Does Time Run Backward in Other Universes?

Scientific American in recent years seems to be quite fond of parallel universes, with major articles promoting the multiverse here, here and here (commentary on this blog here and here). Their latest issue continues in this vein with an article by Sean Carroll entitled Does Time Run Backward in Other Universes?, which advertises his 2004 work with Jennifer Chen claiming that the multiverse explains the arrow of time. For new blog entries about this, see here for something from Sean, here for a Lubos rant.

As with all claims about the multiverse, the problem is whether they are even in principle scientifically testable or not. If they’re not, they’re not science and promoting them to the public is a bad idea. The only thing I can find in the Scientific American article that addresses the testability issue at all is the following:

As of right now, the jury is out on our model. Cosmologists have contemplated the idea of baby universes for many years, but we do not understand the birthing process. If quantum fluctuations could create new universes, they could also create many other things—for example, an entire galaxy. For a scenario like ours to explain the universe we see, it has to predict that most galaxies arise in the aftermath of big bang–like events and not as lonely fluctuations in an otherwise empty universe. If not, our universe would seem highly unnatural.

This doesn’t seem to have anything to do specifically with the Carroll/Chen claims about the arrow of time, but rather is just a restatement of one of the desired properties of multiverse models, that they don’t lead to “Boltzmann Brains”.

This entry was posted in Multiverse Mania. Bookmark the permalink.

21 Responses to Does Time Run Backward in Other Universes?

  1. Ethan Siegel says:


    Surely this is a question that we don’t take seriously anymore. It isn’t that the laws of physics aren’t time-reversal invariant, it’s that quantum mechanics and “probabilities” have a preferred arrow.

    If I take a free neutron and it decays into a proton, electron, and antineutrino, well, sure, if I took the proton, electron, and antineutrino and fired them at one another with the same momenta the neutron emitted them at, they could re-form into a neutron. But if I run time forwards, the neutron will decay again; if I run time backwards, the neutron will never decay.

    Or, in simpler terms, I can’t un-fry an egg. With a view of this, why is it even interesting to think about the arrow of time?


  2. Peter Woit says:


    If you want to discuss this kind of question about the arrow of time, best to do it over at Sean’s blog. The topic here is what if anything this has to do with the multiverse.

  3. Pingback: Because I'm Right

  4. Kris Krogh says:

    The time issue has also been a topic at Backreaction, with a poll taken: here

  5. DB says:

    There is a long tradition behind the cosmological arrow of time, going back to Boltzmann, and the question which Sean and his colleague attempt to address directly, namely, why did the universe start from such a low-entropy state really does go to the heart of the issue.

    The modern history of this subject dates from the Cornell 1963 conference (described in T. Gold The Nature of Time, 1967). This meeting is remembered for the presence of a Mr.X who, according to Hawking “felt the proceedings were so worthless that he didn’t want his name associated with them. It was an open secret that Mr.X was Richard Feynman” (cf. S.Hawking in “Physical Origins of Time Asymmetry, eds J.J.Halliwell, J.Perez-Mercader, W.H. Zurek” (1993), p.346). In my mischievous way I bring it up because Sean is fond of reminding us that he sits at Dick Feynman’s desk at Caltech:) This latter reference documents the second major conference on the subject in Mazagon, Spain in 1991. It’s an interesting read, with contributions from Wheeler, Gell-Mann and Hartle, Hawking, Zureck, DeWitt and Griffiths (of consistent histories fame) and others, along with a record of post-talk debates. But, as before, it’s very philosophical and I couldn’t extract any evidence that anyone was in the business of making scientific predictions. Instead, they appeared to spend a fair of time arguing over semantics, e.g. decoherence, consistent histories, what is “reality” etc..

    Sean’s paper is very much in the tradition and style of such contributions, but I liked it because unlike many others it really tries to home in on the core issue – why the initial low entropy.

    Ultimately however, most of this quantum cosmological palaver relies on Everett’s many-worlds interpretation of quantum mechanics and the multiverse which flows from that.

    But as we know, there is no credible experimental mechanism to tell us whether the Copenhagen interpretation should be favoured over the Many-Worlds one (or any of the other stabs at this topic). David Deutsch, who filled in the gaps in Everett’s original arguments claims there is (Int. J. Theor. Phys. 1985), but other dispute this.

    All I ask is, show me a reasonable experiment capable of telling us which interpretation (if any) is favoured by nature. Until then, it’s just philosophy with a high-tech veneer.

  6. Peter Woit says:


    My problem is not with the discussion of the entropy problem of early cosmology, but with the idea of writing articles for a major US popular science magazine promoting the multiverse and Boltzmann brain argumentation. This gives people the idea that this kind of empty speculation is what science is, impressing those who can’t tell the difference between science and science fiction, and turning off those who can.

  7. Professor R says:

    I’m reading through Sean’s article on my lunchour. It’s a very good artilce, highly informative. But the context worries me too. I feel that most science magazines make very little distinction btween theories that are grounded in experiment, and theories which are promising.

    It must be very confusing for the reader – in fact, I know it is. Every time I refer to say, GR, in a public lecture, it’s clear most have no idea of the supporting evidence. It all gets lumped in with ‘all those other theories’, from string theory to the multiverse.

    I sometimes wonder if magazines should have a special section marked ‘speculative’ – articles on topics that are a long way from experimental verification….Cormac

  8. DB says:

    I completely agree, Peter. My comments were directed at his paper with Chen.

  9. D R Lunsford says:

    I just had an example of this last night. I ran into a friend at a tavern, and he excitedly told me how he learned from “Discover Magazine” that a new accelerator was going to test string theory. He’s a smart person but has no training in math or science. Who could blame him for believing what he reads? The editors of these publications do not care what the content is, as long as it sells. Even “Sky and Telescope” is guilty of hyping pseudo-science to generate sales.


  10. Peter Woit says:


    I don’t think this is the fault of the editors of these magazines, they are just printing what prominent string theorists are saying. Quite a few of these are out there promoting the idea that “the LHC will test string theory”, with the great majority of physicists who know better keeping quiet.

    I suppose after all one should perhaps encourage this nonsense. If you could get everyone to believe that the “LHC will test string theory”, and as expected no evidence for string theory shows up there, maybe everyone would then accept that string theory had failed and move on.

  11. I have co-authored and published a couple of papers, in the Physics track of international conferences, on the Arrow of Time which deal with the putative Multiverse, including this one, which grapples with the specfic paper of Carroll and Chen:

    Title: Comparative Quantum Cosmology: Causality, Singularity, and Boundary Conditions

    Authors: Philip V. Fellman, Jonathan Vos Post, Christine M. Carmichael, Andrew Carmichael Post
    Comments: 17 pages, 2 figures. 7th International Conference on Complex Systems
    Subjects: General Relativity and Quantum Cosmology (gr-qc)

    (Submitted on 26 Oct 2007)

    Abstract: In this review article we compare the recent work of Peter Lynds, “On a finite universe with no beginning or end”, with that of Stephen Hawking, primarily “Quantum Cosmology, M-Theory, and the Anthropic Principle”, and two foundational works by Sean M. Carroll and Jennifer Chen, “Does Inflation Provide Natural Conditions for the Universe” and “Spontaneous Inflation and the Origin of the Arrow of Time”, in order to evaluate their comparative treatments of the nature and role of causality, time ordering, thermodynamic reversibility, singularities and boundary conditions in the formation of the early
    universe. We briefly reference Smolin and Kauffman’s recent arguments with respect to possible processes of “evolutionary selection” in early universe formation as an alternative explanation to key elements of Hawking’s earlier “M-Theory”, and its attendant anthropic
    principle. We also briefly excerpt a short section of Smolin’s recent work on topology in quantum loop gravity, simply as an illustrative
    example of the type of complex quantum topological transformation which he offers as a theoretical alternative to string theory in
    quantum cosmology.

  12. Professor R says:

    Peter, re
    “I don’t think this is the fault of the editors of these magazines, they are just printing what prominent string theorists are saying”

    I think that’s we disagree. Granted, one can’t expect science journos and their editors to have a good knowledge of every part of science. But they are surely aware that each branch tends to overemphasise its own importance, and overlook its flaws.

    In other words, they shouldn’t be just printing what prominent string theorists are saying. Surely it’s the job of a good editor to play devil’s advocate to some degree, and bring the article back down to earth. To me, it seems that this professional skepticism has gone missing in today’s science journalism…Cormac

  13. Kris Krogh says:

    Scientific American used to have the brilliant, independent-minded physicist Philip Morrison, who helped keep things grounded in reality. Unfortunately, he passed away several years ago.

  14. Thomas Larsson says:

    I suppose after all one should perhaps encourage this nonsense. If you could get everyone to believe that the “LHC will test string theory”, and as expected no evidence for string theory shows up there, maybe everyone would then accept that string theory had failed and move on.

    Larsson’s theorem uses string theory to make a falsifiable prediction about the LHC. I am particularly proud that I already in 2007 could correctly predict that poor Lubos would lose his experimental-susy-by-2006 bet.

    Of course, if the LHC does find sparticles I will be proven wrong. It is in the nature of falsifiable predictions that they are, well, falsifiable.

  15. mike harney says:

    Does anybody know if we are spending government money on multiverse theories? If so, I would like to put in a grant of my own for some untested, highly unrealistic notions.

  16. Haelfix says:

    So I gather you don’t like the idea of eternal inflation either. A lot of people don’t, but you must admit its a logical possibility (if you have inflation at one place, what stops you from having inflation somewhere else).

    It *is* testable to a certain degree, though we need to have better resolution of gaussian profiles and presumably gravity wave detectors to really distinguish between the class of models

  17. Haelfix says:

    Replace eternal with chaotic for the post to make sense.

  18. Peter Woit says:

    Mike Harney,

    To the extent government grant money is being spent on multiverse models, I think the amounts are quite small. Actually, I’m quite curious how multiverse grant proposals are doing before NSF and DOE panels. Searching on “multiverse” in the NSF database of funded proposals turns up nothing. Getting popular science magazines to promote ones multiverse research is one thing, getting other physicists to give it the kind of uniformly high marks needed to get a grant funded is something different.


    My comments were about Sean’s multiverse/arrow of time proposal, which as far as I can tell is completely empty of any predictive power or any sort of testability. Different inflationary models have different degrees to which they make distinctive predictions, but at least in those cases there is something to discuss.

  19. Smedley says:

    Point of interest: you ought probably no longer link to anything on Lubos Motl’s blog, as he has set it up so that any visitors with your blog as their referred are redirected to this (rather unfortunate and vitriolic) page:

    Stay classy, Motl.

  20. Arun says:

    ’tis why physicists find second careers on Wall Street – they excel in speculation.

  21. Pingback: Not Even Wrong » Blog Archive » Hints of ‘time before Big Bang’

Comments are closed.