Once Before Time

There’s a new popular book out this week entitled Once Before Time: a whole story of the universe, by Martin Bojowald, promoting his ideas about “Loop Quantum Cosmology”. It’s a translation of the original German edition, Zuruck vor den Urknall, published last year.

The topic of the book is work by Bojowald on toy models using loop quantum gravity that avoid the Big Bang initial singularity of classical general relativity. For a much shorter version of all this, see his 2008 Scientific American cover story Big Bang or Big Bounce?

There’s a very deep human desire to understand origins and thus to trace the history of the universe back before the earliest periods for which cosmological theory and observations have provided some degree of scientific understanding. Unfortunately this has led in recent years to a flood of over-hyped claims by physicists claiming to have a scientifically viable theory of what happened “Before the Big Bang”. To qualify as legitimate science, such claims need to be backed up by some conventional sort of evidence. This might take the form of experimental predictions, testable either now or in principle in the future. It might also take the form of a highly constrained and beautiful theory whose success in other realms makes a compelling case that it could also explain experimentally inaccessible phenomena. I don’t know of any example of such pre-Big Bang scenarios now being sold to the public that comes even close to having such backing.

The cover of Bojowald’s book tells us about Loop Quantum Cosmology:

Now the theory is poised to formulate hypotheses we can actually test.

I’m no sure exactly what that is supposed to mean, but it appears to be misleading hype, not corresponding to anything actually in the book. The text of the book itself wavers back and forth, sometimes explaining the overwhelming problems one faces if trying to extract some kind of prediction out of the LQC framework and emphasizing how speculative it all is, at other times expressing ungrounded optimism that somehow these problems will be overcome. It ends on an upbeat, hopeful note

Will we ever, with a precision that meets scientific standards, see the shape of the universe before the big bang? The answer to such questions remains open. We have a multitude of indications and mathematical models for what might have happened. A diverse set of results within quantum gravity has revealed different phenomena important for revealing what happened at the big bang. But for a reliable extrapolation, parameters would be required with a precision far out of reach of current measurement accuracies. This does not, however, mean that it is impossible to answer questions about the complete prehistory of the universe. Cosmology as well as theoretical investigations are currently moving forward and will result in unforeseen insights. Among them might well be experimentally confirmed knowledge of the universe before the big bang.

but I found nothing in the book to justify this optimism. The few allusions to specific attempts to find some relation to something observable are vague and suffer from the book’s nearly complete lack of any references to more technical sources of information.

About fifteen years ago, in The End of Science, John Horgan described the field of fundamental physics as degenerating into what he called “ironic science”, something more like literature, art or philosophy than traditional science, pursued in a “speculative post-empirical mode”. At the time I thought he was going too far, but Bojowald’s book provides an unfortunate confirmation of the phenomenon Horgan was describing. It’s written in a rather dense and sometimes impenetrable style, featuring quotations from Nietzsche, some science-fiction set off in italics, and a few pictures of contemporary art-works supposedly relevant to the argument. Attempts are made to claim a role for pre-Socratic philosophy, with LQC finally providing a means of going beyond the pre-Socratics:

Otherwise, one can find among the pre-Socratics most of the elements of modern cosmology. Only with quantum gravity did truly new elements enter the game.

Aficionados of the loop quantum gravity – string wars will find various accurate comments about string theory and the sociology of science, and Bojowald also describes an interesting insider’s point of view on the story of the development of loop quantum gravity and the scientific figures behind it. He’s quite right that it’s a fascinating possible approach to quantum gravity well worth pursuing, but the applications to cosmology seem to me not even close to being ready for prime-time and this kind of treatment in a popular book.

Update: There’s a review by Brian Clegg at the Wall Street Journal here.

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

25 Responses to Once Before Time

  1. Marcus says:

    Posted on arxiv today:

    Observing the Big Bounce with Tensor Modes in the Cosmic Microwave Background: Phenomenology and Fundamental LQC Parameters
    Julien Grain, A. Barrau, T. Cailleteau, J. Mielczarek
    12 pages, 5 figures
    (Submitted on 8 Nov 2010)
    “Cosmological models where the standard Big Bang is replaced by a bounce have been studied for decades. The situation has however dramatically changed in the last years for two reasons. First, because new ways to probe the early Universe have emerged, in particular thanks to the Cosmic Microwave Background (CMB). Second, because some well grounded theories — especially Loop Quantum Cosmology — unambiguously predict a bounce, at least for homogeneous models. In this article, we investigate into the details the phenomenological parameters that could be constrained or measured by next-generation B-mode CMB experiments. We point out that an important observational window could be opened. We then show that those constraints can be converted into very meaningful limits on the fundamental Loop Quantum Cosmology (LQC) parameters. This establishes the early universe as an invaluable quantum gravity laboratory.”

  2. Peter Woit says:


    Seriously evaluating that paper requires more time and expertise than I have available. A quick look at it though indicates that it is discussing the possibility of features in a possible B-mode power spectrum in the CMB that would differentiate between a conventional inflationary potential scenario, and an LQC-inspired one. LQC doesn’t seem to give any actual prediction of the size of these features, that’s a free paramater.

    At this point, there’s no observation of CMB B-modes at all, and these authors don’t seem to be claiming relevance to anything measurable by current experiments (Planck). So, the bottom line is that some possible next generation experiments (how many years out?) might see not only the power spectrum predicted by inflation, but some variation of this IF unknown numbers parametrizing our ignorance about LQC are in the right ranges.

    I don’t think any of that is much reason for optimism that we’re going to be seeing pre-Big Bang effects anytime soon (or ever). But, even if Bojowald is much more optimistic, the fact remains that no specific results like that in this paper are described in the book. A serious description of CMB B-modes and the challenges of observing them and finding interesting signals in them would have made the book a lot more interesting.

  3. Marcus says:

    *A serious description of CMB B-modes and the challenges of observing them and finding interesting signals in them would have made the book a lot more interesting.*

    A good point, Peter. Would make any book on the topic (LQG cosmology) more interesting. I haven’t read Bojowald’s book and can say nothing about it, but I have read the paper by Aurelien Barrau and friends. It’s first rate.

  4. Daniel Tung says:

    Hmm..How about Penrose’s book ? He also talked about the time before big bang, in which he suggested to be just the end of another period of universe, and that the same thing will happen to the end of our current universe.

  5. Bee says:

    Hi Peter,

    You might find it interesting to check section 2.4.7 and 3.3 of my recent paper


    It summarizes the work Marcus is also referring to. If you have a bounce scenario, the bounce would leave a distinct signal in the tensor mode spectrum. That is in principle observable, though, as always, it will eventually only constrain the parameters of the model. In this regard, also have a look at this paper for constraints from presently available data


    Now one might plausibly say that the range in which the model can be tested is so far not a particularly strong constraint, but at least it’s something. Best,


  6. Tim van Beek says:

    …the applications to cosmology seem to me not even close to being ready for prime-time and this kind of treatment in a popular book.

    Here is what I know from newspaper stories published in Germany last year: Bojowald did LQC calcuations about the big bang for his PhD, which is an acceptable topic in the sense that the PhD candidate can prove that he/she understands some physics and can do reasearch/calculations on his/her own.

    Then, for reasons unkown to me, several newspapers interviewed him or published articles about this work: My impression is that people simply wanted to publish something about theoretical physics for a change, and Bojowald’s work held the promise to be interesting because it is about the “origin of the universe and everything”. It would seem that some publisher succeeded in convincing Bojowald to use the ensuing publicity to write just another “popular science” book about highly speculative ideas that no one needs 🙁

    Well, Bojowald did write an interesting (to theoretical physicists!) survey of this area in living reviews:
    Loop Quantum Cosmology.

  7. Physicsphile says:

    I haven’t read the book or the paper, but its fair to say the CMB B-modes potentially offer a unique new window into the early Universe. But, Unfortunately, their amplitude will probably be too small for us to ever detect.

  8. Marcus says:

    Here are white papers/websites discussing the proposed B-pol and CMBpol missions


    There you can see what kind of time-frame and instrumentation they are talking about.

  9. Peter Woit says:

    Thanks Bee,

    I just took another look at the book to see if I missed something. The section discussing possible early universe experimental tests is about four pages long and very vague. No specific mention of B-mode polarization at all (earlier in the book there’s a vague mention of CMB polarization).

  10. Anonymous says:

    Horgan’s “The End of Science” is really the end of Moore’s law for science, the end of “one revolution after another” exponential growth for science, which prevailed the 20th century but can’t possibly sustain forever.

  11. k97 says:

    “Now the theory is poised to formulate hypotheses we can actually test.”

    How is any of this different from Kane 1997 or Kane 2010?

  12. Marcus says:

    *How is any of this different from Kane 1997 or Kane 2010?*

    Rhetorical question? I haven’t read the book and I suspect neither have you, so we can’t actually tell *how it is different* (if it is different in kind) out of context.

    One obvious difference COULD arise if Bojowald should point to a small body of literature by phenomenologists/observational cosmologists. I know of about a dozen papers along similar lines as the latest by Barrau and friends. The authors are not Loop community—they write about observational testing of other stuff including string.

    Obviously Barrau would like to see the B-Pol mission funded. It was a proposal for the 2015-2025 timeframe, in response to a NASA call for proposals.

    In the current economic condition what is going to happen to projects which take a closer look at the CMB?

    I am not advocating. You asked *what is the difference*. I am replying that there is a hard dollar bread-butter difference. Personally I don’t think Kane 2010 was talking about one specific testing project, like putting one specific CMB observatory in orbit. In the 2015-2025 timeframe. Check the B-Pol mission link out for concrete detail.

    So I think in answer to your question that if there is a difference, it is probably a difference in the degree of concreteness. And I think that Bojowald and Barrau and others are going to have to build on that, and make that point, if they are going to get Lqc early universe models tested by observation, or any other early universe models, in the present economic/political situation.

  13. Bee says:

    Hi Peter,

    Well, I think the results are fairly new, so possible he just didn’t have them when the book was written. Best,


  14. Peter Woit says:


    Maybe that’s it, but then it’s still somewhat of a mystery why the cover of the book claims that LQC was “poised to formulate hypotheses we can actually test”, since I haven’t heard of any others besides the B-mode ones, and couldn’t find any in the book.

  15. cormac says:

    It’s not a great translation of the title, is it?

  16. chris says:


    I really like your scepticism against all sorts of hype – not just string related. it is good to see that among all the “publish or perish” crazieness that has gotten hold of high energy during the dire dataless decades there is still a voice of reason.

    LQG people are amazing. after 20 years without being able to find the Hamiltonian constraint they still feel superior to string theorists and point at their lack of progress.

  17. Marcus says:

    Please try to keep your comments reality-based 😀

    *LQG people are amazing. after 20 years without being able to find the Hamiltonian constraint they still feel superior to string theorists and point at their lack of progress.*

    I don’t see Loop researchers “pointing the finger” at stringers. They have plenty to do in their own field, and some temporarily cross over now and then if they see an interesting stringy problem to tackle. Can you cite some recent instance of finger-pointing?

    There may or may not be some inner feelings of good fortune, or of having a superior rate of progress, but it’s hard to tell inner feelings. Every Loops conference so far has had a plenary talk by a string theorist. Could be their respect and openness is good policy.

    The progress in Loop just in the past three years 2008-2010 has been remarkable. Approaches have converged and what has emerged is looking very much like coherent theory ready for testing.
    See for example 1004.1780 and 1010.1939

    Something that could, in fact, be falsified by astrophysical observation. But I think it would be naive to start making bets at this stage. My attitude is to keep my eyes open, watch the scene develop, and refrain from trying to pick winners. Emotional/partisan statements tend to sound stupid to me.

  18. Marcus says:

    I forgot. Loops 2009 in Beijing broke that pattern. They didn’t have a plenary talk by someone from the string community at that conference. Otherwise it’s been a regular custom starting in 2005 with Robbert Dijkgraaf. Just a little thing–may not seem very important to you if you have a “two-camps” mentality.

  19. Bob says:

    Marcus, no-one in the field will take the Mielczarek et al calculations seriously – from what I understand even among the LQC cohort he’s considered a joke.

  20. Moore says:

    Horgan published a brief update to the End of Science in Discover magazine a few years ago for those interested:


    No, he doesn’t really give any ground; yes, he references this blog.

  21. stavrogin says:

    marcus cites two of rovelli’s papers as a proof of how a coherent picture is emerging in LQG. These papers have absolutely nothing to do with hamiltonian constraint of LQG. These works are summerising recent progress in spin foam models whose relation to canonical LQG to put it modestly, quite overhyped.

  22. Marcus says:

    Bob, what you posted is absurd. Mielczarek is not the lead author of the paper I mentioned, which follows in a line of some half-dozen by Barrau and Grain. Barrau is the senior author. Grain started out working with him when he was a PhD student as I recall.

    The Barrau Grain papers are taken seriously, and cited, by LQC folks last time I checked. You can probably find a citation by Ashtekar.

    The third author Cailleteau of the paper I mentioned is I believe a PhD student of Barrau, or postdoc working with him.

    I don’t remember seeing Mielczarek’s earlier work (before this collaboration) cited in Loop papers, so can’t offer an opinion, but it is irrelevant since this paper is out of Barrau’s group.

    Barrau gave a talk on this at ICHEP 2010 in Paris this summer.

    Here is a 7 page paper on the topic published by Physical Review Letters if you want a sample:
    Cosmological footprints of loop quantum gravity

    Have to say, Bob, that your post strikes me as dishonest to the point of being mildly disgusting. Thanks for the opportunity to reply.

  23. Marcus says:

    What I said was *Approaches have converged and what has emerged is looking very much like coherent theory ready for testing.
    See for example 1004.1780 and 1010.1939*

    There has been considerable convergence of spinfoam LQG with the older canonical version. This is detailed in the two papers I mention here, which constitute a brief status report. The application to cosmology is now being based on the spinfoam (“path integral”) formulation as you can see from recent papers by Ashtekar, Rovelli and others.

    I haven’t seen any claims that the convergence is complete or that there is an exact equivalence. Why don’t you see if you can find some statement about it in either of those two papers which you would call “hype”? It has been an important trend, to which Lewandowski’s work has contributed a lot, but the statements reporting on it that I have seen in the research literature have been amply qualified and rather cautiously worded. See if you can find a sample of “hype” to show us.

    I think the main point to make is that the picture with (esp. spinfoam) LQG and its application to cosmology has changed significantly in the past couple of years and that there is a robust implication for cosmology that looks testable. (See Bee Hossenfelder’s comments earlier.)

  24. Peter Woit says:

    Bojowald has a new paper out this evening claiming something about falsifiable predictions about the CMB based on LQC. It’s


    and it’s hard to compare its claims to those of the recent paper Marcus referred to, since Bojowald doesn’t refer to it.

    I agree with Marcus’s general evaluation of comments like that of “Bob”, and had thought of deleting that comment. Please, do not use anonymity to engage in personal attacks.

    I took a quick look at the new Bojowald paper, but it looks like it would take a lot more time to figure out whether there’s a there there than I have the willingness to devote to such a project. By now there’s a long history of claims about quantum gravity imprints in the CMB. Some of the first I paid attention to were those of my colleague Brian Greene and his collaborators ten years ago


    I’m afraid I’m still a skeptic about all efforts of this kind. The information so far read off of the CMB is impressive, but corresponds to relatively low-energy physics, very far from telling us about higher energies scales, much less the Planck scale.

  25. Terra says:

    The proliferation of QG models in which we can actually compute things (that includes strings, LQG, and a number of other approaches) has lead us to believe that quantum gravity respects unitarity — since we’ve got no idea how we’d do quantum calculations without unitarity, we’re like a bunch of drunks searching for our keys under the streetlight, because that’s where the light is.

    (Although most people work with orthogonal basis sets, unitarity is the property that’s actually needed for a basis set to work… It’s very possible to work with phase space representations that aren’t orthogonal, although you open up another can of worms when you do that.)

    Anyhow, if you believe in unitarity, there’s no information loss in a black hole, no singularity in a black hole, and no singularity at the beginning of the universe…. These results hold up whatever model you’re using.

    Paul Ginsparg wrote an editorial where he pointed out that we don’t have any proof that QG (in the real world) respects unitarity — all of the QG approaches we have now are still castles in the air. On the other hand, if you believe in unitarity, you get such wonderful results that it’s hard to give up.

    Of course, that’s the tragedy of modern physics; the interior of black holes is one of the most exciting things right now, but nobody will ever get to go inside one, see what it’s like and tell us the tale. 😉

Comments are closed.