Grading String Theory

Commenter Shantanu points to this video of a recent colloquium by Andy Strominger at Harvard, which includes some extensive comments on the current state of string theory. Strominger is one of the most prominent string theorists in the business, and has been working in the field for more than a quarter-century since the first “Superstring Revolution” of 1984. In the talk (at about the 52 minute point), Strominger gives a “report card” for string theory, where he assigns it 3 As, 2 Bs, 3 Ds and 2 Fs, for an average grade of about C. It gets an F for making no unambiguous testable predictions, a D for prospects of saying anything about LHC, an F for the CC (Strominger isn’t sold on the anthropic landscape) and a D for cosmology. Some of the high grades are debatable, with an audience member pointing out to him that there was a tension between his A for “Not being ruled out as theory of nature” and F for no testable predictions. Strominger repeatedly claimed that most string theorists would agree with him on these grades (except maybe the F for the CC).

As an overall evaluation, he said that it was debatable whether this was a passing or failing report card, then arguing:

But this is the only student in class, so if you flunk her you have to shut the school down.

Along the same lines, a bit earlier in the talk one of his slides characterizes all that theorists can do as “go home and watch TV” if they believe in the landscape (“String theory is everything”) or if they think string theory is a failure (“String theory is nothing”). The positive argument he was trying to make is that there still is something for string theorists to do even after they are forced to give up on particle physics: they can try applying AdS/CFT and black holes to other areas of physics (nuclear physics, solid state physics, fluid mechanics).

I think Strominger is right that his grades and point of view about string theory are now conventional wisdom among leading theorists. What I find striking about this is the argument that if you are forced to give up on string theory, you have to “shut the school down” or “go home and watch TV”. More than 25 years of working on string theory has left Strominger and others somehow believing that there is no conceivable alternative. The failure of string theory as a theory of particle physics leads them to the conclusion that they must not abandon string theory, but instead must abandon particle physics and try and apply string theory to other fields. The obvious conclusion that string theory is just one speculative idea, and that its failure just means you have to try others, is one that they still do not seem willing to face up to.

This entry was posted in Uncategorized. Bookmark the permalink.

42 Responses to Grading String Theory

  1. Kea says:

    Yes, it is fascinating to see people dictating to Nature the impossibility of alternatives, merely on the assumption that nobody else could possibly do better than they have. I feel that Nature would rather we were somewhat more productive than TV couch potatoes, since TV watching uses less brain power than sleeping. Well, Nature has a simple cure for this kind of behaviour, othewise known as Extinction.

  2. Anonymous says:

    So what’s the alternative theory that gets better grades on these issues?
    By the way, abandoning particle physics is not really a bad idea, if you accept the premise that a field of inquiry should only receive resources proportional to the amount of new experimental evidence arising in this field.

  3. hsearles says:

    Reminds me of the Karl Popper quote: “Whenever a theory appears to you as the only possible one, take this as a sign that you have neither understood the theory nor the problem which it was intended to solve.”

  4. Peter Woit says:


    You could also argue that fields should get resources inversely proportional to amount of new experimental evidence, on the grounds that if there are lots of new experimental clues, any unfunded idiot will be able to make progress. It’s only when there are few experimental clues that the work is hard, so should get more resources.

    Also, by your criterion, mathematical physics should never be funded at all.

    The argument that “sure, string theory is a failed idea, but since we don’t have better ones, we’re going to keep doing it anyway” has been the main thing keeping string theory research afloat for a very long time now. It wasn’t a good argument 15 years ago, and with the landscape, etc., it’s not getting any better.

  5. The alternative theory with better grades is the Standard Model.

  6. Yatima says:

    That’s quite an interesting little talk but it’s hard to figure what the linkage of black hole theory or string theory to solid state physics, fluid dynamics and whatnot is all about. It’s also difficult to ascertain what the author classifies as belonging to string theory and what not. A colored diagram with the mathematical bridges linking the main idea blocks and equations would be useful.

  7. flunkemall says:

    The Standard Model is the baseline. The challenge is to find a successful extension. That determines the grades.

  8. Dave Miller says:


    What do you think of the view that string theory is simply a framework for theories, but that you need some additional ideas to narrow it down to get a specific theory? It’s an analogy to QFT: i.e., no one criticizes QFT on the grounds that thee are an infinite number of QFTs. We just accept QFT as a framework that, fortunately, includes one specific example that works extremely well — the Standard Model.

    Perhaps even string theory enthusiasts should accept the fact that the community just needs to “sleep on it” for a few years to come up with some novel ideas. After all, QFT goes back all the way to 1930 or so, but it wasn’t until the ’60s that the Standard Model was actually invented.

    Dave Miller in Sacramento

  9. neo says:

    I, for one, do not think physicists and mathematicians should give up on string theory. They should stop the hype and stop blocking the consideration of alternatives, but not give up. One should wonder though, absent evidence, when you go down a path that gets ever more convoluted, whether you are on the right path. At least the QFT folks were led into a labyrinthic theory by the existence of hard facts.

  10. DB says:

    “The obvious conclusion that string theory is just one speculative idea, and that its failure just means you have to try others, is one that they still do not seem willing to face up to. ”

    Indeed it is. But Strominger is 55 years old and therefore completely passed it in terms of making fundamental new contributions to theoretical physics. So for him to abandon String Theory is to acknowledge that his whole research career has been a failure. Since String theory has been going for thirty years most of his colleagues are in the same boat. Psychologically he, and his fellow workers can’t face up to this. And practically speaking, it would mean the end of their mentoring role – the only meaningful role they have left. And this, more than any other factor, explains their bizarre committment to a failed theory.

  11. Jose says:

    “You could also argue that fields should get resources inversely proportional to amount of new experimental evidence…”

    …in which case string theory should get infinite funding.

  12. I know that this is not “a place for people to promote their favorite ideas about fundamental physics”, but surely the names “loop quantum gravity”, on the quantum gravity side, and “noncommutative geometry”, on the unification side, must have come to the ears of Strominger and other promoters of the TINA argument. I would understand if they said “there is no good alternative”, since this would clearly be the expression of an opinion, but choosing instead not to even mention the other theories surely qualifies as propaganda.

  13. M Brane says:

    In many ways, string theorists are like wives and string theory itself is like the husband that beats them.

    “But I love him”
    “I can’t leave him, because where would I go?”
    “He only does those things because he’s incomplete, and I can FIX him”
    “I’m too old to start seeing someone new”

    However in some violent relationships, the surprising happens: The husband leaves. So it may end up that string theory itself may leave the theorists and not the other way round.

    Indeed since 1998 , when string theory said “I’m just going to the shop to get some CC-igaretes” , the theorists have been waiting for it to come back by keeping it’s dinner warm in the landscape oven.

  14. Seth Thatcher says:

    M brane:

    I think the abandoned wife of string theory has lately been internet dating “Multiverse” in “String theory’s” absence to pick up CC-igarettes. “Multiverse” sounds good and is quite interesting, but he is elusive and refuses to meet her at the local Starbucks for a face to face. She’s never seen him and can’t verify who he is or that she he even really exists, but she’s hopeful.


  15. Peter Woit says:


    If you’re going claim to just be a “framework”, the question becomes whether you’re a useful framework, i.e, whether you’re able to provide a testable scientific explanation of anything. Claiming that “God did it” is also a framework, just not a very useful one.

    The QFT framework historically used by physicists is not “pick any random, arbitrarily complicated QFT and try and use it. What is really powerful is the idea: pick one of a small number of the simplest QFTs and compare these to the world. This is what physicists have been doing since the earliest days of quantum theory, with great success in terms of getting a scientific explanation this way. The simplest string theories are just dead-wrong, so you’re forced to look at more and more complicated ones, evading confrontation with experiment.

  16. So it appears that Strominger, Woit, Motl, and everyone else are in broad agreement about what is string theory, and what it has and has not accomplished. Motl even agrees with Strominger’s report card. Controversy only arises when someone argues that a failed theory is still worth pursuing if all the experts say that it is the only game in town.

  17. Smurf says:


    …therefore completely passed it in terms of making fundamental new contributions to theoretical physics…

    Hidden conformal symmetry of the Kerr black hole (2010), Kerr/CFT correspondence (2008), Chiral Gravity Conjecture & its proof (2008), OSV conjecture (2004) & its proof (2006), 4D/5D black hole connection (2005),… Enough for you?

  18. JC says:

    At this point, I doubt people like Witten, Strominger, etc … are going to publicly renounce string theory. Nobody wants to admit their entire research career was all in vain.

    From a more pragmatic perspective, they don’t want to “nuke” the careers of their younger colleagues, previous students, and postdocs, whom are still climbing up the academic system (ie. looking for an assistant professor job, on tenure track, etc …). They also don’t want to give the funding agencies any more excuses to reduce or outright cancel funding for string theory.

    With that being said, the earliest I can think for somebody like a Witten or Strominger to renounce string theory, would be after they are retired and no longer have their offices at the university, IAS, etc …

    For a die hard fanatical string theorist, they probably will not renounce string theory at all. Though if they do renounce, for whatever reasons (ie. seeing the light, etc …), it will most likely be when they are on their deadbeds.

  19. Dave Miller says:

    Peter wrote to me:
    > The QFT framework historically used by physicists is not “pick any random, arbitrarily complicated QFT and try and use it. What is really powerful is the idea: pick one of a small number of the simplest QFTs and compare these to the world. This is what physicists have been doing since the earliest days of quantum theory, with great success in terms of getting a scientific explanation this way. The simplest string theories are just dead-wrong, so you’re forced to look at more and more complicated ones, evading confrontation with experiment.

    Yeah. I think what probably will happen, and probably should happen, is that people will gradually drift away from string theory, simply because no one really knows what to do with it. However, it seems to me quite possible that, forty years from now, some bright young kid will be working on LQG or non-commutative geometry or whatever the hot thing is then, and will see that it connects with string theory in a cool way, and that this was the missing idea that kept string theory from being useful.

    Stranger things have happened: it took a long time after Riemann and Gauss worked out the initial ideas in differential geometry until Einstein saw how to use those ideas in general relativity. Of course, GR has an indefinite metric, as suggested by special relativity, and this gives a structure not imagined by Riemann or Gauss. It seems to me that something like that may happen someday in string theory: some old ideas banging around from string theory may mate with some entirely new ideas from some other source and produce some viable offspring.

    I don’t think I’m really arguing with you here: I don’t think you’ve ever suggested that we should simply burn all the books and journals with string theory stuff in them to prevent future generations from being polluted with string theory! Obviously, string theory has not worked, but I do not think it is crazy to suspect that some ideas from string theory may eventually make it into real elementary-particle physics. At any rate, I doubt it is possible to hide the stuff that has been worked out in string theory from future generations of theorists, even if they find it of limited interest.

    The other point I’d like to raise is basic math issues about string theory: I’m still bothered by the different quantization procedures (Gupta-Bleuler-like quantization, light-cone gauge, etc.), by the question of a true string field theory, by whether or not “M theory” actually exists, and, indeed, by the question of what string theory actually looks like in spacetime (the jump from 2-D quantization in the sigma-tau plane to actual 9+1 or 25+ 1 or whatever-dimensional spacetime is slid over pretty slickly in everything I have seen).

    You recall Streater and Wightman’s “PCT, Spin and Statistics, and All That”? There was nothing about real-world field theories there (certainly not gauge theories), just an attempt to figure out what sort of critters field theories are.

    It seems to me that there are a few decades (at least!) of similar work in string theory, and, to my tastes, that might be worthwhile. Most of us don’t pooh-pooh the work put into proving that phi^4 theory is trivial in 3+1 dimensions: it did not teach us anything about the real world, but it did teach us something about field theories.

    Similarly, if someone could actually present “M-theory” in a well-defined way and show that it exists mathematically, I’d find that interesting as work in mathematical physics, even if it never connects with the real world.

    Alas, string theory does not seem to attract too many people with the personalities of Streater and Wightman.


  20. Tom Whicker says:

    Chiral Gravity has been proved? Does this mean that Uncle Al has been vindicated???

  21. Anonymous says:

    Similarly, if someone could actually present “M-theory” in a well-defined way and show that it exists mathematically, I’d find that interesting as work in mathematical physics, even if it never connects with the real world.

    By AdS/CFT, this is probably at least as hard as showing the existence of Super Yang-Mills theory. What a hopelessly difficult challenge.

  22. Dave Miller says:

    Anonymous wrote, re my earlier post:
    >By AdS/CFT, this [showing M-theory exists] is probably at least as hard as showing the existence of Super Yang-Mills theory. What a hopelessly difficult challenge.

    I should have been clearer with my Streater/Wightman example. I don’t expect anyone to work out string theory with the level of math rigor that Streater/Wightman used.

    But, it would be nice to have some demonstration that there is such a thing as M-theory at the same level of understanding we all used to accept for field theory – some single Lagrangian or *something* that really is a single presentation of M-theory. As far as I can tell, there have been rather wild jumps to possible phenomenology, brane-worlds, the landscape, and all that, without even knowing if M-theory exists in terms of the relaxed standards of rigor physicists usually accept (or used to accept).

    Similarly, “first-quantized” string theory seems actually to be a classical field theory of strings (just as Klein-Gordon theory actually is a classical field theory that needs quantization). However, if you try to work out the details based on the standard approaches – light-scale “quantization,” Gupta-Bleuler-like imposition of the constraints, etc. – well, I have not seen that done, and it looks to me to be fairly difficult.

    I suspect that some string theorists would claim that this is obvious, and that I am just too dumb to see it.

    Maybe so. In which case, surely one of them could have mercy on us poor dumb theorists and write it out for us at the same level as any of the old classic field-theory textbooks.

    I don’t think they can. I have most of the standard string-theory texts – GSW, Polchinski, etc. It’s just not there.

    If string theory is real, even at the level of mathematical physics, it should be possible to work this stuff out and write it up. Trying to do so might clarify a lot of what is going on.

    I’m an agnostic on this: I don’t know if the basic ideas of string theory make sense or not.

    But, it would be nice if someone were actually trying to work this out.


  23. anonymous says:

    What grades does he claim are A’s and B’s?

    There’s a huge difference between LHC and cosmology.
    LHC physics, while very important, is at an energy scale (TeV) for which string theory
    may well have little to say. In very early universe cosmology there is a greater possibility for connection with data, since the subject depends on high energy physics (Planck-suppressed operators). In this subject, there are testable signatures of some well-defined theories; moreover the theory has motivated
    new, more systematic studies of field theory and CMB signatures
    which are central in that field. The speaker has not worked on this much, and may not have thought carefully about it.

    It is surprising that he would claim to speak for others on this; I would guess the grades would vary wildly.

  24. Peter Woit says:


    Strominger’s grading of string cosmology with a D I think accurately reflects both its lack of success and the general opinion of the subject of cosmologists, string theorists and others. The only disagreement among most physicists would be whether a D or an F is the right grade. Trying to claim that he doesn’t know what he’s talking about is just absurd.


    I think giving a public colloquium in which you award your own subject Ds and Fs for its progress in saying anything about particle physics and describe the future of this subject as its adherents having to go home and watch TV is about as close to admitting failure as you are ever going to get. What’s at issue now is what you do after admitting this failure. Strominger is arguing that theorists should leave particle physics and try applying techniques they have learned from string theory to other subjects. I’d be happier if he and others argued that people should be looking for new techniques.

  25. DB says:


    “From a more pragmatic perspective, they don’t want to “nuke” the careers of their younger colleagues, previous students, and postdocs, whom are still climbing up the academic system”

    I think it’s more insidious that this. The reason, I believe, for the relentless incantation of the mantra “this is the only game in town” is a cynical pr-style attempt to influence young talent into taking up String Theory. I think this is the only rational explanation for their behaviour. The old burnt-out hands, who now sit in very influential positions, know that the only real chance for progress is to attract the young talent that might be capable of making the breakthrough that they have failed to make and are now too old to realistically hope of achieving. The alternative is that their long toil of thirty years of research is consigned to the dustbin of history.

    Putting myself in their place, I can see why they are driven to relentlessly spin and hype their field. If by some miracle a young genius was to sort out what appear to be completely intractable problems with the field, the shameful history of hype and spin would soon be forgiven and they would be celebrated as pioneering visionaries.


    Your filter as to what constitutes “fundamental new contributions to theoretical physics” appears to be set at a considerably lower threshold to mine. Historically, really important new breakthroughs by theoretical physicists over the age of 45 are extremely rare.

  26. piscator says:

    Not sure that anonymous really appreciates particle physics. The inflationary scales are no more Planckian than the TeV scales.

    In large classes of models (all versions of gravity mediation), *all* the soft terms – i.e. the terms in the Lagrangian that in the best of all possible worlds the LHC will measure – arise as Planck-suppressed operators.

    In both susy and cosmology, understanding the physics requires understanding Planck-suppressed operators. How well defined the models are is another question, but if you want to understand the origin and physics of these Planckian operators, there is still only one game in town (sorry Peter).

  27. anonymous says:

    piscator: TeV physics may or may not involve SUSY, which if it exists may or may not involve gravity mediation. Inflationary cosmology necessarily involves Planck-suppressed operators, in the cases protected by symmetry, an infinite class. However I do agree with you that particle physics is very interesting and we could get lucky in terms of connections to string theory. SUSY grew out of string theory as well, as a spinoff.

    Peter: actually cosmologists do take this seriously. WMAP papers and various cosmo white papers written by the experts in the field refer at an appropriate level to string theoretic mechanisms and their corresponding signatures, for standard reasons. Moreover, string theory contributed in a clear way to the development of the field theoretic classification of single-field inflation mechanisms.
    This is all standard by now in the cosmo community. Unfortunately, it is not absurd to say that string theorists including perhaps Strominger are not all caught up on this, but CMB physicists by and large are.

  28. piscator says:


    OK, we aren’t that far from each other. Personally I feel cosmo people place too much emphasis on string-motivated ideas, and pheno people too little, but we can agree to differ.

    Inflation is in some sense twice as hard as susy breaking though, because for inflation you need controlled susy breaking to start with, and then controlled inflationary dynamics on top of that.

  29. careful says:

    Wow. The one thing I would think all theorists would have learned by the age of 55, is to never claim to speak for a whole community.

    I know a lot of string theorists. I have even been called a string theorist on occasion. I am not sure what “grades” I would give this huge framework (which in my opinion has contributed a lot to theoretical physics, but is not yet in anything close to a finished form). But I am pretty darned sure that if you took a poll of string theorists, you would get a nice, diverse statistical distribution of answers about where the theory has done best/is most promising, and where it has been most deficient (or what areas are most overgrazed to little continuing benefit).

    I suspect Strominger, for instance, might emphasize calculations of highly idealized “black hole” entropies (counts of field theory BPS states) as a great success, where others might instead be happiest with models of cosmology (which promise eventual contact of a more physical nature), or particle physics, or geometric engineering of various known and previously unknown field theory results, or even relations to pure mathematics. I am not advocating any of these views, but it is clear from the arXiv that many people view each as a prime motivation to work in this area.

    This is as it should be; since there has been no smoking gun success at the level of making contact with nature, any agreement would be purely sociological, and I’m happy to say that we haven’t reached that level of groupthink yet.

  30. vasisdas says:

    Amusing colloquium but very political, with lots of preferential references to present and former Harvard people. Does he really believe that black holes are the harmonic oscillators of the 21st century? They seem pretty 20th century to me, just like Strominger’s powerpointless slides. I doubt that referencing Subir repeatedly for using black holes to solve CMT will help extend the lifetime of this peculiar fad. Of course, trying to explain the Big Bang is what string theory should be good for. He might not like the progress on this so far, but this is a far more likely arena for eventual string theory contact with real world than LHC energy scales. If LHC discovers some new physics that will be great, but having it distinguish between different BSM scenarios (let alone string constructions) is unlikely. So, grading both as D looks misleading.

  31. Shantanu says:

    Peter, here is another interesting talk by Avi Loeb on advice to young astrophysicists. Although this is more astrophysics based(although string theory and landscape is mentioned), the same method applies
    to particle physics also
    Someone should give a similar talk about particle physics

  32. Anonymous says:

    As well as being an accomplished scientist, Loeb has a very open mind. Among other points, he thinks dark matter/dark energy may not eventually be proven correct, and our current understanding of cosmology leading to the theory of galaxy formation may well turn out to be another theory of epicycles.

    Ironically, string theorists seem to be absolutely certain that there is a cosmological constant, while astrophysicists, who actually study this issue directly, at least do not view competing theories like quintessence as crackpottery.

  33. lun says:

    Dave Miller’s comment is very appropriate given Witten’s paper of a few days ago

    Strictly speaking, it is not on string theory at all, let alone of string theory as a theory of quantum gravity. However, objects which arise naturally in string theory are shown to be somehow equivalent to objects which arise in certain quantum field theories and even quantum mechanics.

    Seiberg’s dictum, if we find something beyond string theory well call it string theory, is not COMPLETELY unjustified.

  34. Chris W. says:

    The writeup that Avi Loeb refers to in his talk is arXiv:1008.1586.

  35. Geoff says:

    Hi Peter –

    My bias is that string theory does too much violence to GR, but there is an interesting problem that makes me wonder if string theory isn’t entirely useless from a computational standpoint. If I had been a student of DeWitt and he asked me to calculate graviton-graviton scattering as described here:

    it would have taken over 500 terms and I’m quite sure I would have never gotten it right. If someone told me there was a quick way to get the right answer with only one term, I certainly would have used it without regard as to whether the underlying theory solved the unification problem.

  36. Coin says:

    “Among other points, he thinks dark matter/dark energy may not eventually be proven correct”

    Hm. Loeb’s general philosophy on research is fascinating, but I’m surprised to find people still doubting dark matter. I’d be curious how any alternative to the dark matter hypothesis could possibly be made compatible with the bullet cluster/MACS J0025 observations…


    “SUSY grew out of string theory as well, as a spinoff.”


  37. Peter Woit says:


    Sure, the applications of string theory giving sums of lots of Feynman diagrams in terms of one world-sheet diagram are impressive. It’s worth pointing out that this is still a non-trivial calculation (to really connect to GR, you should quantize the 10d superstring and then extract 4d info from it).
    Lots of things coming out of string theory are useful, just not the ones coming out of conjectured unification via string theory.

    I don’t know who Yau’s GR instructor was, don’t think he says in the book, may be wrong.


    Dark matter has nothing to do with this posting, which is one of the problems with string theory unification…

    I wrote extensively about the history of 4d SUSY in my book. It was first discovered in the Soviet Union, through motivations having nothing to do with string theory. Later it was independently discovered in the West, there string theory was part of the motivation.

  38. np says:

    It’s NP (Physics) day today. Not string theory by any stretch of the imagination, but I suggest it’s time for a new post.

  39. Peter Woit says:


    Sorry, but the Nobel news is about something I know nothing about. Earlier this year I was at a series of talks which was supposed to include Geim, but he didn’t make it because of the volcano. So, I remain ignorant and I fear you’ll have to find better informed blogs for insight on today’s news.

  40. zanzibar says:

    RE: Nobel prize (graphene) and Black Holes (!?)

    Interestingly, Motl’s blog on the Nobel prize contains material apropos this thread’s topic:

    Just four days ago, graphene was also mentioned on TRF because it appeared in Andy Strominger’s talk – as an example of a physical system for which the string-theoretical AdS/CFT correspondence has been useful.

    That’s partly because graphene is a nearly perfect fluid – much like the other systems that are well covered by the holographic dual (and reinterpreted as a black hole). This fact is just another example of the tight relationships with string theory and its brand of quantum gravity that the most important state-of-the-art discoveries even in as faraway disciplines as atomic or condensed matter physics display.


    (Sorry about the double posting, but how do I quote material via this blog’s editing interface?)

  41. ohwilleke says:

    The good news is that there actually more kids in the classroom (LQG and entropic gravity most notably) than there were a decade ago.

  42. MathPhys says:

    I personally liked Strominger’s seminar and it is clear to me that he has thought about the issues that he was talking about, deeply and for a very long time.

    It is natural that he referred frequently to contributions by Harvard people. The seminar took place at Harvard, the audience were mostly Harvard students and he was being patriotic, so to speak. It is also a fact that he was mostly talking about traditional, non-stringy physics.

    His opinions on string theory in the final part of the talk were, well, his opinions and he’s entitled to them. What’s he expected to say? Someone else’s opinions? I think he was very reasonable.

Comments are closed.