I think it's time to write a short essay why I am really disappointed about string theory. It is not because ST is wrong (we don't know if it is wrong; and we have no proof that other theories such as the SM are right). It is not because it is complicated (the electroweak model is complicated as well), and it is not because I do not understand it (there are not many people who should claim that a theory is wrong simply b/c they do not understand) It has something to do with its potential and with keeping (or neglecting) promises. As far as I can see string theory (whatever this means - ST, F-, M-, ...) is the only candidate with the potential to unify all interactions including gravity. Other approaches may seem more straightforward, less complicated or exotic, closer to the well-known realm of the SM, ... but none is able to unify all interactions. Loop Quantum Gravity (which I really like and which I understand quite well) is a very promising candidate for QG - but as far as I can see only for QG. Topological braiding of q-deformed spin-networks from which all elementary particles can emerge is a brilliant idea - but up to know wishful thinking. Non-commutative geometry does not expalin why there is a unique nc structure (or I have overlooked this as I am not an expert in this subject) Asymptotic safety is fine for QG, but it does not explain the symmetry structure, masses, coupling constants, mixing angles etc. of the SM. SUGRA is neither unique nor finite (perhaps finite order by order, but not for the whole series) - nor elegant if restricted to four dimensions What I like about string theory. I think the greatest achievement of string theory (provided that it's true; afaik we do not have a sound proof, whether it's true) is that string theory turns most (all?) possible theories including gravity from theories into solutions. If the ideas regarding SUGRA, swampland etc. are correct, then there is a huge class of theories (in the classical sense) that can be "derived" from string theory. And perhaps even the opposite is true: theories which cannot be "derived" from string theory will turn out to be inconsistent. From that point of view I should be a fan of string theory, shouldn't I? So here's what I don't like about string theory. I don't like that string theory comes with an enormous mathematical and physical apparatus, w/o being able to give us a hint why we should believe in this apparatus (10/11 dim., SUSY, CY, ...). For me there is only one good reason to become a believer, namely to follow my argument from above - that string theory defines a unique framework from which all candidate theories can emerge. So it's about a promises! So essentially I like string theory b/c it makes these promises - and I don't like string theory b/c it only makes these promises! Here are some questions for which I would like to see progress [I only list problems that are not inherent to string theory; I don't care how to count CY spaces b/c this is not a physical question; I don't care about the definition of the higher genus measure for the amplitudes in superspace as this is not relevant physically - it is a problem not created by nature but by the string theory formalism; I don't care about AdS/CFT b/c this is not our universe; so I will list questions asked by nature] Why do we live in a 3+1 dimensional space-time? How is the big bang singularity resolved? Why do we see the matter content and interactions we see? (why three fermion generations, why the symmetry group of the standard model, why the Higgs (or not?), ...? do we live in a 3+1 dimensional space-time? What is the mechanism breaking symmetries and selecting the true vacuum? (which according to string theory defines the above mentioned interactions and structures) What is the microscopically picture regarding dynamical spacetime including black holes, entropy of the gravitational field etc.? Of course I am prepared for answers regarding landscapes (Susskind) and mathematical universes (Tegmark). But frankly: I will never accept these arguments. This is regarding string theory, therefore I expect answers in the context of string theory (if my daughter has to go upstairs in order to shampoo I don't accept discussions regarding justice; that does not mean that I am unintersted in justice - I am - but not in the context of telling a six-year old girls to go upstairs in order to shampoo!) So my expectation is that string theory does something very natural: be aware of the true problems of nature, provide ideas how to address them, provide a status or summary regarding progress and obstacles. Instead of listing obstacles (which may sound biased or even rude) I would like to ask the string theory audience here in this forum for their assessment. What are the major achievements of string theory? Are there predictions subject to (accessable to) experimental verification / falsification both in principle and in practice? Are there physical phenoma which (once observed) would kill string theory? Are there predictions specific for the string theory context (nothing that may follow from SUSY as SUSY could be true even w/o string theory) What are the short-term / long-term research programs? What are the major obstacles inherent to string theory preventing the theory from delivering on its promises? What will be the final theory in terms of strings - a theory, or a framework to create theories? A last remark for all those who are still with me and did not stop reading: this is all about progress in physics and waiting for illumination. It is not about fighting against a theory - doing that would require less thinking and writing ... Regards & Thanks Tom
The big problem with answering these deep questions is that we don't yet have a working definition for string theory. What we have so far are only bits and pieces, HINTS that something must be going on. Just because ads/cft makes some correct calculations, doesn't mean that ads/cft is THE string theory. Similarly just because some CY compactifications lead to SM (or MSSM) type physics, doesn't mean that must be the way to do it. There are also of course the twistor strings which are a whole different beast. All of these different frameworks have to come together somehow, they can't all be describing different (but sometimes even the same) parts of reality without being connected. That's what we need to understand before even attempting to answer physical questions. As for some major achievements, how about the KLT relations? If strings "don't exist", why did that work? Think about what it means for something to "exist" anyway. How do we know particles "exist"? We don't, in fact we have no idea what these particles "really" are. Maybe these particles are actually tiny vibrating apples made of energy - we don't care. We just come up with a model that attempts to make correct calculations, up to a particular accuracy. If the model works, its constituents are said to "exist". This is what we're doing with string theory in ads/cft. If those calculations lead to something correct, there you have it, strings "exist". Whether or not you'll see them in an experiment, and whether or not our universe is AdS.
The attitude of string theorists towards twistor strings is either as a superficial mathematical, not good beyond 1 loop calculations or just as a flawed physical theory. This is not an opinion of mine, I actually asked important string theorists about this. And that goes along what Witten said about it being incomplete. As KLT relations, they might be or not imply the need for strings. Apparently, it works for SUGRA N8 in relation to SYM N4, in 4d, up to 4 loops, beyond that no one is mathematically sure because there is no proof of renormalization. Some say that counter terms of 7 loops or not may rely on terms that are not due supersymmetry.
It looks like you're trying to bend over backwards to deny the use of strings. Next thing you know, even if string theory makes a prediction about new particles or something which are later found, you'll say ah that doesn't mean strings exists, we can have those particles by just putting them in the SM by hand. The KLT relations, at tree level (which doesn't care much about susy), predicted a relationship between gravitons and gluons. There's no field theory understanding (as far as I know) of why this should be true. So you use string theory to derive a non-trivial relationship, and then you say we don't actually need strings for that to work? So let me get this straight: -if you use string theory to explain stuff about known theories, that doesn't mean strings exists - since the theory as well as the particular fact can exist independently from strings? -if you use string theory to describe possible new things - that's silly because we haven't seen any of the new things? As for twistor strings - yes it doesn't work beyond loop level yet. But at tree level it works pretty nicely. Is that a coincidence, a fluke, or what? Just this particular fluke has been more useful than everything eg LQG has done so far, so I don't see why we should disregard it so quickly.
I appreciate very much what you say; this is my impression as well, but I would like to learn something like that from string theorists. If thus is true than it's still too early to ask physical questions. So perhaps string theory is not wrong, but just too difficult to understand. The problem is again about promises. If you propose such complex theory, the results must WOW you in order to justify this complexity. Do you really need strings; I thought SUGRA was enough. I agree. I don't agree.You can't start with a wrong AdS universe, do some calculations and hope that this proves anything. If you start from "the moon is made of green cheese" and if you are able to derive the standard model, your proposition remains wrong! I agree that AdS/CFT is a major achievement, but more a mathematical than a physical one.
No one is saying that AdS/CFT means that the universe is AdS. It's just a toy model. There are other variants of gauge/string dualities, ads/cft is just one example. You can think of it as just a framework used for solving SYM. Another example would be building a sphere to solve a matrix problem. It doesn't mean that our universe has anything to do with either the sphere or the matrix. You can still use this construction though for practical reasons. All this construction says is that there is probably a deep connection between spheres and matrices, or something like that. Just like apparently there is a deep connection between particles and gauge groups.
These authors speculate on getting de Sitter solutions from AdS/CFT - is there a handwavy way that a layman could understand this? http://arxiv.org/abs/0908.0756 "We then generalize these to AdS4/CFT3 duals, and suggest extensions of the method to obtain de Sitter solutions."
You are making this up. But you put up pretty much things very well straight. That's the usual scientific skepticism. Yes. But you do know that just like contour integrals exist in real life twisotrs are held as a mathematical device (which also works by simplifying contour integrals...). Twistor strings are not strings, at best U(1) instatons according to Berkovits. By Witten's twistor string used to build it are not really strings, but topological strings. You have to defend String Theory by means of the F-Theory derived GUTs.
F theory is just one part of string theory. I don't see any reason why you would consider it more string theory than the other parts. Anything that resembles strings and has applications (whether it's gluon amplitudes, quantum gravity, unification, or condensed matter) is good string theory to me.
Because that`s the one that actually is making advances where it was supposed to be making, including the mass relations between generations, quantum corrections that are not supposed to be not existent without the theory, that is, among all of the other uses, things that leads falsifiable predictions.
Well, no one really cares what it was supposed to be doing. It was actually only supposed to explain mesons.. We don't know enough string theory to predict masses and stuff like that, that's the short story. And that's why everyone is trying to understand it from different perspectives. I would find it very surprising if some sort of holography wasn't needed in the full theory which gives the SM. I think there's deeper reason why we have both ads/cft and f theory type of formulations. The chances of both of them existing independently are very slim, I'd say. F theory and ads/cft are just the tip of the iceberg. By the way, I forgot to remind you about the dual conformal invariance of YM amplitudes which lead to the discovery of the T fermionic duality for strings. Like I said before, since gauge theories and string theory are the same thing, whatever you find in one you'll find in the other. So the whole discussion of who is working on exactly what is pretty silly to me. It seems to me like most anti-string people care more about who is working on string theory than the string theorists. Also ads/cft can lead to falsifiable predictions, about SYM (or maybe quark gluon plasma). Why doesn't that count?
I don't speak with any authority on this, not being a string theorist, so take the following for what its worth. "Why do we live in a 3+1 dimensional space-time? " Unknown, but a good attempt at a guess might be string gas cosmology. "How is the big bang singularity resolved?" I don't know. Some more benign singularities seem to be resolved or smoothed out by perturbative string theory, but others (like apparently the bb singularity) really require the full nonperturbative analysis (see AdS/CFT or Matrix theory) to even begin to talk about. "Why do we see the matter content and interactions we see? (why three fermion generations, why the symmetry group of the standard model, why the Higgs (or not?), ...?" Certain promising vacua give very specific answers to those questions. Typically in the form of a statement about the geometry of the underlying space. Of course you can always ask, well why that particular geometry and you go right back to the vacuum selection problem which as you know is currently unsolved. Nevertheless, that any vacua possess this sort of capability and there are by now hundreds of papers that have identified more or less mssm or close to msssm physics; is highly nontrivial and suggestive. "What is the mechanism breaking symmetries and selecting the true vacuum? " Again, unknown depending on what you mean. For instance cosmology likely has something to say about the vacuum selection method, and it might simply be more field theory phenomenology that has to do with say SuSY breaking. "What is the microscopically picture regarding dynamical spacetime including black holes, entropy of the gravitational field etc." Entropy is a macroscopic phenomenon, given by the usual laws of thermodynamics. Black holes satisfy the usual Hawking-Beckenstein entropy bound (which you can actually rigorously show with st). AdS/CFT is probably the framework where the most well defined answers come from regarding what to expect form exact black hole microphysics, but of course this game is still highly incomplete (sometimes you only know answers from one side of the duality).
Thanks a lot. This is exactly what we should do. Try be honest both on progress and open issues. What about the assessment? Second list of bullet points?
Ok this seems to start out as a fair discussion and the setup is perfectly reasonable. Let me get through the numerous points over time, there's too much to say in a single shot. Comment on mathematics: Even ordinary quantum mechanics appears to the outsider as "too mathemaical" (Hilbert Spaces, Matrices) since he/she can't explain it in simple terms to Mom. But that's not the issue, isn't it? It's the proper language to use. So why do we need mathemathical structures like Susy, CY, higher dimensions? I guess here a severe misunderstanding is taking place, even many string theorists are making it. The simple point is that SUSY makes things tractable. One can use the powerful methods of algebraic geometry, holomorphy and so on. Most of the progress in recent years in strings and gauge theory relies on that. Since this is a highly developed branch in mathematics it is natural that much of this has been made use of. But IMHO all this geometrization is a toy model; a very powerful and fruitful one which has allowed to get many important insights in different problems like non-perturbative gauge dyamics, black holes, string dualities, AdS/CFT. But IMHO the confusion is that many people take it too literal and for the real thing. The real thing, the non-susy standard model incl gravity, may not be ever tractable to high accuracy. It is simply because non-susy dynamics is infinitely more complicated, esp when we deal with gravity; only with great luck we may find an approximate (broken) susy at low energies that makes things to some extent tractable. I think that this SUSY-Toy machinery has be proven invaluable, much more than outsiders can possibly recognize. The string physicists are very excited about the progress that was achieved, like state counting in black holes (which eg shows that the theory is consistent and nothing is missing). So they say: great we begin to understand black holes!!! Now other people sit in armchairs and complain that only SUSY black holes were investigated, and no experimental predicitions were made, so all of this is worth nothing; a failure, a dead horse, a waste of ressources. But that's not the point - at this stage it is about to learn how things work! And it is as exciting as non-trivial to see a glimpse of it! So, summarizing, the "heavy mathematics" mostly concerns algebraic geometry, but this is just a powerful toolset to deal with the supersymmertic toy models. One needs to use it for doing actual computations, which are important for understand how, or if, things work. Whether this whole susy framework has anything directly to do with nature, is a different question, and some people try to cook up susy constructions to describe the standard model. It's too early to tell whether this real-world application with bear out or not, I'd be skeptical, but this is just one part of the string theory program and definitely worth a try. Comment on extra dimensions: this is another great piece of confusion. One may loosely say that extra dimensions are one way to parametrize the extra matter, non-gravitational fields. Only a small subset of 4d string constructions have a direct interpretation as geometrical compactifications, and one may formulate everything in a language where extra dimesions don't even appearâ€¦ better just call this sector of the theory "internal" degrees of freedom. And this purported disadvantage of string theory is in fact the opposite, namely a blessing: strings need those internal degrees of freedom for consistency, and in a sense "predict" extra matter. This is what we actually want to have. In "alternative" approaches, such as LQG, AFAIK there exists no consistency requirement that would demand extra matter; so this is put in by hand without any guiding principle - I see this as a desaster for achieving a unification of matter with gravity. No conceptual solution in sight here! Also poeple often do not realize, that due to the dualities, one and the same theory can have different interpretations in terms of geometry and dimensions; in fact there is no absolute meaning of this! One and the same theory may be described as a 4d gauge theory, or 10d string theory compactified on an AdS space; if fact it seems to be a general phenomenon for "ordinary" gauge theories that their strong coupling limit has a higher dimensional interpretation. If one does not like it, one is cordially invited to close the eyes; but others will go on and try to see what can be learned from this extra-dimensional perspective. And a lot can, probably the holographic property of QFT's was one if the most important discoveries in many decades. So in short, extra dimensional physics is automatically built in ordinary gauge theories, this is a computational fact and not an ideology, so there is no way to avoid it for any serious researcher.
Thanks for the long and elaborated answer. Two comments are in order: 1) I can agree with extra dimensions interpreted as internal degrees of freedom. Let's compare it with qm (Hilbert spaces). Assume for a moment that qm would not be able to make experimentally falsifiable predictions. Then we would wonder what these wave functions and Hilbert spaces should be, where they are and how we can measure them. I guess we would come to the conclusion that they are unphysical. 2) It's different with SUSY (already with SUSY in MSSM) and with superstrings. Here the problem is that these entities are not "hidden" mathematical entities but physical ones: directly shows up SUSY in the particle spectrum. And here I see the following problems: SUSY is elegant - as long as you do not try to break it (OK, this is not a very good point :-) SUSY / MSSM does not need string theory (OK, refer to my argument from above: it may be a progress to have MSSM as one solution of string theory instead of just another theory introduced by hand) SUSY has to be verified experimentally sooner or later I think the last bullet point is a serious issue: as long as there is no experimental support, string theory (and even SUSY) is somehow a solution hunting for a problem; and if one does not find SUSY at the LHS one can again say "that one will find it at higher energies"; so in the end it's about a promise again. I studied SU(N) gauge theories for some time, even large-N limit. What about the following idea: is it possible that string theory is nothing else but a large-N approximation of certain supersymmetric gauge theories? If yes, would this kill string theory as a fundamental theory? Taking all your ideas into consideration my conclusion is that string theory is still in an early state of its development; it makes progress, but many people are uncomfortable with its velocity.
The problem with the theory is that it thinks the universe does calculus every time. In genetics there are only four base units and that creates all living things. What if there is something simple that creates all of matter from energy?
It is important to add that the four base units themselves are not so simple on a molecular level. We use abstract theories to visualize them.
(I assume you are referring to ads/cft - f theory for example couldn't care less - in principle - if ads/cft was just plain wrong) There are many reasons to believe that ads/cft can be extended to non-susy, non large N. For example because you can break some pieces of susy, and still get right answers. Polchinski here gives a discussion and a number of references on this sort of stuff http://arxiv.org/PS_cache/gr-qc/pdf/0602/0602037v3.pdf (might want to skip to page 8 to "Lessons, generalizations, and open questions")
No, I am referring to ordinary QCD in the large-N limit. Thanks for the link; I've studied this paper several times but it does not answer my question if it's possible that instead of having some SUSY as low-energy limit of a fundamental string theory, string theory itself is only a (large-N) approximation to a certain fundamental SUSY. If this would be the case string theory would become of less interest. I just want to entrtain this possibility as it eventually dates back to the origin of string theory were one tried to understand hadrons in terms of strings. Assume for a moment that it's true and that somebody finds a proof that some ST is nothing else but a large-N approx to the MSSM.