Physicists Develop Test For “String Theory”

Press releases claiming that a “test for string theory” has been found appear with some regularity, notwithstanding the fact that no one actually knows how to test string theory. The latest one comes from the University of California at San Diego, where the press office today put out a press release entitled Physicists Develop Test For “String Theory”. The story has been picked up by the media, appearing here and here and probably soon in many other places.

This latest claim about a “test for string theory” is quite remarkable and even more bogus than usual. It is based on a paper which has nothing to with string theory and doesn’t do a string theory calculation at all. The paper first appeared on the arXiv last April with the title Falsifying String Theory Through WW Scattering, and was extensively discussed here. In October a new version of the paper was put on the arXiv, with a changed title Falsifying Models of New Physics via WW Scattering (and this was discussed here). I’m guessing that the removal of the claims about string theory from the title was due to a referee at PRL not being willing to go along with such a title, although maybe there’s more to the story and if so I’d be curious to know what it is.

The year is just beginning, but I’m already willing to award this press release the title of “most outrageously misleading string theory hype of 2007”. It is going to be extremely hard for anyone else to match it.

Update: The Distler et. al. overhyped press release continues to spread misinformation to the public, getting more and more ridiculous as it spreads. The blog reports about string theory that:

Until now, experimental verification has not been possible; but researchers at the University of California, Carnegie Mellon University, and the University of Texas are planning a definitive test with the future launch of the Large Hadron Collider…

This then made it to Slashdot, which put out a story under the headline String Theory Put to the Test, which starts off with:

… scientists have come up with a definitive test that could prove or disprove string theory. The project is described as…

and then goes on to give a description of the LHC project.

I think the people responsible for this should be ashamed of themselves.

: Not to be outdone by UCSD, Carnegie-Mellon has also issued a press release about this. More also here and here.

Update: More at Digg, SpaceDaily, Science Frontline, etc., etc.

Update: Yet another major university issues a misleading press release about this: from the University of Texas Team of Theoretical Physicists Develop a Test for String Theory.

Update: The Resonaances blog has a posting explaining what is actually in the Distler et. al. paper, while describing the press releases, with their pretensions that the authors have found a way to test string theory at the LHC, as “hilarious”.

Update: Sabine Hossenfelder wrote in to point out that New Scientist now has an article about this, with the title New particle accelerator could rule out string theory. The article quotes hype from string theorist Allan Adams as well as from Distler, ignoring Distler’s co-authors and describing him as “leader of the team” that solved the problem no one else had been able to solve, figuring out how to test string theory at the LHC. Funny, but as far as I can tell, this great advance in the testability of string theory is not being covered at any of the string theory blogs. I wonder why…

This entry was posted in This Week's Hype. Bookmark the permalink.

89 Responses to Physicists Develop Test For “String Theory”

  1. Arun says:

    The abstract says:

    “We show that the coefficients of operators in the electroweak chiral Lagrangian can be bounded if the underlying theory obeys the usual assumptions of Lorentz invariance, analyticity, unitarity and crossing to arbitrarily short distances. Violations of these bounds can be explained by either the existence of new physics below the naive cut-off of the the effective theory, or by the breakdown of one of these assumptions in the short distance theory. As a corollary, if no light resonances are found, then a measured violation of the bound would falsify generic models of string theory. ”

    I repeat what I posted earlier, that in the absence of string theorists, we’d greet the violation of the bounds with gleeful cries of new physics below the naive cut-off.

  2. Amos Dettonville says:

    I must say this is an ingenious approach to establishing the falsifiability of a theory, one that I had not thought of before. I wonder if Popper ever thought of it, or commented on it. There are some extremely fundamental premises underlying essentially all scientific thought, and so an observed violation of one or more of those premises would automatically falsify just about every scientific theory. For example, according to my theory, the electrostatic force of a charged particle falls off (at least approximately) as the square of the distance, so if someone discovered that it actually falls off as the cube of the distance, my theory would be falsified. Likewise, according to my theory, it’s impossible to construct a perpetual motion machine, so if someone ever constructs a perpetual motion machine, my theory is falsified. I can dream up endless “tests of my theory” this way… but does this really count as falsifiability in the sense of a robust scientific theory?

    Maybe the notion of falsifiability should include an extra condition: If a particular empirical finding would falsify not only my theory, but also virtually every other theory, then it shouldn’t be counted as a test of my theory in particular. For example, a test of Lorentz covariance, or analyticity, or anything of this fundamental nature, ought to be simply described as tests of those properties. It seems inappropriate to label such tests as “tests of string theory”.

    I’m actually surprised that the press releases haven’t taken the opportunity to call these “Tests to Falsify Einstein’s Theory of Relativity!”, since a violation of Lorentz covariance would surely be more significant as a violation of relativity (an actual existing theory) than of “string theory” (a hypothetical theory that may or may not actually exist, and that may or may not actually be falsified by this result, depending on what, if anything, the theory turns out to be).

  3. Aaron Bergman says:

    I repeat what I posted earlier, that in the absence of string theorists, we’d greet the violation of the bounds with gleeful cries of new physics below the naive cut-off.

    I’d think you’d find that, if you were to talk to string theorists, they would be extremely excited if new physics, any new physics, would be found.

    [If something would falsify] virtually every other theory, then it shouldn’t be counted as a test of my theory in particular.

    I’d think a modicum of research would show that there are plenty of attempts by people to construct theories that violate most subsets of unitarity, Lorentz invariance and analyticity. Lee Smolin, to pick an example, seems to believe (at least some portion of the time) that LQG is not Lorentz invariant.

  4. Amos Dettonville says:

    It doesn’t take any “research” to know that people have examined the possibility that (for example) Lorentz invariance might be violated. Indeed, Lorentz himself theorized about this. But surely Lorentz invariance is (and has been for quite some time) accepted as one of the best supported empirical facts, and it is implicit in all the *standard* (for lack of a better word) modern theories (as opposed to research porgrams or ideas for theories) of physics. I’d think a modicum of thought would suffice to realize that if Lorentz invariance were ever found to fail, any implications it may (or may not) have for the highly speculative research program known as “string theory” (or LQG etc) would be utterly insignificant compared with the implications for the very foundations of modern physics, i.e., actual theories that actually exist today. A falsification of relativity would be staggering news… but not because of what it does or doesn’t say about “string theory”.

    My point is that it’s silly to tout failure of Lorentz invariance as falsification of the hypothetical “string theory” (whatever that may or may not turn out to be), when in fact it would falsify the foundations of all the successful EXISTING theories of physics.

  5. amused says:

    press release,

    The odds are against any of those things turning up, don’t you think? If it turns out, as many people seem to expect, that string theory can accomodate whatever new physics is found at the LHC, then these bounds at least give one way that it could still potentially be falsified.


    “I repeat what I posted earlier, that in the absence of string theorists, we’d greet the violation of the bounds with gleeful cries of new physics below the naive cut-off.”

    If the violation isn’t accompanied by the mentioned “light resonances” then the conclusion about breakdown of one of the assumptions on the short-distance theory is inescapable. (And existence of string theorists was not one of those assumptions…)
    If the light resonances are there, then the bounds would need to be modified – that’s what I was alluding to in the “some complications arise” part of my last comment.

  6. N. Nakanishi says:

    One should not confuse the Lorentz invariance of the fundamental action with that of the physical S-matrix. As I emphasized previously (January 27, 2007 at 6.14 am), Lorentz invariance can be violated SPONTANEOUSLY in the ordinary framework of QFT without contradicting the fundamental principle that the action integral is Lorentz invariant. Therefore, this theory can be a theory which coincides with the standard theory in lower energies. Furthermore, if one accepts the violation of Lorentz invariance, it is possible to remove all ultraviolet divergences without violating the unitarity of the physical S-matrix.

  7. woit says:


    I think that pretty much everyone except Distler et. al. who has ever thought about what it means to test a scientific theory understands that your test of the theory is supposed to involve characteristic features of that theory, and that you’re supposed to actually do a calculation involving such characteristic features in order to come up with anything that can legitimately be called a “test”.

    The way some string theorists recently seem willing to throw out the most basic things about what it means to do science in order to make rather absurd claims for their research never ceases to amaze me.

  8. Garbage says:

    Peter & Amos,

    I disagree with the claim that the bounds are not a test for whatever New Physics (NP) is out there in the UV. The magic of the EFT formalism is that one doesnt really need to know what is it like but what sort of basic principle respects. If the bounds are violated these principles must be re-thought, or the other possibility is NP at a much lower scale than we think.
    To partially answer press release at the same time. If these basic assumptions break down naively at the Planck scale, it isnt a priori true that they wont show up at lower scales. For instance, the breaking of LI could generate operators in teh EFT which are not Planck supressed (This is easy to understand once the 1/Mpl gets killed with the UV cuttoff scale).
    Unitarity in the other hand goes tied to the notion of evolution. Therefore, it is also possible that Planck scale violations get enhanced by
    a mixture of its time and energy dependence (recall the disp relation will not get thru unless SS*=Id).
    Yet another possibility is that the fall off of the scattering amplitude does not obey the Froissart bound and, as in the cae of LI, small violations will get amplified by the cutoff scale.

    In spite of all the possible scenarios we could come out with, the EFT spirit tells us we can still do some predictions based on simple assumptions without a complete knowledge of the UV completion. If the bounds are violated we will know one or more of these assumptions are wrong, or there is a light resonance out there we havent found. This is a robust prediction which will falsify all the models of NP which rely on such premises. All we know so far of ST falls into that class. LQG or others, might not. This is a non trivial test of the High Energy behavior of scattering amplitudes, needless to say does not conflict with classical GR since, for instance, the assumed LI in LQG are of quantum origin. As such, this is a test of quantum gravity or GUT if you wish, for which ST claims to give a description of…


  9. confused says:

    I don’t understand what you’re all so upset about with this paper. The authors note that not all low energy effective field theories can be UV completed to some string theory. This is important and it runs contrary to what most anti-string people claim. Furthermore, it’s not true that any quantum theory of gravity which anyone takes seriously will automatically satisfy unitarity, lorentz invariance, etc – LQG doesn’t. So this seems to provide some nontrivial information about physics beyond the standard model. Obviously this can’t be used to verify string theory but it could, in principle, falsify it (modulo assumptions which look quite reasonable to me). This is EXACTLY what you guys are always complaining about: string theory can’t be falsified. But when somebody presents some small progress in the direction of showing that string theory can be falsified you people all complain about it, rather than praise the authors for at least trying.

    As for the hype, yes, of course, the title somewhat overstates the claim but it’s not an outright lie (they don’t need to do a full string theory calculation to know that string theory respects unitarity, etc). All the authors have done is what everybody does: try to motivate people to actually read their paper. Do you also have “ethical” objections to giving your paper a playful/funny title in an attempt to attract attention?

  10. Peter Woit says:

    confused and Garbage,

    My problem is not with the actual content of the paper. The authors are claiming to have some bounds which, in the extremely unlikely event they are violated, would indicate the existence of some new physics. The problem is that this is very different than having a “test of string theory”. They don’t actually use string theory in any way at all in their paper, in particular they don’t show that string theory satisfies the assumptions they are making (and no one knows if whatever non-perturbative string theory is, whether it will satisfy these assumptions). It’s quite simple: the paper is simply not about string theory at all, so it can’t in any sense be “progress in the direction of showing that string theory can be falsified”. There’s nothing wrong with trying to show string theory can be falsified, there is something wrong with claiming you can show this when you can’t.

    “All the authors have done is what everybody does: try to motivate people to actually read their paper. Do you also have “ethical” objections to giving your paper a playful/funny title in an attempt to attract attention?”

    Unfortunately the title of the press release doesn’t seem to be a joke, although it really is one. Yes, I do see an ethical problem with giving a paper a misleading, dishonest title in order to get people to read it and it appears that a PRL referee had the same problem. I see an even bigger problem with issuing misleading and dishonest press releases, which is what the authors have done here.

  11. onymous says:

    no one knows if whatever non-perturbative string theory is, whether it will satisfy these assumptions

    You keep saying this, but we do know that string theory is a theory satisfying the rules of quantum mechanics, so I don’t see how unitarity could ever be violated by nonperturbative effects. It also seems hard to believe that analyticity would ever be violated, because it encodes causality properties. Lorentz invariance, you might be able to raise questions about, I suppose (maybe you can turn on a small background B field or something?). So if you were going to try to argue that a realistic string theory could violate one of these principles, I guess that Lorentz invariance would be the place to start. In any case, within the paradigm of string theory on 4D Minkowski space times a compact manifold, the assumptions seem pretty much unimpeachable, right? No matter what nonperturbative physics is happening to stabilize the compact manifold, it doesn’t change the far short-distance properties of the theory. “Nonperturbative” doesn’t automatically mean “we have no clue.”

  12. Peter Woit says:


    Getting 4D Minkowski space X a stabilized compact manifold is just one aspect of what you hope your non-perturbative string theory, whatever it is, will look like at large distances. You don’t know what is going to happen at arbitrarily short distances. With very little trouble you should be able to find many places where prominent string theorists such as Gross, Witten, Seiberg, etc. go on about how understanding non-perturbative string theory will require us to give up our conventional notions of what space and time are. I don’t see how you can reconcile this with claiming that you know that non-perturbative string theory involves a 4D Minkowski space structure and associated analyticity properties down to arbitrarily short distances. It seems to me that Gross, Witten, Seiberg, etc are saying the exact opposite, and that they would be happy to claim that violation of the bounds in question is evidence for string theory, not that it falsifies it.

  13. Amos Dettonville says:

    Just to be clear: I don’t dispute that the violation of Lorentz invariance would falsify one of the basic premises of any theory or any research program that takes Lorentz invariance as one of its basic premises. I’m just observing that this includes essentially all *existing* successful theories, as well as many/most research programs for new theories. Lorentz invariance is a very fundamental feature of existing physical theories, and the fact that its violation would undermine some speculative directions of research into future theories (“new physics”) strikes me as secondary to the fact that it would falsify the foundations of all existing theories.

    Perhaps one can take the phenomenological view that existing theories are not based on any degree of Lorentz invariance beyond what has been experimentally verified, so if a violation were found outside those limits, it wouldn’t affect existing theories. But that seems like a rather obtuse interpretation of current theories. Surely the point of a scientific theory, like relativity, is not just to be a catalogue of known empirical results, but to predict things beyond what has already been measured, and to place the phenomena in a coherent conceptual framework. If a measurement conflicts with the prediction of a theory, even in a region beyond what has been tested before, then the theory is falsified. We may still accept that the theory “works” approximately within a limited range, but it is nevertheless falsified in its previous sense.

    I hope we can all agree that relativity predicts that Lorentz invariance is never genuinely violated (more or less by definition), so if a violation is found, relativity is falsified. All I’m saying is that, to me, the falsification of relativity itself would be far more significant than the undermining of one or more speculative research programs based on (among other things) relativity. Notice that the paper in question points to Mattingly’s 2005 paper in Living Reviews in *Relativity* for a discussion of possible Lorentz invariance and its implications.

    Note to N. Nakanishi: You commented about something that might be termed “violations of Lorentz invariance” in existing QFT, but the authors of the paper in question seem to conceed that the kind of Lorentz invariance they are talking about would in fact conflict with QFT, so they seem to be talking about what I would call a genuine violation of Lorentz invariance (see Mattingly’s paper for the meaning of “genuine”).

    Note to confused: I’m uneasy about giving playful/funny titles to scientific papers “in an attempt to attract attention”. I suppose if the title is understood to be an inside joke, then it would be okay, but I don’t think it is being presented as a joke when reported in the popular press. Reporters don’t seem to “get” the fact that it is a joke, i.e., that the falsification being discussed would undermine not just some avenues of string research but essentially all existing theories of physics. My objection to this practice is that it doesn’t just attract attention, or serve as an amusement, it also seems to mislead. I’m all for whimsy, but I think we should strive to avoid using whimsy to mislead people… and there seems little doubt that people have been misled in this case (see all those press releases).

  14. Garbage says:

    “…It seems to me that Gross, Witten, Seiberg, etc are saying the exact opposite, and that they would be happy to claim that violation of the bounds in question is evidence for string theory, not that it falsifies it.”

    At least not the kind of ST we have so far and that’s definitely a powerful claim. If the bounds are violated, and ST wants to stay up there, better start taking those ‘give-up’ ideas seriously.

    In the other hand, there is people who would say ADS/CFT provides a
    non-perturbative definition of ST. If that is the case, violations of the bounds would certainly throw that option out of the window, since we know the FT side is well behaved (In fact that is what convinced Hawking that BH evaporation was unitary after all. Although that might not be the case).

    The paper uses assumptions about the UV completion of the SM. So far ST is *perhaps* the only theory at hand, which happens to obey these in most of its current forms. Therefore, the bounds are a test for what we know or hope of ST. I agree there is too much of a hype behind this, but yet I see no incorrect statements in the paper nor in the authors’ claims in the press.

    I would be happy to see more paper like this coming out, and certainly it would be great if the bounds are violated and the string community starts to ‘give-up’ in any way 🙂

  15. Garbage says:

    LI can be violated in the vacuum (or a given state) and yet be there as a full symmetry of the action. In LQG the starting action is Einstein theory, its quantization introduces the foamy structure which might break LI. The LQG people would claim Einstein theory is recovered in the (h goes to 0) classical limit. The same people would tell you that the space of physical state is invariant under the whole diff group, and gauss law (local Lorentz). The same could happen to Unitarity depending on the choice of time variables even though there is an underlying unitary theory.
    For the scattering amplitudes, String theory so far doesnt seem to violate LI nor the other assumptions (see the original Adams et al. paper).


  16. woit says:


    “At least not the kind of ST we have so far and that’s definitely a powerful claim”

    The “kind of ST we have so far”, perturbative string theory, AdS/CFT, and a few versions of M-theory in special backgrounds, are already falsified. You can’t get the standard model as low energy physics out of any of them. Claims that string theory can lead to the standard model have to invoke some hoped for, but not yet understood non-perturbative theory. The question of “is string theory falsifiable” is only non-vacuous if it refers to conjectural versions of string theory that solve the problems of known versions of string theory. For these conjectural versions you can’t say one way or another whether they satisfy the conditions at issue.

  17. onymous says:

    “Claims that string theory can lead to the standard model have to invoke some hoped for, but not yet understood non-perturbative theory.”

    You do know about the work of Braun et al on the heterotic MSSM, yes? Or of Verlinde and Wijnholt? There’s not exactly “not yet understood non-perturbative theory” that’s crucial there. There’s no mystery about the short-distance physics. The models aren’t perfect yet, true, but they’re not as far from reality as you seem to think.

    (Cosmology might require not-yet-understood physics, but I don’t see any obvious connection to the short-distance questions we’re talking about here.)

  18. woit says:


    The models you discuss involve an ad-hoc choice of background, in particular an ad-hoc choice of a 4d Lorentz invariant background. This background is supposed to be determined by the unknown non-perturbative theory fundamental theory. As far as I know, they are not yet successful at reproducing the standard model.

    The question is not whether people have looked at models with 4d Lorentz invariance, the question is whether string theory has to lead to models with this invariance, and there is no argument for this.

  19. Garbage says:

    “The question is not whether people have looked at models with 4d Lorentz invariance, the question is whether string theory has to lead to models with this invariance, and there is no argument for this.”

    There will be if the bounds are violated….forget about it….


  20. Bob McNees says:

    I know I’m late to the party, but I feel like I need to comment on this. There are four versions of the paper available at the arXiv. None of them contains the phrase “test of string theory”. They do, however, refer to the potential for “falsifying string theory”.

    I’m going to assume that everyone here understands the difference between these two things.

    The “press release” appears to have been put together by UCSD’s news service and subsequently picked up by their counterparts at CMU and UT. In fairness, I can see how some of the quotes, taken without reference to either the rest of the press release or the actual paper, deserve to be argued with. But Peter’s “most outrageously misleading string theory hype of 2007″ also contains these quotes:

    “If the bounds are satisfied, we would still not know that string theory is correct,” said Distler. “But, if the bounds are violated, we would know that string theory, as it is currently understood, could not be correct”

    “In other words, string theory—as articulated in its current form—would be proven impossible.”

    Wow. So basically, the most outrageously misleading statement about string theory for 2007 is that an experiment can not confirm string theory, but it can falsify it. How did such a radical idea ever make it through the peer-review process?

    I can’t imagine why Peter – who is, of course, free to delete this post if he likes – would be upset by the prospect of a legitimate proposal for falsifying string theory (Yes, I know … along with several other theories). I mean, it’s not like he’s getting royalty checks from a book whose title implies that this is not possible, even in principle.

  21. N. Nakanishi says:

    Reply to Amos Dettonville: You seem to misunderstand my claim.
    There are two, mutuall equivalent, perturbation theories, “old” and “Feynman”. The latter is manifestly covariant, that is, if the action is Lorentz invariant then the “Feynman” S-matrix is automatically Lorentz invariant, as long as no new situation is encountered. The necessary new situation encountered is the introduction of complex delta function. This is not an artificial procedure. Indeed, if one calculates the S-matrix by means of the “old” perturbation theory, one obtains exactly the same Lorentz-noninvariant result without introducing any new concept.

  22. Amos Dettonville says:

    Reply to N. Nakanishi:

    Maybe you can help me understand what you’re saying if you would comment on the following quote from Mattingly’s 2005 review paper on Lorentz Invariance (which is cited in the “string falsification” paper):

    “As we have seen, over the last decade or two a tremendous amount of progress has been made in tests of Lorentz invariance. Currently, we have no experimental evidence that Lorentz symmetry is not an exact symmetry in nature. ”

    Do you agree or disagree with this statement? Are you saying there IS evidence of violation of Lorentz Invariance? Or are you saying current theory entails such violation, even though it has not yet been detected? Or are you saying something else completely? If the latter, could you explain the relevance to the subject under discussion?

  23. N. Nakanishi says:

    Reply 2 to Amos Dettonville:
    I never positively assert the violation of Lorentz invariance, but I think that the main subject of this debate is the possible violation of Lorentz invariance. I believe that the spontaneous violation of Lorentz invariance is much more probable than the extra dimensions, string theory, etc. My aim is to remind people of the fact that even if LHC experiment indicates the violation of Lorentz invariance, it is still possible to explain it in the framework of the manifestly Lorentz-covariant QFT.
    (As an experimental evidence of the possible violation of Lorentz invariance, some people point out the observation of extremely high-energy cosmic-ray particles, because they should collide with 3K cosmic background photons.)

  24. woit says:


    You totally ignore the scientific argument repeatedly made here that the Distler et. al. paper does not provide a way of falsifying string theory (an argument that the PRL referee seems to have agreed with). While ignoring the scientific issue, your only argument is to make a sleazy accusation that I’m just trying to make money.

    Completely pathetic.

  25. Bob McNees says:

    I don’t ignore the argument, I disagree with it. Perturbative string theory gives you an S-Matrix with well-defined properties. If you demonstrate a violation of these properties, you have falsified the theory. Yes, a lot of other things go along for the ride. If you see the kinds of violations they discuss, it would be hard to reconcile those observations with conventional QFT. One way of falsifying string theory is showing that properties it should have at all scales are violated at the scales you have just gained access to.

    That is a scientific argument. You are free to say “they should come up with something that targets string theory more precisely, as opposed to such a wide-swath of theory space”. Sure, that would be great. But it doesn’t make their argument unscientific.

    Is it the press release you disagree with? The way it might be interpreted in the popular media? Because that’s where I think you are being hypocritical. Where was your indignation when the press releases were being issued on your behalf? You just wrote a book describing your views on string theory, and you gave it a title implying that string theory cannot be falsified. The publisher issued press releases and advertisements, as publishers are wont to do, claiming that you *show* that string theory cannot be falsified. I read your book and considered your complaints, and I don’t agree that you have demonstrated a “lack of falsifiability”. Why do you get to announce that in a press releases?

    I don’t think you are “just trying to make money”. I think you genuinely want people to know what you think about string theory. You just chose a forum that uses advertising and press releases to drum up interest, and now you’re upset that it might cut the other way. At least people can download and read the peer-reviewed article for free.

  26. Peter Woit says:


    You continue to ignore the scientific argument I was making. Perturbative string theory is already falsified, it makes predictions that disagree with experiment.

    I gave my book a title implying that string theory cannot be falsified, because, in its current state, the way it is being pursued, it can’t. I’m willing to stand by that claim, and have spent a lot of time discussing and defending it here and elsewhere and will continue to do so.

    The marketing materials prepared by a publisher trying to sell a book are not directly comparable to a press release issued by a university on behalf of one of its faculty members, but if you want to make that comparison, fine. I don’t think I’ve seen everything the marketing people at my two publishers have written and may be using, but I am willing to stand behind everything that they’ve shown me and asked me to look over. This is mainly the jacket copy on the books. I argued with the US publisher and insisted that they use language that, while it was not the way I would want to present my arguments, was language that I could stand behind and back up. I assume that the authors of this paper similarly are willing to stand behind what has been issued in their name. If they’re not, it would be a good idea for them to say so publicly.

    It’s a very simple distinction: String theory IS “not even wrong”, it is not testable in its current state, and honest researchers in the field acknowledge this.

  27. D R Lunsford says:

    Bob McNees said

    None of them contains the phrase “test of string theory”. They do, however, refer to the potential for “falsifying string theory”. I’m going to assume that everyone here understands the difference between these two things.

    This is just sophistry. There is no difference other than one of language. I am going to assume you know the difference between actual physical experimentation and the vagaries of language.


  28. Jack Lothian says:

    My wife has had hundreds of press releases issued in her name & she has read everyone of them before release & edited most of them before release. This is a pretty common practice. The idea that press releases are generated by faceless gnomes who never communicate with the so-called authors of the release is not a true picture of how most press releases are issued. Shame on you for suggesting this, even our lowest politicians know that they can not get away with this kind of defense. It ranks right up there with the dog ate my homework excuse. I believe the odds are that these authors saw this press release before it was sent out & they approved it. Thus the authors publically endorsed the view that their paper presented a “test of string theory”.

    I agree with Lunsford that your arguments are a form of sophistry.

  29. urs says:

    Peter wrote:

    Perturbative string theory […] makes predictions […].



  30. Q says:

    The first […] is ‘is already falsified,’
    The second […] ‘that disagree with experiment’.

    It’s supersymmetric unification scheme predicts a massive cosmological constant.

    The SU(3) force predicted by string for low energy, using the known SU(2) and U(1) force strengths, is wrong by current experimental data (four times outside the experimental standard deviation).

    Should Peter have called his book simply ‘Wrong’? ‘Not Even Wrong’ refers only to the non-falsifiable ‘predictions’ like unification of all forces at an energy of 10^19 GeV, extra dimensions rolled up in Planck scale manifolds, prediction of gravitons, superpartners, branes, etc., that can’t be checked falsifiably by experimentation…

  31. woit says:


    If you actually have an argument, it would be helpful if you would make it. I’ve made one: purely perturbative string theory doesn’t give standard model physics, it’s wrong. Attempts to get around this by invoking complicated backgrounds, generated by non-perturbative effects like branes are non-predictive and not even wrong. If you have an answer to this argument, let’s hear it.

  32. Thomas Larsson says:

    Wrong, completely wrong, not even wrong.

    Bosonic string, superstring, M-theory.

  33. Pingback: Bloggernacle Times » This Week in Science and Religion

  34. Let’s say Gross comes up with some very good math to link string theory to QCD (perhaps giving QCD it’s own emergent spacetime while string theory has some other spacetime)… would this then allow Gross to justifiably claim QCD as a test of string theory?

  35. Peter Woit says:

    The “test of string theory” discussed here has nothing to do with a string theory dual for QCD. There may very well be such a thing, and if it can be shown to be exactly dual to QCD, then every test of QCD would be a test of that string theory. The claims in the Distler et al. paper are not about this, but about testing the idea of string theory as a unified theory of particle physics and gravity

  36. Pingback: Not Even Wrong » Blog Archive » Shameless Enthusiasm

  37. Arun says:

    What’s your opinion of this “verification of string theory” method?

  38. woit says:


    I haven’t looked at the paper very carefully, but I guess Shiu/Underwood. are claiming that in certain brane-inflation models, for certain ranges of parameters, you can in principle see effects of the warped compactification on the CMB. I have no idea whether these effects are something one could imagine practically measuring anytime in the forseeable future for realistic parameter values.

    They don’t claim in the PRL paper to be able to “test string theory”, but they have issued a press release, which seems to me an unwise thing for theorists to do to promote this kind of very speculative result:

    The press release is entitled “Physicists find a way to “see” extra dimensions”, which is nowhere near as misleading as the Distler et. al. press releases. It does contain claims about string theory that are likely to mislead people, especially Shiu’s quote that “This provides a rare opportunity in which string theory can be tested.” It would have been a good idea for him to make clear that string theory is not being tested, but a very specific “string-inspired” model of extra dimensions, and that not seeing the effects they study (which is extremely likely) doesn’t in any way provide evidence against string theory.

    This is kind of like the endless claims one heard after Randall-Sundrum that the LHC would “test string theory” since in principle extra dimensions could be of TeV scale, and in principle they could come from string theory. People seem to have stopped making that particular “test of string theory” claim for one reason or another.

    It will be interesting to see how many inaccurate press stories are generated by this press release.

  39. Pingback: Not Even Wrong » Blog Archive » Odds and Ends

Comments are closed.