“So what will you do if string theory is wrong?”

In a new preprint of an article entitled “So what will you do if string theory is wrong?”, to appear in the American Journal of Physics, string theorist Moataz Emam gives a striking answer to the question of the title. He envisions a future in which it has been shown that the string theory landscape can’t describe the universe, but string theorists continue to explore it anyway, breaking off from physics departments to found new string theory departments:

So even if someone shows that the universe cannot be based on string theory, I suspect that people will continue to work on it. It might no longer be considered physics, nor will mathematicians consider it to be pure mathematics. I can imagine that string theory in that case may become its own new discipline; that is, a mathematical science that is devoted to the study of the structure of physical theory and the development of computational tools to be used in the real world. The theory would be studied by physicists and mathematicians who might no longer consider themselves either. They will continue to derive beautiful mathematical formulas and feed them to the mathematicians next door. They also might, every once in a while, point out interesting and important properties concerning the nature of a physical theory which might guide the physicists exploring the actual theory of everything over in the next building.

Whether or not string theory describes nature, there is no doubt that we have stumbled upon an exceptionally huge and elegant structure which might be very difficult to abandon. The formation of a new science or discipline is something that happens continually. For example, most statisticians do not consider themselves mathematicians. In many academic institutions departments of mathematics now call themselves “mathematics and statistics.” Some have already detached into separate departments of statistics. Perhaps the future holds a similar fate for the unphysical as well as not-so-purely-mathematical new science of string theory.

This kind of argument may convince physics departments that string theorists don’t belong there, while at the same time not convincing university administrations to start a separate string theory department. Already this spring the news from the Theoretical Particle Physics Rumor Mill is pretty grim for string theorists, with virtually all tenure-track positions going to phenomenologists.

I have some sympathy for the argument that there are mathematically interesting aspects of string theory (these don’t include the string theory landscape), but the way for people to pursue such topics is to get some serious mathematical training and go to work in a math department.

The argument Emam is making reflects in somewhat extreme form a prevalent opinion among string theorists, that the failure of hopes for the theory, even if real, is not something that requires them to change what they are doing. This attitude is all too likely to lead to disaster.

Update: A colleague pointed out this graphic from Wired magazine. Note the lower right-hand corner…

Update: Over at Dmitry Podolsky’s blog, in the context of a discussion of how Lubos’s blog makes much more sense than this one, Jacques Distler explains what it’s like for string theorists these days trying to recruit students:

Unfortunately, I’ve seen a number of prospective graduate students, who spent their undergraduate days as avid readers of Woit’s blog, and whose perspective on high energy physics is now so hopelessly divorced from reality that the best one can do is smile and nod one’s head pleasantly and say, “I hear the condensed matter group has openings.”

This entry was posted in Uncategorized. Bookmark the permalink.

87 Responses to “So what will you do if string theory is wrong?”

  1. Moshe says:

    Not sure I agree this is the only interest in these systems, but in any event there is currently a new set of ideas of dealing with strongly interacting systems. It may be of some interest for people with experience dealing with such systems to explore these ideas and discover what they are good for.

  2. Peter Woit says:


    The problem with the argument that physics departments have only stopped hiring string theorists this spring is because of the LHC is that the LHC turn-on has been imminent for quite a few years, and departments that wanted to move in that direction have been doing so for quite a while.

    However, it is only this spring that string theorists have started posting articles on the arXiv about what they plan to do once it is shown that string theory is wrong…

  3. Peter Orland says:

    Sure, new things need to be explored. The ideas you mentioned aren’t being neglected, while there are other approaches to strongly-interacting problems which have excited less interest, possibly because they don’t start from string theory (using exact form factors is one of those of which I am familiar).

  4. Peter Orland says:

    Moshe – If you are interested, the form factor program is reviewed in
    an excellent article by Essler and Konik.

  5. Moshe says:

    Thanks Peter O., I’ll take a look.

  6. Eric says:

    I don’t think you can claim that physics departments have stopped hiring string theorists. For example, Hawaii has just made an offer to a string theorist, and I’m sure others will be forthcoming.

  7. Eric says:

    Again, what about Lehman College, CUNY? Can we not consider Dan Kabat a string theorist?

  8. Peter Woit says:


    I’m not claiming that no physics department is hiring any string theorists, just that dramatically fewer are getting tenure-track jobs this spring than was typical in recent hiring seasons. I’ve given you my source for this information: the rumor-mill web-site. Go check it for yourself.

  9. Eric says:

    Perhaps you should have a closer look at the rumour mill site yourself, as the information I was just giving you came directly from there. Even if there are fewer string theorists hired this spring, I don’t think this would be that unusual, as these things are cyclic.

  10. Coin says:

    Sorry if this is kind of a dumb question, but this is something I’ve been a bit confused about for a while: What exactly is the difference between a “Phenomenologist” and an “Experimentalist”?

  11. Peter Woit says:


    Experimentalists design, build, operate experiments and analyze the data from them. Deciding which theorists are “phenomenologists” is not so clear, the term covers a wide range of activity. But in general the idea is that they are theorists working not on theoretical or mathematical foundations, not on models far from reality, but on better understanding and extracting predictions from testable models that can be compared with experiment. At one end of the subject, their activities probably become indistinguishable from some of the activities engaged in by experimentalists as part of their data analysis.

  12. Peter Shor says:

    Peter Woit is right; I certainly wasn’t calling for all string theorists to be fired. I was trying to point out that Moataz Emam’s call for string theorists to form their own departments (and I don’t quite know how seriously to take him on this) was not only completely implausible, but it also wasn’t even in the best interest of the string theorists. If anybody tries to form a separate Department of String Theory, I would expect and hope that most of the physicists (string theorists and others alike) in the affected Department of Physics would resist.

  13. Thomas Larsson says:

    Dimitry and Peter O.

    I am well aware that the 3D Ising model is Kramers-Wannier dual to the 3D Ising gauge model and admits a random surface representation. Even better, one can write down a funny O(N) or U(N) lattice gauge model on a 3D brick lattice, with a log action, that can be exactly mapped onto a model of self-avoiding random surfaces; the weight of each graph equals N^chi u^A, where chi is the Euler characteristic, u is related to the coupling constant, and A the surface area.

    Does this construction solve U(N) gauge theory, or make me into a string theorist? Hardly. Just because I have mapped one untractable model into another does not mean that I have solved it. If anything, this shows that perturbative string theory is inadequate for this problem because steric repulsion is crucial here (self-avoiding surfaces do not self-intersect!).

    A method has only really contributed to our understanding of a model if it helps to extract some kind of quantitative information about it, not necessarily critical exponents and not necessarily exactly. The methods that Peter O. mentioned do that, as do high- and low-temperature expansions, real-space RG and computer simulations. However, AFAIK no quantitative results about 3D Ising have come out of string theory.

  14. Peter Orland says:


    As you can see from my earlier comment, I largely agree with your sentiment that string/random surface approaches to field theory/stat mech problems have not been successful. In fact, as I said, it has been much easier to obtain string excitations from field theory than the other way around (except in AdS/CFT approaches, which have not yet made a connection with real physics problems. Maybe the work Moshe referred to will change the situation). I must admit, however, that I still have some fondness for this approach, in spite of its lack of success, since it captured my imagination in grad school. It was a very daring suggestion, which, unfortunately, may not lead anywhere.

  15. As someone with credentials as a scientist and as a writer, I see the argument for splitting String Theory Departments away from Physics Department, to allow Physics Departments to return to the “Real World” and to allow the String Theorists to pursue their quest; as analogous to an argument that Critical Theory Departments split away from English Literature Departments, to allow Eng Lit to focus again on poems, stories, and novels, and be free from Deconstructionism, Feminism, Pan-Africanism, Asian Pacific Islandism, Gay Theory, Postcolonialism, and the like, while the Critical Theory professors and grad students and postdocs explore a rich sociocultural and semiotic landscape unfettered by having to read and explain prose and poetry that naive millions of readers actually enjoy buying and reading.

  16. JC says:

    Is that quote on Podolsky’s blog really by Jacques Distler himself? Or is that quote just somebody else posting under Distler’s name, trying to make him look bad?

    (Offhand, I can’t really tell either way).

  17. Peter Woit says:


    If it’s not Jacques, it’s someone doing a pitch-perfect impersonation….

  18. Gil Kalai says:

    Peter, this thread shows the difficulty in discussing hypothetical scenarios. Emam referred to a hypothetical situation that “someone shows that the universe cannot be based on string theory” so he was certainly not calling to form separate departments now.

    Overall, Enam’s analysis is not convincing. I think he is simply wrong in his assumption that if someone shows that the universe cannot be based on string theory many string theorist will continue to do what they do. Showing that the universe cannot be based on string theory (if this will be the way things will go) will represent a major breakthrough which is one of the potential fruits of string theory research. Probably such a (hypothetical) development will cause most string theory to work on something else (that can still be called “string theory”). Those who will continue to work on string theory (or “old-fashioned string theory”) will have a place in physics department just like many other theoretical physicists who work on models which while not describing the universe, still give important physics insights and develop important mathematical physics. Also Emam’s comparison with statistics does not make sense.

    Here is an analogy from mathematics. For many centuries algebraists tried to find formulas for solving polynomial equations. If somebody had come say in 1600 and said: “guys in view of the many years of failure, try to think about the possibility that such formulas are impossibles” this could be a little useful. If the same guy had said “algebraists have failed and they should be fired” (opening the door to more astrologists and alchemists), then this could be a little harmful. But what was really needed is an understanding why there are no formulas to solve polynomial equations and not just vague speculations and meditations.
    Once this was understood nobody continued the old algebra endeavor but there was plenty of work left for algebraists, in fact, this was the beginning of modern algebra.

    The turning point and the beginning of modern algebra was not just the “gut feeling” that the old endeavor might have failed but the detailed technical understanding why it cannot succeed.

  19. Peter Woit says:


    String theory, as an idea about unification, is not in danger of failing because of mathematical contradiction. It is failing because it is becoming clear that, as an idea about unification, it is essentially vacuous. The reasons for this are well understood (e.g. the landscape). I don’t think the author of this article or anyone else envisions string theory being shown to be mathematically wrong, or being falsified by showing that it predicts something that disagrees with experiment. What is happening is that conviction within the physics community is growing that there is no hope to get a real prediction out of string theory, thus the failure. Die-hards will keep claiming that hope remains, but there will be fewer and fewer of them, and at some point the consensus of the community will be that the cause is hopeless.

    Right now, I think many of the leaders of the subject are well aware of how bad the problems are, but are holding out from giving up on the theory, waiting to see what the LHC says, hoping that LHC results will change the current picture. If the LHC doesn’t come up with something that somehow gives new hope to the string theory unification program, I think it will be conclusively dead, although some people will insist on pursuing it. I would guess that it is exactly this all too possible situation that Emam is thinking about when he brings up the possibility of string theory being shown to be “wrong”.

  20. Gil Kalai says:

    Dear Peter,

    It is entirely rational to wait to the LHC outcomes and I think most people interested in the scientific endeavor are hoping that it will give new data and lead to important new insights. I have noticed that your opinion on string theory is negative, but your description of the convictions within the physics and string theory communities may very well be biased by your own personal position. At least your description differs from the convictions I sense when talking to many colleagues in these areas. (My little comment to Peter (Shor) was that Enam does not call string theorists to form independent departments but rather discusses a scenario which is hypothetical from his point of view.)

  21. Kea says:


    It is not at all reasonable to wait and see what happens at the LHC. Any responsible tax funded theoretician should be working their ass off to come up with quantitative predictions before 2009.

  22. Thomas Larsson says:

    Peter O,

    I am sorry if my post came across as more negative than it was meant to be. The random surface approach also captured my attention a few years after it captured yours (when I was in grad school), and I find it frustrating that it has not panned out. When I now think about the problem again after many years, it seems like the obstacle is the self-avoidance constraint, which induces an interaction which is very non-local on the worldsheet, although it is local in the ambient space.

  23. Gil Kalai says:

    Peter, of course the nature of scientific proofs is different in different disciplines so mathematical proofs just appeared in my analogy since the analogy was drawn from the area of mathematics.

    Ahh, I think I see a serious difficulty with your approach. You cannot imagine a way that string theory will be definitely proved to be wrong by physicists and therefore you assume that its weaknesses already suffice. (But just like the case of solving polynomial equations, in view of the prominence and successes of string theory much stronger scientific arguments will be needed.)

    Similarly, you cannot imagine how string theory can prevail as a viable physics theory and therefore you conclude that it failed. But string theory can win in ways we cannot imagine at present and can also fail in ways we cannot imagine at present.

  24. anon. says:

    ‘But string theory can win in ways we cannot imagine at present and can also fail in ways we cannot imagine at present.’

    This kind of statement is what you expect to find in a religious tract. It’s not really scientific, but is more in the category of belief systems/cults/wishful thinking/crackpotism. If string theory ‘wins’ or is falsified, then it will be deserving of a media mention. Until then, it’s not a scientific theory.

    Feynman told a story of ‘Cargo Cult Science’. During WWII, South Sea islands were used as military airports, and the islanders had business. After the war ended, some islanders tried to attract business by encouraging passing planes to come down to land by building improvised runways with control towers.

    Their efforts looked professional from a distance, but up close they were fake. It’s the same with string theory. The mathematics and alleged physical successes make it superficially look like a scientific discipline, but it isn’t because it can’t predict anything checkable.

  25. Zathras says:

    I’ve seen a number of prospective graduate students, who spent their undergraduate days as avid readers of Woit’s blog, and whose perspective on high energy physics is now so hopelessly divorced from reality that the best one can do is smile and nod one’s head pleasantly and say, “I hear the condensed matter group has openings.”

    That this blog has steered young students away from string theory is the best endorsement of any blog ever. Sounds like a clear charge of you’re corrupting the youth.

  26. Pingback: 42. Recalling a couple of facts about 2D and 3D Ising models

  27. Mitch Miller says:

    Peter Woit Wrote:
    “Right now, I think many of the leaders of the subject are well aware of how bad the problems are, but are holding out from giving up on the theory, waiting to see what the LHC says, hoping that LHC results will change the current picture. If the LHC doesn’t come up with something that somehow gives new hope to the string theory unification program, I think it will be conclusively dead, although some people will insist on pursuing it.”

    Wow, that is not the impression I get at all. Everybody seems to already know what the LHC will find (Higgs for sure, maybe Susy and small chance of extra dimensions and even smaller chance of something crazy and non standard model). i don’t see how anything bad for string theory can come out of the LHC. If extra dimensions are found, it will be great, if not its no big deal as the lower bound for these effects (or anything else related to string theory) to be measureable is not even close to being reached by the LHC. If something truely unexpected is found, everybody will be scrambling to explain it, not just string theorists. As a young grad student that is what I’m routing for, unlikely as it is 🙂

  28. Gil says:

    Dear Anon, please note that the statement you quote and criticize

    “String theory can win in ways we cannot imagine at present and can also fail in ways we cannot imagine at present.”

    is not a statement within a scientific theory, but a statement about scientific theories. As such, I think it is perfectly fine and can be supported by examples.

  29. anon. says:

    Gil, string ‘theory’ is a misnomer. Stringy ideas are an empty framework which fits 10^500 theories. To get a specific string theory, you need to specify all of the moduli of the Calabi-Yau manifold of 6 compactified dimensions. If there was a string theory then your position would be sensible (and this blog wouldn’t exist).

  30. Kris Krogh says:


    It’s true something unimaginable may save string theory. Would you like to invest in Company XYZ? The stock is performing terribly, but something unimaginable may save it.

  31. Gil Kalai says:

    If this looks more accurate to you, anon, replace “string theory’ by ‘the string framework’.

    It is possibe that the ‘string framework’ will lead to a definite win (and thus perhaps to a ‘string theory’) in ways we cannot expect, and it is also possible that it will be definitely shown, in unexpected manners, that the ‘string framework’ is inedequate for its (ultimate) purposes.

    There are also various possible forms of partial victory for ‘the string framework’, in terms of new insights and horizons in mathematics and physics which come short of a definite ‘theory’ to your taste. Some of these partial victories are already in place.

  32. Peter Woit says:

    The debate about string theory on this thread has become more or less completely content-free. Enough.

    My apologies for not providing fresh material for discussion. I’m busy with other things, and there has been remarkably little news on the math-physics front.

  33. nbutsomedy says:

    “I’ve seen a number of prospective graduate students, who spent their undergraduate days as avid readers of Woit’s blog, and whose perspective on high energy physics is now so hopelessly divorced from reality that the best one can do is smile and nod one’s head pleasantly and say, “I hear the condensed matter group has openings.”

    Thank God (anyway, not a believer) !!!! A wise step for the students. Both monetary and scientifically.

  34. TCO says:

    Why the disdain for condensed matter from the first poster at the other site? Look at what Feynman did with helium. What Einstein did with thermoelectric effect. BCS theory. Fermi-Dirac. Bose-Einstein. How about figuring out cuprate superconductors, studs? Come on smarties. Direct your theory to something physically tangible.

    -solid state chemist

  35. A deep statement by Atiyah, discussed by Gowers, recently in turn referenced by Terry Tao, explains why it might be a bad idea for String Theory (if “proven wrong”) reconsolidate itself as distinct departments away from Physics.

    Tao: “In fields such as nonlinear PDE (which Perelman’s result can broadly be included in), individual theorems tend not to be directly applicable much beyond their original intended use, but general ideas, strategies, tricks, and paradigms are often far broader. A similar point (using combinatorics as the primary example) is discussed in Gowers’ ‘two cultures’ paper


    Gowers citing Atiyah:

    “The ultimate justification for doing Mathematics is intimately related to its overall unity. If we grant that, on purely utilitarian grounds, mathematics justifies itself by some of its applications, then the whole of mathematics acquires a rationale provided it remains a connected whole. Any part that drifts away from the main body of the field has then to justify itself in a more direct fashion.”

    If String Theory has drifted away from the main body of Physics, or Mathematics, then (however beautiful it may be) it is sociologically subject to a demand for direct justification.

  36. S Halayka says:

    For what it’s worth, this paper seems to highlight the possibility of experimental verifiability at a scale of around 20 orders of magnitude larger than the Planck scale:


  37. S Halayka says:

    The paper mentioned above has been criticized to death by a trusted string researcher. My apologies for jumping the gun.

Comments are closed.