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

Is string theory really science?

  1. Feb 21, 2012 #1
    The scientific method says that if experiments or observations falsify a theory, that theory must be rejected.
    String theory requires space to have 9 dimensions. But observation shows that our space has only 3 dimensions. This alone should falsify string theory.

    I know that scientists have amended an ad-hoc hypothesis that 6 of the dimensions are curled up too tiny to observe. But there is no experimental evidence of this. And the theory does not predict or explain why 3 dimensions should be large and 6 tiny.

    If you add enough ad-hoc hypotheses, any theory could be accepted. The Earth-centered universe could still be believed if you add enough epicycles and other unsupported ideas to explain the motion of the planets. But science usually rejects this kind of fixing up of a theory (mainly due to Occam's Razor). So, without ad-hoc hypotheses, shouldn't string theory also be rejected?

    Does any future experiment even have the potential to falsify string theory? If so, what might that experiement be? If not, then string theory is not science. It is a religion.
     
  2. jcsd
  3. Feb 21, 2012 #2
    Please show what observation shows that space has 3 dimensions.

    Is there any experimental evidence against it?

    Please show your work. Where can I find a refernce to such a theory. Does it still predict the correct positions of celestial bodies, even the ones that were unknown in antiquity?

    Occam's Razor is not part of the scientific method.

    Maybe. I don't know that much string theory.
     
  4. Feb 21, 2012 #3
    In principle, yes, but it's observation instead of experiments [at least for now]. Stringy effects at high energy could have contributed in the very early universe, and with the expansion of the universe even tiny effects can be magnified. In the current non-stringy standard cosmology, you can use quantum field theory to study what is the effect of quantum fluctuation to matter perturbation, and this is basically the seeds that later form galaxies. In certain models of string theory, stringy effects will modify this, and cosmological observations can in principle falsify some of these models.There may also be imprints on the cosmic microwave background that differ from non-stringy models [http://arxiv.org/abs/0704.0647v1] [Broken]. Of course all these are works and ideas in progress, but in principle they can falsify or at least put some constraints on string theory.

    See also http://www.math.columbia.edu/~woit/testable.pdf.
     
    Last edited by a moderator: May 5, 2017
  5. Feb 22, 2012 #4

    haushofer

    User Avatar
    Science Advisor

    I think string theory should be regarded as a framework or paradigm, not as a single theory (alas the name!). In that sense it could be compared to quantum field theory, which is also a paradigm. A better name for String Theory would perhaps be "string paradigm".

    In that sense you can wonder what it means to "falsify string theory". You should be able to falsify specific theories in the string paradigm, but to falsify the paradigm itself would be very hard. How would one for instance "falsify the framework of quantum fields"?
     
    Last edited: Feb 22, 2012
  6. Feb 22, 2012 #5

    haushofer

    User Avatar
    Science Advisor

    For centuries, observation had shown that time is absolute, that nature is continuous, that the universe is static, that continents don't drift, etc. etc. Physics showed us how to go beyond intuition, and showed us that the contraries of all these statements are actually true.

    Why would space be 3-dimensional? Because we accidentally, as humans, live at energy scales at which it seems 3-dimensional?

    Don't get me wrong: I don't know how many dimensions space has. I only know that ST is a promising candidate for quantum gravity and a very interesting playground for physics beyond the standard model. ST is very conservative; you extend the notion of point particles and just apply all the lessons learned in the last century. This conservative route turns out to have "fancy" consequences.

    It's not ad-hoc. It is required by the theory, as we understand it.

    Could you show such a model which can describe all phenomena described by modern cosmology?

    See my post one up.
     
  7. Feb 23, 2012 #6
    First, to answer your question. The earth-centered universe is an idea proposed by Ptolemy that the planets revolve on small circles centered on other circles (epicycles) around earth. Later, it was proven that the paths of all planets and comets (any path at all) could be reconstructed from a large finite number of epicycles. This is similar to Fourier analysis recontructing any function from sin and cos waves. One reason science doesn't believe this theory anymore is that it is too complicated and has too many hypotheses. Newton's law of gravity and motion is much simpler.

    String theory also adds too many ad-hoc hypothesis. Strings require 9 space dimesions but we see only 3, so scientists propose that 6 are curled up tiny. This may be true, but the theory doesn't require this curling up and there is no evidence for it. Now some people have proposed that all 9 dimensions are large but we are stuck to a 3-D membrane. So which is it? String theory doesn't say. Strings predict many new particles that haven't been found, so scientists propose that they must have huge masses. This may be true, but the theory doesn't require large masses and there is no experimental evidence for them. There seems to be too many "unsupported speculations" being made just to try to save the theory.

    I agree with Haushofer in that string theory/M theory is a paradigm, or philosophy. It is not a specific fundamental theory. We can not prove or falsify a paradigm. Physicists must develop one definite specific string theory that can be tested by experiments/observations.
    I also agree with Yenchin that observations of cosmic background radiation might test string theory. But, we still need some definite prediction from the theory to test it.

    I know that many wierd ideas in science have turned out to be true. But these have been confirmed by experiments/observations before scientists accepted them. String theorists seem to be saying "just believe all these wierd things because we say so". That is not the method of science; that is the method of religion.
     
  8. Feb 23, 2012 #7

    Nabeshin

    User Avatar
    Science Advisor

    String theory is indeed a framework, but I think it's still subject to falsification. In principle at least, one could derive some sort of general result from the framework of ST which simply is inconsistent with our known universe, falsifying the whole thing. Much more likely, however, is that we rule out large swaths of string theories from observation.

    Also, the fact that spacetime is 11 dimensional is not in the slightest a hypothesis, but rather a PREDICTION from the theory itself. This is rather startling, because I know of no other theory which actually predicts the dimensionality of spacetime (correctly or otherwise). It would certainly be fantastic if we could show that, in general, things compactify to a universe resembling ours, but this is of course a tall order. I think the analogy to QFT is really good here -- we can imagine tons of QFTs, like people are doing right now with compactifications of string theory, and simply hope to find ones that are either 1) interesting, 2) similar to our universe, or 3) falsifiable.

    I'm really surprised that you don't like string theory and find it arbitrary, given that the Standard Model is perhaps one of the most ad-hoc theories I can think of. Sure, it has great predictive power, but we had to put a lot into it (masses and particle content!) to get this power out. String theory, on the other hand, requires one (dimensionful!) parameter: the string length. Now, it would certainly be nice if the theory could, from that, predict the dynamics of its dimensions to reveal 3 macroscopic dimensions like we see, and perhaps it can, but the jury's still out.

    Main motivations for scientists believing so strongly in the validity of string theory despite its lack of experimental validation are obviously deeply theoretical. But such concerns led to Einstein developing GR, it's just that with that case, predictions were much easier to make and falsify. I think it's way too soon to get on either side of the bandwagon though, and I very much dislike it when popularizers of science like Greene make it sound as if we KNOW ST is correct. I don't think anyone is saying 'just believe this because we say so', so much as, 'look at all these amazing things within string theory, what are the odds that this theory is wrong and still contains all of this?'
     
  9. Feb 23, 2012 #8

    tom.stoer

    User Avatar
    Science Advisor

    Let me add one remark regarding the relation of QFTs like the SM and string theory.

    QFTs, mostly (partially supersymmetric) gauge theories can be constructed for different interactions but always w/o gravity. The construction is well-understood, but the question which specific symmetry structure you want to use is ad hoc. I would say that it's common believe that interactions w/o gravity have to be described using gauge theories.

    Taking gravity into account the situation becomes much more complicated. It could be that there is a similar construction principle call supergravity (SUGRA) which is much more restrictive regarding matter content, interactions and dimensionality of spacetime.

    The big difference between gauge theories and SUGRA is renormalizibility. For gauge theories we know (infinitly) many cancidates which are perturbatively renormalizable. For SUGRA it's not yet clear, but perhaps there is not even one!

    No look at strings: what the string theory framework essentially does is to convert a large class of supergravity-gauge-theories which have been constructed ad hioc into vacuum solutions of one underlying theory
     
  10. Feb 23, 2012 #9

    tom.stoer

    User Avatar
    Science Advisor

    Let me add one remark regarding the relation of QFTs like the SM and string theory.

    QFTs, mostly (partially supersymmetric) gauge theories can be constructed for different interactions but always w/o gravity. The construction is well-understood, but the question which specific symmetry structure you want to use is ad hoc. I would say that it's common believe that interactions w/o gravity have to be described using gauge theories.

    Taking gravity into account the situation becomes much more complicated. It could be that there is a similar construction principle call supergravity (SUGRA) which is much more restrictive regarding matter content, interactions and dimensionality of spacetime.

    The big difference between gauge theories and SUGRA is renormalizibility. For gauge theories we know (infinitly) many cancidates which are perturbatively renormalizable. For SUGRA it's not yet clear, but perhaps there is not even one!

    No look at strings: what the string theory framework essentially does is to convert a large class of supergravity-gauge-theories which have been constructed ad hioc into vacuum solutions of one underlying theory
     
  11. Feb 23, 2012 #10

    tom.stoer

    User Avatar
    Science Advisor

    Let me add one remark regarding the relation of QFTs like the SM and string theory.

    1) QFTs, mostly (partially supersymmetric) gauge theories can be constructed for different interactions but always w/o gravity. The construction is well-understood, but the question which specific symmetry structure you want to use is ad hoc.

    I would say that it's common believe that interactions w/o gravity have to be described using gauge theories.

    2) Taking gravity into account the situation becomes much more complicated. It could be that there is a similar construction principle call supergravity (SUGRA) which is much more restrictive regarding matter content, interactions and dimensionality of spacetime.

    The big difference between gauge theories and SUGRA is renormalizibility. For gauge theories we know (infinitly) many cancidates which are perturbatively renormalizable. For SUGRA it's not yet clear, but perhaps there is not even one!

    I would say that it's widely believed that the low-energy sector of interactions w/ gravity taken into account have to be described using SUGRA-gauge theories.

    3) Now look at strings: what the string theory framework essentially does is to convert a large class of supergravity-gauge-theories (which have been constructed ad hoc) into vacuum solutions of one underlying theory!!! In addition another large class of theories (infinitly many) which could be constructed ad hoc is ruled out b/c they cannot be derived from string theory. I bet that you cannot derive a world with gravity but w/o other fundamental forces; neither can you derive a theory like the SM in flat Minkowski spacetime w/o gravity. Therfeore string theory turns a construction principle into solutions.

    It's like having infinitly many low-energy theories describing water, iron, ice, wood, ... and then you discover QED. QED does not predict whether we observe water or ice, but it provides a unique framework from which all these different states of matter van be derived in principle.

    I would say that many string theorists believe that this is one cenmtral feature of the string paradigm. Somehow the theory looses predictive power b/c it cannot tell you which low-energy theory you will observe, but it does explain which theories can exist in principle.
     
  12. Feb 23, 2012 #11
    I think the word 'theory' is a misnomer. It suggest a scientific grounding in solid evidence, which string theory lacks. Absolutely, there are many good reasons to assume that string theory will ultimately prove to be correct, but as of yet, there is no such experimental proof. There is only 'well, everything seems to fit if we use it'. That's not a bad thing, but, like I said, I would rather we not use the word theory to describe it.

    (I mean, can we really blame the creationists for saying evolution is 'just a theory' when we use the word theory for pretty much everything ourselves?)
     
  13. Feb 23, 2012 #12
    I rather share kochanskij's point of view. strings are not what we'd expect from a ToE. it needs too many hypothesys and makes no enough testable predictions.

    Hypothesys:
    - 11 dim
    - calabi-yau (why calabi-yau and not something else?)
    - susy (+ hundreds of free paramaters)
    - particules are strings
    - 100^500 theories

    All of this do not even make possible to derive the standart model in its full details (+ possible extenions). All we have are models looking more of less like the SM.

    In the contrary, a good theory should decreased the free parameters (if you believe in science and not in landscape)
    Maybe one day a principle will be found to make strings able to single out a testable theory, I hope so. But since 20 years is has not be the case.

    SM was constructed based on experience. Free paramters need to be fixed experimentally but it made thousands of testable predictions. It is a terrificly efficient theory. That what we ask to Strings, predictions must be at least as great as hypothesys are.
     
  14. Feb 23, 2012 #13

    haushofer

    User Avatar
    Science Advisor

    But how much of a breakthrough would it be if it would be shown that ST could "reproduce the standard model"? ST lies way, way, way beyond the energy scales of the standard model. What would we really learn if we could explicitly derive the SM from ST, besides that it is then explicitly shown to be possible?
     
  15. Feb 23, 2012 #14
    Deriving SM would be a breakthrough, but it is the strict minimum we can ask to a ToE.
    If a theory aiming to be a ToE is not able to explain what we already know to be true, then the theory is manifestly wrong or at least incomplete. I would say ST is still incomplete in that sense.

    To come back to the question of this thread: Does it mean ST is not science ?
    I don't think so, IMO it is an incomplete and untestable theory for the moment. Maybe one day it will proved to contain the SM and we could call it a "possible true" theory, but it is not case...

    I find much more promising and efficient for instance NCG which lives at planck scale too but does derive the full SM + Gravity and makes predictions + postdictions (mass of the top...)
    Predicted Higgs mass was already ruled-out, but it might be fixed if new particules are discovered at the LHC. No Landscape problem, not 10 dimensions with Calabi-Yau... and the SM + Gravity fully derived.
     
  16. Feb 23, 2012 #15

    tom.stoer

    User Avatar
    Science Advisor

    Can anybody derive solid state physics from QCD + QED?
     
  17. Feb 23, 2012 #16

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    Isn't that one of the odd numbered exercises in Physics 253?
     
  18. Feb 23, 2012 #17
    I would have to say that, at this point, calling it a theory is a misnomer. Does this make a huge difference to anything? No.
     
  19. Feb 24, 2012 #18
    I think string theory is more of a mathimatical trick than science. The goal being to uncover underlying possiblities in physics to give a clue or solution to unexplained phenomena in physics. In doing this, it has failed. All it has shown us is that physics can be explained in about any number of dimensions and could be equally valid. I think this is a result of mathmatics itself and not physics. All we have found is that physics can work the same way in about any number of dimensions mathmatically. It is too complex to be logically changed in a logical way that would in turn describe our world more accurately, so then there would never be any reason to assume that reality has to be made up of more or less dimensions. Even if it was, it could still probably be described just as accurately in four dimensions because the mathimatics could allow that to happen. So, then there would never be a reason to think the universe should have a certain amount of dimensions, since mathmatics allows any number of dimensions to be descibed with a different number of dimensions equally.
     
  20. Feb 24, 2012 #19

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    This is just false. Various numbers of dimension are not 'put into string theory'. Instead, the number of dimensions comes out as prediction of the string framework, and there is only one choice for the most general framework: 11. In principle, this is a falsifiable prediction. Unfortunately, for now, it is not a testable prediction (because of campactification).
     
  21. Feb 24, 2012 #20
    For the n-th time: extra dimensions should not be taken too literally. Rather they represent one way to parametrize "internal" degrees of freedom; or "matter fields". Generically they even dont have in interpretation in terms of higher dimensions; only in limiting cases this is so.

    So what the necessity of those "extra dimensions" translates to, is that string theory _requires_ extra matter fields besides pure gravity. This is a feature rather than a drawback! AFAIK no other approach to gravity seems to be able to come up with such a constraint, which then means no predictions at all for particle physics.

    Whoever is obsessed by not liking "extra dimensions", is invited to think about them simply as a natural device to generate quarks, leptons, higgs fields.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook