rovelli's thinking, and I agree to his initial sentiment too. But then after a while it gets blurry.

  17. Mar 18, 2008 #16


    User Avatar

    I found out that penrose's idea was apparently used by witten to reduce the dimensionality in string theory just a few years ago? (I have not read it, I JUST spotted this)

    Edward Witten, "Perturbative Gauge Theory As A String Theory In Twistor Space"
    -- http://arxiv.org/abs/hep-th/0312171

    It sounds like a possible realisation of a personal objection of mine to unitarity and continuum models, that is it likely to inflate the degrees of freedom that are non-physical. This seems consistent with the idea that some of the apparent degrees of freedom in string theory are non-physical?

    Anyone else that finds this interesting?

  18. Apr 3, 2008 #17


    User Avatar

    I justy got penrose book as well as some other books I didn't have, including smolins three roards to QG. I was surprised to find that there seems to be almost not formulas in smolins book. And Penrose dedicates the foreword to explain what a rational number is, and it was less math than I expected. I think I understand Coin's comments now. It's hardly a book I'm going to read front to back, at least not this year. I'll probably start locating the presentation specific to Penrose QM/entropy/Gravity ideas, which was the motivation for getting the book, it will be interesting.

  19. Apr 20, 2008 #18


    User Avatar

    I've slowly started too think about Penrose's gravity induced collapse. Everything is very fuzzy, but I do think there may be something to this.

    He tries to discuss the stability of a superposition, and argues that this stability and thus destabilisation time, depending on how time is defined, is related to the gravitational energy between the possibilities.

    I don't appreciate a too litteral attempt and try to actually calculate the gravitational energy by semi-classical means and then use the HUP. I rather suspect he has something more clever in mind. I associate my thinking also to his 1971 paper where he argues for a combinatorical approach. In such an approach the complexity alone relates to generalised "diffusion times" and I like to think of time evolution as a generalised type of diffusion. I think penrose ideas may be in line with associating inertia with information capacity.

    I somehow think that perhaps the better angle is to look for a deeper understanding of gravity and the arrow of time from information principles, rather than the other way around.

    I'm going to think more about this. I really like Penrose associations between gravity, the second law and the arrow of time.

    If anyone knows more recent papers relating to this I'm interested. This seems the kind of fundamental reflections that I am looking for. It's easy to associate this to Martin Bojowald's idea of an alternative entropy measure. (Marcus posted about that https://www.physicsforums.com/showthread.php?t=225627) I think there may be a connection here.

    What is funny is that I don't appreciate their strong cosmology perspective, but regardless of this I see something interesting that I can relate to, and it regardes the information view. Here the choice of measures is central.

    It seems several questions are really the one and the same here...

    - the physical information basis for superposition
    - the emergence of an arrow of time
    - gravity connection when we consider parametrisations of this flow

    I would really love to see fundamental papers on this, that treats this on the fundamental level without mixing it up with other models say cosmology etc... IMO these things should have a meaning in itself... this is what intrigues me.

  20. Apr 21, 2008 #19


    User Avatar

    Since it's known that the statistics coming out of the superposition principle can not be explained by classical statistics on a single probabiltiy space. If the superpostion principle at times is the answer, then what is the question? Can an understanding of the original of the superposition and borns probability rule be a route to answer penrose's puzzle?

    If we associate the probability space with a microstructure and combinatorics, then wouldn't this puzzle suggest that that nature doesn't work with simple probabiltiy spaces? Perhaps the issues is that our application of standard probability to physics isn't quite right.

    If that is so, what's the point where things went wrong? How is our event space, and thus thus microstructure and probabiltiy space constructed and selected? If this is instead considered to be a dynamical process, perhaps the concepts that seems to be successful in QM, but yet not satisfactory understood could be seen to actually make sense?

    Could the superpostion be understand by a dynamical transformation of the initial microstructure? And why does this happen? What drives this?

    This issue with probability interpretations, seems to me the point where alot of people start to get shaky and want to think that this is not physics, and there is simply no benefit in questioning this.

    I think there are good reasons to question this. And maybe this is related to Penrose's intuition too.

    Rather than thinking that gravity will "explain" superposition, I'm thinking that analysing this should shed light on both things in a neutral way. Maybe gravity is even a consequence once the "probabilistic" foundations are reworked to match reality.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook