The First Evidence For String Theory? Not.

Over the last couple years there has been a large amount of hype about cosmic strings, including press releases from Santa Barbara, and a story in New Scientist about The First Evidence for String Theory?.

The Santa Barbara press release from June 2004 concerned a paper by Polchinski and others about potentially observable fundamental strings of cosmic size. It stated that “LIGO… could provide support for string theory within two years.” There are five months left for this prediction to work out.

The New Scientist story was about an astronomical object optimistically given the name “CSL-1” (for Capodimonte-Sternberg-Lens candidate), that supposedly might be a galaxy lensed by a cosmic string, causing it to appear doubled. Of course the much less exotic and much more likely possibility was always that it was just two similar looking neighboring galaxies. Recently the Hubble space telescope was used to take a closer look at CSL-1, and as reported here and here the Hubble image clearly shows that it’s not a cosmic string, just two nearby galaxies.

This entry was posted in Uncategorized. Bookmark the permalink.

74 Responses to The First Evidence For String Theory? Not.

  1. Christine says:

    Who Says:

    I note that this research community does not have much visibility on the Web: for instance there is no QG blog and I don’t know of any central bulletin-board for QG news information and comment.

    Yes, you are right about that. The fact that I couldn´t find a blog with discussions directly related to what you call QG* was the reason that I myself started one (which includes other matters as well), even though I am quite aware that it will probably take some time (viz. several years) until my blog reaches the level of quality and usefulness that you mention (and that I very much would like to!). BTW, you have been a very nice contributor to my blog and that is what really enriches a blog, useful comments, specially in its initial stages!

    I don’t see why universities are not scrambling to hire these people in junior faculty positions.

    Life is not like that, unfortunately I know it well.

    Best wishes

  2. Kasper Olsen says:

    Aaron: I think that LQG and CDT are related in the following way; they are both attempts at a “background independent” approach to quantum gravity; and furthermore, they don’t appear to pose any restrictions on the type of matter that couples to gravity…

    Gravity is important – but there is, to be honest, more that Matters 😉

  3. Aaron Bergman says:

    You know, if I never hear the term ‘background independent’ ever again, my life would be just that bit happier.

    As for the types of matter allowable, it’s not at all obvious to me that the existence of a UV fixed point would be insensitive to the types of matter allowable. I’m more amenable to the idea than Jacques is, but it’s far from an apparent point.

  4. Chris W. says:

    [You’ll want to skip this, Aaron.]

    The original form of the requirement of background independence was introduced by Leibniz. Closely related ideas and arguments were developed by Mach, and these have arguably been the most influential.

    On the face of it, Newton won this argument with Leibniz; the success of Newton’s mechanics and theory of gravitation appeared to show that misgivings about the soundness of his assumption of absolute space and time were misguided. Mach’s criticisms made clear to Einstein that they were not. What Mach pointed out was that the operational significance of the notion of an inertial trajectory, and by extension an inertial frame of reference, were deeply unclear in the presence of universal gravitation and in light of the unobservability of absolute position in space. There is an operational resolution to this difficulty, but it is inherently dynamical. Julian Barbour has argued persuasively that this is in fact the essence of what Mach was getting at.

    All this has been taken as irrelevant to particle physics, where the operational validity of an effectively (if not literally) inertial laboratory frame is taken for granted. The problem cannot be easily evaded however; one cannot dismiss the issue by simply saying, well, some of us may like the idea of background independence but it may not be the way nature is. The precedent of Newtonian gravity + mechanics suggests that this is misstating the issue. To say that there is a fixed background spacetime implies that one has found a way to observe it, and has done so. However, we have not done so, and the existence of such a spacetime is fundamentally inconsistent with general relativity.

    In particle physics, before we concerned ourselves with quantum gravity, we could say “well, maybe there is a fixed background spacetime, or maybe there is merely an effective background; what difference does it make?” In quantum gravity we must understand the dynamical origin of that effective background. If we simply assume it as an absolute—eg, because we only know how to formulate quantum mechanics in the presence of a fixed background with certain symmetries—then we assume its absolute role in the classical limit, and its observability in that limit. But all our experience suggests that this is not possible, and this impossibility is built into the deep structure of general relativity.

    If string/M-theory or particle physics in general could squarely confront this issue—say, by showing that a background is used as a formal device in certain formulations, but is irrelevant to the theory’s physical interpretation or observational consequences—then I think you could be spared further heated discussion of “background independence”.

    On the role of matter in all this, I’ll simply point out that matter is what we can observe, and that all our operational conclusions about spacetime as a physical entity are based on this. So in some sense matter and spacetime are bound to each other. One is tempted to say that one is derived from the other, or a manifestation of the other. String and M-theory incorporate matter, or provide the substrate for effective field theories to arise, but they treat gravity in this context as just another effective field theory in a background spacetime—do they not? This is what needs to be clarified. The massively proliferating complications of the last 10 (or so) years certainly don’t help matters.

  5. Thomas Larsson says:

    At the level of statements of results, the basic claims of LQG and CDT are generally accepted. I assume you agree with this as no critic has shown a technical error in an important result-such as those listed in my review. There is of course disagreement on their significance.
    Let us put the question this way: it is a non-trivial question whether there could be any examples of background independent QFT’s.

    The technical objection is that polymer representations of the diffeomorphism group are not of lowest-weight type, unless all other representations of symmetry groups in quantum physics. Hence, although there is no doubt that LQG models exist and are background independent, they apparently fail to be genuinely quantum in this respect. It is easy to argue that no locally non-trivial strictly background independent QFT exists at all. Mathematically, the 1D version of this argument is that the only unitary irrep of the centerless Virasoro algebra is the trivial one.

    The solution to the problem of QG lies elsewhere: one must introduce an explicit representation for the observer’s trajectory in spacetime. This leads to the resolution of three classes of conceptual problems:

    1. It resolves various problems of time.
    2. It resolves problems with the Copenhagen interpretation, by moving the observer to the quantum side of the Heisenberg cut.
    3. It makes it possible to formulate diff anomalies in 4D, which we know must exist in analogy with 2D gravity, because the multi-dimensional Virasoro cocycles are functionals of the observer’s trajectory.

    For details, see e.g. this post or this forthcoming book. In order not to litter Peter’s blog with off-topic material, objections may be posted here.

  6. Aaron Bergman says:

    If string/M-theory or particle physics in general could squarely confront this issue—say, by showing that a background is used as a formal device in certain formulations, but is irrelevant to the theory’s physical interpretation or observational consequences—then I think you could be spared further heated discussion of “background independence”.

    Please read this thread. You can start with my post #70 if you want to skip to the relevant stuff.

    The quick summary is that string theory looks exactly as if the background is only an artifact of th formulation. A more interesting question is the dependence on asymptotics and that is dealt with extensively in the above comment thread.

  7. Urs says:

    urs said:

    It would also be impressive to have a theory which would allow to derive most, if not all, distances of the planets from our sun.

    Yes, it would be impressive. Let me propose such a theory…F=ma… As input we need […]

    [… ] the conditions shortly after the big bang

    Exactly, that’s my point. You cannot derive it from first principles, but you can derive it given specified initial conditions.

    Hence it is not, as Kepler once hoped, a unique solution, but one out of an infinite “landscape” of solutions. In the end, the question “why” the earth has the distance from the sun that it has is certainly not independent from the a posteriori fact that carbon-based systems observe it being there. (That’s supposed to be an uncontroversial tautology.)

    This does not imply that the same state of affairs applies to the parameters of the standard model. But it may help illustrate that it is nothing exotic for a theory to yield unique solutions only after some initial conditions have been specified. It is in fact the generic case.

    We may hope for more, but it does not mean to abandon science to expect less.

  8. Thomas Larsson says:

    unless all other representations
    unlike all other representations

  9. dan says:

    so given relevant initial inputs, is string theory able to derive other features of the standard model?

    “This does not imply that the same state of affairs applies to the parameters of the standard model. But it may help illustrate that it is nothing exotic for a theory to yield unique solutions only after some initial conditions have been specified. It is in fact the generic case.”

  10. Urs says:

    so given relevant initial inputs, is string theory able to derive other features of the standard model?

    As you may know, fixing a compactification for strings (heterotic, say), specifies completely a corresponding low energy field theory which is observed as the ‘standard model’ of the world described by this solution.

    So for instance the known compactifications which come close to the standard model all make a prediction for the number and nature of the Higgs fields in these theories.

    But most models so far also predict a bunch of massless scalars (moduli). The most recent model by Braun et al, which produces otherwise exactly the structure of the standard model, as far as I understand, has reduced these to a mere 13. But that’s still 13 too many to be phenomenologically viable.

    Which, incidentally, means that these solutions have been falsified.

  11. Who says:

    Christine said

    The fact that I couldn´t find a blog with discussions directly related to what you call QG* was the reason that I myself started one (which includes other matters as well),…

    I should have mentioned your new “Background Independence” blog as potentially just the sort of news and discussion-board for quantum gravity that I had in mind! BTW the blog is visually very nice.

    An erratum: I accidentally transposed digits when I meant to write
    and wrote 0502016 instead—back on 15 Jan 2:32P

    BTW Christine, another thing, 2005 was a significant year for “QG*” in which a lot of progress was made. You could ask for nominations, at your blog, for the most important paper. Then, if only I showed up at the blog and proposed a candidate, my nominee would automatically win 🙂

  12. Who says:

    Lee (14 Jan 4:41P)

    Dear Dan,

    Actually LQG and related research programs (such as CDT or programs based on quantum information ideas) are progressing rapidly,

    Dan (14 Jan 7:13P)

    I am curious: how close is LQG on its semiclassical description, dynamics and contact with standard model particle physics?

    Moshe (15 Jan 10:30A)


    I believe the non-trivial question for LQG and similar approaches is whether they are indeed quantum theories of gravity, in other words do they contain semi-classical states and propagating gravitons in 4dim. I think lots of people will become interested in any quantum model that can be shown to contain gravity as we know it, whether it is BI or not. Since there is some progress on this recently, maybe one can be cautiously optimistic.

    Lee (15 Jan 6:49A)

    Of course LQG includes spin foam models.

    Carlo Rovelli
    GENERAL RELATIVISTIC QUANTUM PHYSICS: Background independent scattering amplitudes, boundary formalism, local particles and partial observables
    Abstract: (i) I present some preliminary results on background independent calculations of particle scattering amplitudes. In particular, I discuss the derivation of the graviton propagator, from loop quantum gravity and the spinfoam formalism. (ii) I illustrate the boundary formulation of quantum field theory, its role in a background independent context, and how “particles” emerge in this language. (iii) More in general, I discuss how systematic physical predictions can be extracted from a general relativistic quantum field theory: I illustrate the notion of “partial observable”, and discuss the issue of the physical significance of the spectrum of these observables, which controls the interpretation of the area and volume discreteness.”

    Effective 3d Quantum Gravity and Non-Commutative Quantum Field Theory
    Laurent Freidel, Etera R. Livine

    “We show that the effective dynamics of matter fields coupled to 3d quantum gravity is described after integration over the gravitational degrees of freedom by a braided non-commutative quantum field theory symmetric under a kappa-deformation of the Poincaré group.”

    “We will use these results to write down the Feynman graph amplitudes of a massive scalar field evolving in the 3d quantum geometry and we will show how they reduce to the standard Feynman graph amplitudes of the standard quantum field theory in a no-gravity limit G -> 0. On one hand, that solves the problem of the semi-classical limit of 3d quantum gravity, and on the other hand, this formulation allows us to compute explicitly the quantum gravity corrections to the standard quantum field theory scattering amplitudes and show that the corresponding geometry is effectively non commutative… “

    “… In the full quantum gravity regime, l_P and m_P are fixed and finite. The (semi-)classical limit is defined as

  13. Who says:

    Sorry, the preceding got cut off. Here is the rest

    “…The (semi-)classical limit is defined as h-bar -> 0
    l_P is taken to 0 while m_P remains fixed: we do not expect to recover standard classical flat space-time physics but a modified theory with a universal mass scale m_P as postulated in deformed special relativity.

    The no-gravity limit G -> 0 takes l_P to zero and m_P to infinity but

  14. Who says:

    What I am trying to do is give additional substance to some things others have said in this discussion. What Moshe said regarding “some progress recently” on gravitons connects to Rovelli’s talk at Loops ’05. What Lee said about the 3D case connects to the Freidel Livine paper. And quotes from that paper serve as response to Dan’s question about LQG and the semiclassical limit. Freidel’s system shows the correct behavior both as G goes to zero (“zero gravity” limit) and also as h-bar goes to zero (semiclassical limit). Recent QG work is tending to include matter and use the spinfoam formalism (as illustrated by these examples.)

  15. woit says:


    Please, this posting had nothing at all to do with LQG. This has become way off-topic, and I don’t think you should keep posting comments about this here.

  16. Who says:

    Hi Peter,
    two issues: I was having technical problems. sorry about the multiple posts. the message in the preview kept getting truncated and not going up

    second issue: I think the Freidel Livine paper one of the most important LQG papers of 2005. If we are going to discuss progress in LQG at all, we need to be able to talk about the Rovelli paper (which is uses spinfoam formalism) and the Freidel-Livine paper (also spinfoam).

    See last paragraph of page 1 of F-L:
    “We use the spin foam quantization of 3d quantum gravity provided by the Ponzano-Regge model…”

    See Smolin’s post 15 Jan 6:49A:
    “Of course LQG includes spin foam models.”

    If you are going to allow a discussion of last year’s progress in LQG then you have to allow people to talk about both the Rovelli paper and the F-L paper. Both the Rovelli and the Freidel papers were presented as invited talks at Loops ’05 in October. Both are core representatives of current work in LQG.

  17. Who, I think Peter is saying his original blog post was not about LQG. If Peter had started an LQG thread your comments would probably be no problem. It can get started by somebody else I know.

    John… practicing for my own blog… just kidding

  18. woit says:


    John is right. This was not a posting about LQG, and while I think the idea of an LQG weblog is great, this is not it. I don’t have either the competence or desire to run such a thing. Also, I’m afraid that no matter what the originial topic, you tend to turn the discussion to LQG or cosmological natural selection topics. If you want an in-depth discussion of these topics, you should set up a weblog devoted to them.

    The other problem with this is that it gives lots of other people the idea that this comment section is a forum for them to discuss their favorite ideas, even if they’re not the topic I’m posting about and have only come up tangentially in the discussion. I then end up having to spend a lot of my time deleting people’s comments to stop this, annoying them and wasting my time.

  19. D R Lunsford says:

    Peter – the problem is, there isn’t a neutral zone to talk about ideas – in a strange way you’ve created a neutral zone of sorts, whether you like it or not – this is a credit to you because you are letting physics as it were “speak for itself”, and we all admire you for that.

    So it’s not polite to run on about things here, but we/they/all of us only do it because you have set up fairness as your first principle.


  20. Pingback: Thoughts on science and life » Blog Archive » CSL-1: The End of Cosmic Strings?

  21. Who says:

    Thanks for the reactions. DRL mentions courtesy. I apologize for (unintended) impoliteness and will try to be more considerate. My main concern is that the discussion of QG is often clouded by “Nicolai’s mistake”—he coauthored a paper purporting to give an outsider’s view of LQG and didn’t discuss the spinfoam (path-integral) formulation. Peter then posted a letter from Lee responding to Nicolai in which Lee’s main point was that for the past (on the order of) ten years the spinfoam approach has predominated in LQG research. Hope I don’t overstate here. Most of the quantum gravity researchers I watch use spinfoam formalism, currently including Carlo Rovelli (who has been closely associated in the past with the canonical version of LQG).

    In this thread LQG began being discussed by others and I joined the discussion wanting to respond, supplement and clarify what was said. Dan asked about the current status of LQG and it makes an enormous difference whether or not one includes spinfoam research. Including it also makes the close relationship between LQG and CDT obvious, an issue raised by Aaron and by Kasper Olsen.

    Thanks for your patience, Peter. You do a great job creating a neutral forum space as DRL says.

  22. anon says:

    String theorist Dr Lubos Motl, who knows about propaganda, has now usefully given the following enlightening viewpoint:

    “An important part of all totalitarian systems is an efficient propaganda machine. … to protect the ‘official opinion’ as the only opinion that one is effectively allowed to have.” – STRING THEORIST Dr Lubos Motl,

  23. Who says:

    Hello anon, hi DRL,
    string theorist Hermann Nicolai (who seems the opposite of totalitarian and propagandistic—-an honorable member of the majority so to speak)
    has come out with a new paper giving an overview of the main non-string QG approaches that this time INCLUDES spinfoam research


    this could actually improve the quality of the discussion.

    It used to be that string theorists who wanted to defend against a perceived threat would lump all rivals to string under the heading “LQG” and then discuss only the orthodox canonical LQG essentially of the 1990s

    but for the past 5 years or so (maybe more), people who see themselves as part of the LQG community have largely been doing spinfoam (as they tell us, and one can see it by sampling arxiv). There has been a gradual shift towards path integral approaches (like spinfoam and CDT)

    so in the past a would-be defensive string theorist was often not connecting with reality. And people would quote Hermann Nicolai’s earlier paper of just one year ago
    hep-th/0501114 LQG an outside view
    which unfortunately did NOT discuss spinfoam or much of the activity which the LQG community was pursuing.

    Now maybe defensive string theorists will direct their attack more on the actual rivals instead of at a mannequin of their imagining. Something good is likely to result, says I.

  24. anonymous says:


    Thanks for that link:

    ‘… it is thus perhaps best to view spin foam models as models in their own right, and, in fact, as a novel way of defining a (regularised) path integral in quantum gravity. Even without a clear-cut link to the canonical spin network quantisation programme, it is conceivable that spin foam models can be constructed which possess a proper semi-classical limit in which the relation to classical gravitational physics becomes clear. For this reason, it has even been suggested that spin foam models may provide a possible ‘way out’ if the difficulties with the conventional Hamiltonian approach should really prove insurmountable.’ – hep-th/0601129

Comments are closed.