Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

String Theory

  1. May 25, 2010 #1
    I was just wondering, I keep hearing different opinions on this, but how is the reputation of string theory now?

    I speak to some mathematicians and they completely write it off, while I talk to other people and they say it's still in the works but looks promising.

    Obviously it's not finished nor a complete TOE(at the moment) but is it still a route that should be explored?
  2. jcsd
  3. May 26, 2010 #2


    User Avatar
    Science Advisor

    Yes, you correctly summarized the status of it.
  4. May 26, 2010 #3

    so.. is it still a valid theory that is actually thought to at least be possibly correct?

    or is it just the red headed step child of the physics world?
  5. May 26, 2010 #4


    User Avatar
    Science Advisor

    I doubt that you will get an unbiased summary here.

    I would say that string theory is still an active research program, but that it lost credit due to
    - missing post-dictions of already known phenomena
    - missing falsifiable predictions of new phenomena
    - excuses like the antropic principle as a way to save the theory

    Even if you forget about the antropic reasoning and the landscape (which is not seen as a viable way out by all string researchers) the first two problems remain.

    I would say that many discussions are politically biased. This is due to the fact that the majority of the string community ignored the first two problems over a couple of years more or less completely. A few years ago there were (not only physically motivated) attacks on string theory focussing exactly on these weak points.

    So the problem is that the debate about string theory in the last years was never free of prejudices, from neither side.
  6. Jun 13, 2010 #5


    User Avatar

    I would have to respectfully disagree about "missing falsifiable predictions of new phenomena". There are plenty of predictions, and they are theoretically falsifiable. At this point in time experimental physics has yet to develop a machine which would test these predictions, but that is not a flaw in the theory, that is a flaw in the experimentation.
    There is a difference between a theory which is unfalsifiable in principle (e.g. "God exists") or a theory which is unfalsifiable in practice (e.g. "there is water on Pluto"). The former is philosophy while the latter is science. I think this is the crux of the problem - many critics of string theory make no distinction between a theory for which a test can be designed, even if not implemented, and a theory for which no test can be designed in the first place. There is nothing wrong with a theory which is ahead of our technological capacity. 300 years ago they might have dismissed quantum mechanics as "missing falsifiable predictions of new phenomena". It is nonetheless a resoundingly successful scientific theory (although, of course, not perfect).
    And as for missing "post-dictions of already known phenomena" - I am not sure specifically what you are referring to, but string theory, like loop gravity, technicolor, etc. is not a complete theory. That is precisely why the string theorists should keep working at it.
  7. Jun 13, 2010 #6
    It seems the early promise of greatness for String Theory has not yet been fulfilled.... maybe sort of like the human genome code definition....at ten years for that this month, I don't think a single medical cure has been discovered....nor the specific genetics for any disease. But progress sometimes is elusive until a breakthruogh....like Einsteins "greatest thought"....

    In any case, a number of theoreticians seem to feel the mathematics that has been developed in string theory and even possible physical insights that are suggested is itself worthwhile and may ultimately lead to other approaches. Maybe that's akin the Riemannian geometry which doesn't appear to have been used much until Einstein came along some years later and used it to great effect in General Relativity.

    It seems one great obstacle was initially different and complementary views... the existence of at least five approaches.....those were at least partially reconciled by Ed Witten via "M Theory" but that, too, seems to have not proved to be the cure all as was hoped. So far as I understand it, the mathematics of string theory is so complicated no one knows the exact equations of the theory nor I believe exact solutions.

    "Valid", I am unsure; "possibly correct", I think so...lots of people still work its mysteries.

    Science is so far a progression over time: From Newtonian space and time, then Einstein with continuous GR, then QM with "quantum" components....none seem to be the final answer either.
  8. Jun 13, 2010 #7


    User Avatar
    Science Advisor

    I agree and I see the difference.

    But the problem is that ST is not even theoretically falsifiable. E.g. ST predicts SUSY, but does not tell us at which energy we should expect SUSY. The energy can be arbitrary high. So can you tell me which unique experimentally fasifiable predictions ST makes?

    By unique I mean
    a) predictions for which we necessarily need ST, that means for which e.g. MSSM and/or SUGRA are not enough
    b) predictions which are unique within ST, that means for which ST does not make different predictions depending (e.g.) on different approximations.

    Some examples:
    • we live in 3 large spatial dimensions; ST can't tell us why 3
    • the SM has 3 generations of fermions; ST can't tell us why 3
    • the SM has chiral fermions; ST can't tell us why
    • the SM has U(1)*SU(2)*SU(3) as its gauge group; ST can't tell us why
    • ST can't explain the masses = the Yukawa couplings + Higgs mass, Weinberg angle, CKM etc.

    Can you tell me which structures existing in the SM are explained by ST?

    As I said, my arguments are not unbiased. I don't want to tell anybody to stop or to continue to work on certain theories. But I think that a theory which is investigated by thousands of brilliant physicists for decades w/o progress of practical relevance can't be true. With practical I mean that ST must not only solve problems created by ST itself, but problems and questions relevant in practice (like in my list mentioned above).
  9. Jun 13, 2010 #8
    this is basically what I thought..

    I was wondering if anyone found problems with the theory. One can't dismiss a theory simply because our technology prevents us from being able to experiment it.

    But I guess the question is, would it ever be able to be proven? It's an obvious consensus that no time even remotely soon would we be able to experimentally observe said 'strings' so, the theory could have had all kinks worked out, but it still would be up in the air.

    When we're talking on the level of strings, I don't think anyone even has a far off theoretical idea of ho we would even try to observe them ( I could be wrong)
  10. Jun 13, 2010 #9


    User Avatar

    String theory is not my forte, but as an example of a possibly observable phenomenon predicted exclusively by string theory, consider the predicted B-field corrections to the Maxwell field. If we were able to observe the excitations of an electromagnetic field to sufficient accuracy to determine whether the field is purely Maxwellian or has a B-field correction, that would be clear evidence either in favor of, or refuting string theory.
    I may be wrong, but I was under the impression that the gauge group structure of the standard model is expected to arise from an appropriate choice of Calabi-Yau compactification, at least in type IIA or IIB superstring theory. I also thought (and again, I could be mistaken) that the fermion generations fall out of the mass squared spectrum of different superstring sectors.
    Finally, I am not sure what you mean when you that string theory can't tell us why our observable spatial world has 3 dimensions. On a physical level, we can't observe the other 6 dimensions for many possible reasons, all of which depend on the geometry of those dimensions. If you mean philosophically why our universe emerged in this way from the big bang, I don't believe that string theory has an answer, so touche on that point.
  11. Jun 13, 2010 #10


    User Avatar

    It is also my understanding that certain types of Regge trajectories would be indicative of strings.
  12. Jun 14, 2010 #11


    User Avatar
    Science Advisor

    With appropriate CY acrobatics you can generate (nearly) any gouge group you like; there is no selection principle which tells you which CY to chose, so no prediction for a specific gauge group. The mathematics changes with M-theory, but the problem remains essentially the same.

    The fermion generations are related to the topology of the CY, so you have the same problem that you can't predict them. An additional problem is mass generation: in low-energy effective theories it boils down to SUSY breaking. Afaik how to break it is rather arbitrary (look at the MSSM), so again no prediction of fermion masses.

    Yes, this is what I mean. ST does not provide any argment why 3 out of 9 spatial dimensions are large whereas the other 6 are compactified to CY. The mathematics changes with M-theory, but the problem remains essentially the same.

    That is correct, scattering amplitudes would indicate stringy behaviour. Regge behaviour is approx. true for certain hadrons which (afaik) was a support for ST in the sixties and early seventies. But now we would expect Regge trajectoires for elementary particles; unfortunately the situation is similar to SUSY: ST does not tell us at which energies we should expect Regge behaviour.
  13. Jun 14, 2010 #12
    If the higgs boson is found. Does that mean string theory is done?
  14. Jun 14, 2010 #13


    User Avatar
    Science Advisor

    It does not mean anything regarding ST.
  15. Jun 14, 2010 #14


    User Avatar

    Well, that still leaves us with B-field corrections as a falsifiable prediction of string theory. As for post-dictions, I think I've lost the argument there... :cry:
  16. Jun 14, 2010 #15


    User Avatar
    Science Advisor

    Can you elaborate on this? Does that mean you expect ST corrections to Maxwell equations? Already at the classical level?
  17. Jun 14, 2010 #16
    Hmmm I have many things to say about this.

    But I want to know what you mean by ``missing post-dictions of already known phenomena''.

    What phenomena can string theory NOT explain?
  18. Jun 14, 2010 #17
    Someone please clarify or correct me here if I am wrong but in String Theory... - ( I believe I read this from Brian Green's first book ) - ...the "long hand" equations of ST are so time consuming for a PC to work through that they must use Perturbation theory to gain approximations.

    Assuming the above is true. By using Perturbation theory are they clouding the results making predictions and relationships inaccurate?

    Could anyone (more knowledgeable than me:smile:) elaborate on this or point out anything wrong in my understanding?
  19. Jun 14, 2010 #18


    User Avatar
    Science Advisor

    As there is no derivation of the standard model from ST, ST does not make the predictions the standard model makes; therefore ST does not even predict existence of electrons, quarks, neutrinos, ... ST does not predict their masses, their coupling strength ... ST does not predict confinement, deep inelastic scattering, chiral symmetry breaking, SU(2)*U(1) symmetry breaking, neutrino oscillations etc.

    ST says tat all this could happen, but it does not exclude other possibilities. The standard model may emerge from ST; this has neither been proven nor disproven.

    Thats what I mean by "missing post-dictions".
  20. Jun 14, 2010 #19
    What do you mean by ``standard model''?

    Do you mean matter content, particle interactions and forces?

    Or do you mean the electron mass to arbitrarily many decimal places?

    My other comment is---SHOULD string theory predict things like chiral symmetry breaking, coupling strengths, or confinement? These are all features of the low energy effective field theory, and I think if you're asking string theory to address these questions, you're asking the wrong questions. Right?
  21. Jun 14, 2010 #20

    George Jones

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Full disclosure: I have almost no quantitative knowledge of string theory; after working through the first six chapters of Zwiebach and doing the exercises and problems, life got in the way.

    From the second edition (2009) of Zwiebach (page 481):
    A fair assessment?
  22. Jun 14, 2010 #21
    Of the models that he is discussing, yes, but not of the state of the art of string model building.

    There's more than one way to skin a cat, you know.
  23. Jun 14, 2010 #22


    User Avatar

    What would you say are the right questions?

    Would it make sense to you to say that the questions ST is supposed to answer (ie. the "right questions") is to describe the possible (consistent with ST-ideas) world of models in the high energy domain, and not which of these possibilities that describe reality in the low-energy limit, and that the latter would be determined by experiment, and that the difficulty of the large landscape is due to actual uncertainties, rather than a flawed inference? And that we are bound to keep digging our way manually in the landscape?

    If you disagree how would you phrase briefly the right questions?

  24. Jun 14, 2010 #23


    User Avatar
    Science Advisor

    By standard model I mean the well-established standard model of elementary particle physics: U(1)*SU(2) electro-weak interaction plus SU(3) QCD. Of course this includes the symmetry structure, 6 flavor / 3 generation content, masses, couplings, mixing angles etc. Of course I do not expect all these constants to be derivable with arbitrary numerical accurancy, but I expect at least the overall picture to emerge from ST.

    Regarding low-energy phenomena I am not asking the wrong question. I do not expect ST to predict all these phenoma directly. The standard model as a low-energy effective theory of ST would be fine. But unfortunately there is no way known how to derive the above mentioned structure (the structure could be totally different as well).

    Compare it to QCD: there are a couple of ways to motivate low-energy effective theories based on QCD (chiral perturbation theory). Of course this is not perfect, but by using symmetry arguments, approximations etc. one can at least motivate low-energy theories fitting both to the symmetry structure of QCD and to the low-energy phenomenology. Look at lattice gauge theory: you just plug in a few constants (like nucleon masses) and you are able to calculate all the other masses, decay constants and things like that within a few percent. Look at the Fermi theory of weak interactions, it emerges as "static limit" of the full GSW theory. The Fermi constant is even used to fix parameters of the GSW.

    No such interplay between ST and the SM is known. Neither does ST predict the low-energy symmetry structure or some unkown parameters (like masses, couplings), not does the low energy effective theory allow one to predict or constrain the theory space of ST.

    I know of no known fact of the standard model which was derived from ST. If you know a counter example, please let me know.

    What ST is able to do is to provide a large class of theories (regarding dimensions, gauge groups, symmetries and symmetry breaking), to somehow constrain the theory space and to harmonize these theories with quantum gravity. This is of course a major achievement. D-branes and F-theory provide mechanisms for model building and they come quite close to structures similar to the standard model. So ST certainly makes some predictions like
    - there must be gauge theories at low energies
    - there must be SUSY which can be broken (fully or partially) at low energy theories
    - there can be compactification of spatial dimensions
    - fermion generations emerge from the topology of CY spaces
    - ...
    But all these structures are available w/o strings as well. So I do not see the benefit of introducing the whole complexity of strings just to motivate something which we already know and which can exist w/o strings as well. I do not see the additional benefit coming from strings.
    Last edited: Jun 14, 2010
  25. Jun 14, 2010 #24


    User Avatar
    Science Advisor

    "What would you say are the right questions?"

    No one knows.

    What is somewhat unreasonable is to expect a theory of quantum gravity to control anything about everything regarding low energy physics.

    For instance when we discuss superfluidity, we don't sit there and write up a many body QCD and QED bound state problem. We could in principle, but the main salient features are mostly emergent (emergent here is a buzzword for 'its too complicated to solve, i'll just model the feature semianalytically')

    Now the *hope* is that it does control a few things. Like for instance the amount of generations in the standard model, or perhaps why such and such a group representation is small.

    To first approximation, if I was a physicist working on string theory phenomenology in the 80s, I would have been very satisfied if I could simply get mSugra and maybe a nice GUT group out. From there, various lower energy dynamics and unknown symmetry breakings mechanisms could in principle conspire to give me the standard model (+ answer a bunch of pertinent bySM questions). In that regard i'd be way ahead of the field already. Well, that has been done, redone, and finished a long time ago.

    The state of the art in stringy phenomenology can now (somewhat miraculously) calculate various mass hierarchies and things like that.

    The problem of course is that usually the more accuracy you push a particular class of stringy solutions, the more you find that things aren't that easy (and thats good in a sense). Sometimes you get exotics, sometimes one number comes up wrong, etc etc,

    So lets say you have a nice stringy model A and you almost get the standard model in some approximation (maybe say the Yukawa couplings come out a little off, say maybe an order of magnitude error). The question you ask yourself, is .. Is the vacua i'm using incorrect, am I making a bad approximation somewhere, or am I failing to take into account some emergent low energy phenomena? That's a hard question to disambiguate.

    Worse. Assume you finally did write down a model that reproduced the standard model identically. You could ask the same question (is this just a coincidence and did I really take into account everything).
  26. Jun 14, 2010 #25
    See this paper: http://arxiv.org/abs/0708.2691

    These models have the correct MSSM spectrum including forces, particle content (three generations of quarks and leptons) and interactions, heavy top quarks, and it can be shown that there are no light exotics. All couplings are calculable in principle, as the full yukawa structure is known.

    Well, for one, strings gives you a consistent theory of quantum gravity. To understand how the standard model emerges from gravity is no small feat.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook