Two cheers for string theory

By Sean Carroll | July 21, 2005 8:38 am

I am often surprised at the level of disdain and resentment with which string theory is viewed by non-string-theorists. I’m thinking not so much of people on the street, but of physicists, other scientists, and even other academics. As a physicist who is not personally identified as a string theorist, I get to hear all sorts of disparaging remarks about the field from experimental particle physicists, condensed matter physicists, astrophysicists, chemists, philosophers, and so on. I sometimes wonder whether most string theorists understand all the suspicion directed against them.

It shouldn’t be like this. String theory, with all of its difficulties, is by far the most promising route to one of the most long-lasting and ambitious goals of natural science: a complete understanding of the microscopic laws of nature. In particular, it is by far the most promising way to reconcile gravity and quantum mechanics, the most important unsolved problem in fundamental physics. At the moment, it’s a notably incomplete and frustrating theory, but not without genuinely astonishing successes to its credit.

The basic idea is incredibly simple: instead of imagining that elementary particles are really fundamentally pointlike, imagine that they are one-dimensional loops or line segments — strings. Now just take that idea and try to make it consistent with the rules of relativity and quantum mechanics. Once you set off down this road, you are are inevitably led to a remarkably rich structure: extra dimensions, gauge theories, supersymmetry, new extended objects, dualities, holography, and who knows what else. Most impressively of all, you are led to gravity: one of the modes of a vibrating string corresponds to a massless spin-two particle, whose properties turn out to be that of a graviton. It’s really this feature that separates string theory from any other route to quantum gravity. In other approaches, you generally start with some way of representing curved spacetime and try to quantize it, soon getting more or less stuck. In string theory, you just say the word “strings,” and gravity leaps out at you whether you like it or not.

So why wouldn’t anyone be happy about string theory? For one thing, we don’t understand the theory very well. It’s easy to say “replace particles with strings,” but quantum field theory isn’t really about “particles” — particles are just the observable momentum eigenstates in a perturbative regime, not the fundamental building blocks of the theory. At this point it’s a little unclear what the fundamental building blocks of string theory are; there are some reasonable proposals for complete non-perturbative definitions of the theory (matrix theory and AdS/CFT, for those in the know), but connecting these formulations to a more complete picture isn’t easy.

But most of the grumbles about string theory from other physicists aren’t about a complete non-perturbative definition of the theory — they are about the lack of connection to experiments. One often hears that string theory simply makes no predictions, but that’s clearly false. If you scatter two particles together, string theory unambiguously predicts that the cross-section should look stringy, not like that of fundamental point particles. [With caveats discussed in the comments.] The problem, of course, is that the difference between these two possibilities is only noticeable when the energy of the collision approaches the Planck scale (or really the string scale, likely to be similar) — fantastically far away from what we can actually reach in accelerators. So string theory makes predictions, it’s just that we are as yet unable to test them. In other words, string theory is either right or wrong, it’s our challenge to come up with clever ways to figure out which.

There is a matter of principle here that scientists, of all people, should understand. To wit, our current understanding of nature — based on classical general relativity and the quantum-mechanical Standard Model of particle physics — is simply incoherent. It just doesn’t make logical sense. It is very easy to ask questions to which we don’t know the answer: “What is the gravitational field of an electron?” For that matter, since the Sun is made of elementary particles, we can’t even sensibly talk about the Sun as simultaneously a source of gravity and as a source of light and heat. This is not acceptable. Our goal as scientists is to understand how the world works, and relying simultaneously on theories that are deeply incompatible with each other is nothing to be happy with. Even if it won’t help us make a better TV set or understand the mass of the proton, we need to have a coherent theory of quantum gravity.

Recently there has arisen another sense in which string theory purportedly makes no predictions, associated with the “landscape” of possible string vacuum states. Just as in quantum field theory, the observable spectrum of low-energy string excitations and their interactions (that is to say, particle physics) depends not only on the fundamental string physics, but on the specific vacuum state in which we find ourselves. String theory predicts more spatial dimensions than we directly observe, so one of the characteristics of our vacuum is the way in which the extra dimensions are hidden from our view. It now seems quite plausible that the number of possible ways for this to happen is enormous — perhaps 10500 or so. If true, this puts a damper on the hope that string theory would predict a unique vacuum state, and we could explain (for example) the ratio of the muon mass to the electron mass from first principles.

Well, too bad. It would have been great to make such predictions, but the inability to do so doesn’t render string theory non-scientific. The appropriate comparison for string theory is not to “the Standard Model of Particle Physics,” it’s to “quantum field theory.” Nobody complains that there are a huge number of possible quantum field theories, and we actually have to go out and measure the properties of actual particles rather than calculating them using pure thought. If string theory turns out to be the same way, that’s life.

My own view is that string theorists have been a victim of their own characteristically aggressive form of optimism. Not only, we are told, is string theory a consistent theory of quantum gravity, but it’s a theory of everything, gives us wonderful new insights into gauge theories, and possesses a mathematical beauty that is so compelling that the theory simply must be correct. These kinds of arguments just don’t carry that much weight with the non-converted. If I were in charge of the string theory public-relations machine, I would be emphasizing over and over again the basic feature that we’ve understood for a very long time: it’s the most promising way we know to quantize gravity. If there were multiple very successful ways to quantize gravity, it would be important to distinguish between them experimentally; but so long as the number of successful models is less than or equal to one, it makes perfect sense to make every effort to understand that model.

Which is not to say that we shouldn’t also pursue alternatives. I’m all in favor of supporting research on loop quantum gravity, dynamical triangulations, causal sets, and whatever else smart physicists might personally find promising. As long as we don’t know what the correct theory is, individuals need to use their own judgment about what clues to follow. String theory, starting as it usually does from talk about perturbative excitations propagating in a background spacetime, will not seem especially compelling to someone who thinks that background-independence is the most profound feature of gravity. It’s certainly good to support plucky Apples and Linuxes in the face of the Microsoft-esque dominance of the string theory approach; you simply can’t tell ahead of time when someone will hit on a brilliant new idea.

On the other hand, string theory has thus far been fantastically more fruitful than any other idea. When you get into string theory, one of the things that keeps you going is that you don’t get stuck — the rate of progress waxes and wanes, but the progress is very real. It didn’t have to be true that the five string theories studied in the 1980’s would turn out to all be part of one big theory, but they are. It didn’t have to work out that the entropy of a black hole calculated from semiclassical gravity ala Hawking would be equal to the entropy of a corresponding gas of strings and branes, but it is. It’s clues like these that keep the believers moving forward, hoping to understand both the inner workings of the theory and its ultimate connection with what we observe. We interested outsiders should be cheering them on.

CATEGORIZED UNDER: Academia, Science
  • Alejandro Rivero

    There is a problem of principle with the idea of the string: Where is length? In QFT, length is in the background spacetime, in String theory, it is both in the background and in the particle. For some years strings theoretists promised background free theories, so this problems was to be removed in the future. I do not know if the future is already here. It should, if one thing that Superstrings are already more than thirty years old as a theory.

  • WL

    > It would have been great to make such predictions, but the inability to do so doesn’t render string theory non-scientific. The appropriate comparison for string theory is not to “the Standard Model of Particle Physics,” it’s to “quantum field theory.”

    That’s exactly how I view it – a framework generalizing QFT and
    including gravity. We had overwhelmingly strong clues since 20
    years that string theory, in its current background-dependent
    formulation, would not predict a unique vacuum state but rather the
    opposite of it, but that has never bothered me working on it since

    A more modest goal, namely trying to better understand the *principles
    according to which things work* (gauge theories, black holes,
    D-branes, say) leaves more than plenty of things one can sensibly
    do, even without making predictions that would be experimentally
    testable today. There are so many highly non-trivial phenomena,
    which often can be tested via “theoretical experiments”, that it
    would be foolish not to investigate them and see what one can learn.

  • Clifford

    Wow. I go for a long working lunch, and come back to find that Sean has written a nice long post on strings that I’d promised TM in the comments of another post I’d get around to writing one day. Thanks Sean, this is great! I can go back to lunch and continue thinking about my string theory problem. -cvj

  • Robert

    Ouch, calling string theory the Microsoft of quantum gravity hurts!

  • Robert

    I am not sure that string theory is to be compared with QFT. In the end, it’s really just one theory (or very few) and it is the state rather than the theory that varies. However, this is dangerous to say if all we know to do is pertubation theory around states as the other states might be very far away (dynamically).

    Here, I expanded about this a bit. In short: If there were only one state in string theory, that would be quite scary, as it would (if strings are realized by Nature) predict everything. Everything, including everything you see right now. And smell right now. And (maybe) think right now.

    It there were only one state in Maxwell theory, you could (in princliple) compute what your radio would be playing in an hour.

  • Dan

    Nice job, Sean. Very fair, balanced and informed description of what’s going on. You are so many orders of magnitude more intelligent and reasonable when you talk about physics, rather than politics. Stick to the former please.

  • Cameron


    As a lay-person in the physics realm, could you explain what guage theories are?

  • Levi

    I think you promised a post giving your view of string theory way back in the Preposterous Universe days. Glad to see it here. I love the new blog, even the posts about dimwitted movies (e.g. Mr and Mrs Smith) Only, where are the poems? Where are the pictures? But I do love the new blog, and all of your accomplices.

  • Sean

    Cameron– Gauge theories are basically generalizations of electromagnetism. In E+M, the behavior of both the electric and magnetic fields can be described in terms of a single four-dimensional vector, the “vector potential.” But there is a symmetry: you can shift the vector potential by the derivative of a scalar field, and the corresponding values of the electric and magnetic fields don’t change. That’s gauge symmetry. It can be generalized to more complicated kinds of symmetries, e.g. SU(2) etc. The Standard Model makes heavy use of gauge theories, to describe the strong and weak forces as well as electromagnetism.

  • Geoff

    Ignore Dan. Not only are people who try to tell others what to write about annoying, your politics posts provide a nice counterpoint to the physics. It creates a richer blog. Dan can go get his own site if he doesn’t like what’s written here.

  • Dan

    Stop whining, Geoff. The comment section is for comments on what’s been posted, you know. I like it here, even if I don’t agree with the politics. It’s you snotty elitists left-wing liberals who can’t stand other people’s opinions. I’m quite all right with the diversity. I still like to argue, debate and comment.

  • Kuas

    I fell asleep somewhere in the second or third paragraph of this post. Anyhow, I guess publicly going to bat for string theory is not a bad move when you find yourself on the job market.

  • Philip Downey

    I watch string theory from the sidelines as a journalist and read about if often. There are some aspects that are often a oversold and the scientist should probably frame their comments a little differently.

    As you point out, it’s not a theory of everything. The theory of everything tag sounds like it’s taking away free will, and a lot of people react negatively to that.

    Also, since it’s not testable with today’s accelerators, I always wonder if it should be the string hypothesis. I understand the math is well worked out and waiting for confirmation. Does anybody know when quantum mechanics and general relativity went from hypothesis to theory?

    The conflict between quantum mechanics and general relativity is always mentioned, but the details usually aren’t, or summarized glibly as fuzzy and jumpy quantum/versus flexible spacetime. The light flashed over my head this year when I finally read someone explain that general relativity’s bendable spacetime conflicts with the rigid spacetime of quantum mechanics, where things don’t bend.

    Your new blog is interesting so far. I can’t say I ever like the “Continuing reading post…” option. The full version is only text and it’s not like the whole post slows down loading. And it’s easy to spin the mouse wheel to the next post.

  • Bob McNees

    “I am not sure that string theory is to be compared with QFT”

    Maybe not with if you’re talking about states, but I think it’s a very good comparison to make when people raise the issue of a landscape as signalling some sort of fatal flaw in string theory (This is a point I recently raised on Peter Woit’s blog, maybe foolishly). Our understanding of string theory is far from complete. Holding up a tentative observation as evidence that string theory is wrong or, worse, unscientfic is…silly.

    As Sean mentions you can start writing down field theories and never finish. Imagine that you were to write down all field theories and then try to find a “realistic” theory that describes whatever experiments you are interested in. One way of narrowing down the field is to do experiments and tune the parameters of your generic theory, but it’s not until we understand certain selection principles at work in the space of field theories that we can really get anywhere. Gauge invariance allows us to reduce an infinite number of ostensibly different theories into a single theory (imagine writing down all of the gauge fixed versions of a theory with a continuous gauge fixing parameter). Or suppose, after making some obscene number of measurements, we still find ourselves with an infinite number of field theories consistent with our observations. It’s not until we incorporate ideas like the renormalization group and effective field theory that we really understand what it means to distinguish two theories at the scales set by our experiments. Ideas like this completely change our original goal, which was to identify the unique field theory (out of an infinite number) that describes the results of our experiment. Instead, we make due with the fact that our tools (measurements up to some scale) for picking out a particular theory will never completely pin things down (above that scale). We’ve revised our question by understanding effective field theory.

    If we didn’t understand all of the principles at work in QFT we would be up the creek trying to answer that original question. We wouldn’t even understand why we weren’t making progress on a question that turned out to be naive.
    That wouldn’t represent a failing of QFT, just our understanding of it.
    Asking “which of these field theories is the right one?” seemed like a reasonable question at first, but as we learned more about QFT it needed to be refined. Understanding a theory means, among other things, knowing what kind of questions you can ask and how to ask them.

    With respect to the landscape, string theory is in a similar situation. We have ideas about how to formulate string theory, but certainly not with the degree of generality that we would like. We know what the proper observables are in some situations, but not others. In some ways our understanding is comparable to the hypothetical field theorist in the previous example: we’ve learned a tremendous amount but we know there are important principles that we still don’t understand. The landscape may be an important question for string theory, or it may just be an artifact of our own ignorance.

    Based on the successful aspects of string theory that Sean describes, I think it’s reasonable to suspect that string theory is full of powerful and sophisticated mechanisms that will either eliminate the landscape or change the question. That’s my guess. Can I prove it? Not yet, but I’ll work on trying to. Of course, maybe I’m wrong and string theory contains no means of singling out a unique vacuum: i.e. there really is a landscape. What Sean says still applies. String theory isn’t automatically rendered unscientific, we’ve just learned that it doesn’t have anything to say about one of our questions.

  • Peter Woit

    Hi Sean,

    As you might guess, I strongly disagree with some of your claims about string theory, more specifically:

    ‘One often hears that string theory simply makes no predictions, but that’s clearly false. If you scatter two particles together, string theory unambiguously predicts that the cross-section should look stringy, not like that of fundamental point particles.”

    Well, what exactly is the prediction? Let’s say that tomorrow someone figures out how to make an accelerator that collides electron-positron pairs at 10 times the Planck energy, with as much luminosity as you want. What will the detectors see? You say the cross section won’t be point-like, but will be “stringy”, what exactly does that mean? First of all there’s the problem that you don’t know what the string scale is. It’s a parameter times the Planck scale, but this parameter could be anything. So, whatever your “stringy” effects are, you don’t even know at which energy scale they’ll show up.

    OK, this is just one parameter, and if one could make predictions that just depend on one parameter that would be great. You’re implicitly assuming that this parameter is small, so small that perturbative string theory gives a series that, while divergent, is a useful asymptotic expansion, i.e. cutting it off after the first term or two gives a close approximation to the true result. Why do you believe this is true? You’re assuming that ignoring all non-perturbative effects is a good approximation? Why? Presumably by “stringy” you mean that one will see a spectrum given by the vibrational modes of a string. How do you know that, before you get to the first vibrational mode above the vacuum state you won’t start producing black holes (or “branes” of some other kind)? Sure there are characteristic signatures of perturbative string theory, but my claim is that there aren’t any for non-perturbative string theory and that’s the real thing.

    The existence of an infinite number of very different vacuum states around which you can build perturbative string theory expansions makes the issue more pointed. If you believe that perturbative calculations are valid and so the theory can make predictions, you also have to believe in the infinite number of vacuum states, which makes the theory radically non-predictive at energies we can ever hope to measure. Most string theorists I know would prefer to believe that perturbation theory is no good for determining the vacuum state, that non-perturbative effects will pick out a vacuum that looks like our world. But you can’t have it both ways: you can either believe that perturbation theory is a good approximation (although there’s no good reason for this, the question is undecidable until you have a non-perturbative theory), in which case there’s a “stringy” spectrum, but no hope of predicting anything at accessible energies, or you can believe that non-perturbative effects are important, in which case you can hope for low-energy predictions, but there is no characteristic prediction for what happens at the string scale.

    So, again, you’re claiming that string theory makes predictions that we could check if we had a high enough energy accelerator. What are they, and if you’re using perturbative string theory, how are you dealing with the fact that by itself perturbative string theory is an inconsistent theory?

    I’ll comment on some of the other points in your post separately.

  • Peter Woit

    Bob McNees is ignoring the long response I wrote to him on my blog explaining the problem with his analogy about QFT and string theory. You can read it there, but the fundamental point is simple. QFTs make an infinity of well-defined predictions about low-energy physics. You can go out and compare them to the real world. You then see that one of the simplest QFTs agrees precisely with everything experimentalists have seen. Experimentalists do more experiments and get results agreeing with what this QFT predicts. This is a beautiful example of the scientific method. The utter lack of any connection between string theory and experiment makes it something completely different, and raises real issues of whether it is a science at all, especially if anything like the landscape exists.

  • Peter Woit

    About some of Sean’s other points:

    1. Why there is some “disdain and resentment” about string theory among other physicists.

    First of all, until recently there was very little of this. I’ve now spent twenty years complaining to physicists of all different kinds about string theory. Until the last couple years, the reaction I got from most string theorists was that I was an idiot, too stupid to understand the theory. The reaction from non-string theorists was that they weren’t so sure themselves about string theory, but many string theorists were geniuses and I should be careful about criticizing my betters. I made attempts to publish some rather moderate criticisms of string theory, and found these thwarted, generally by non-string theorists who were pretty convinced that only a crackpot would be claiming that string theory was completely on the wrong track.

    Things have changed a lot during the past few years. First of all, because of the internet, venues have appeared in which the problems with string theory can be laid out extensively and in public. Secondly, despite what Sean says, progress on string theory has virtually come to a halt, and many of the people doing it have become discouraged or even left the field. Finally, the apparent existence of the landscape has led to a large number of prominent string theorists engaging in what is obviously pseudo-science. The words many physicists, string theorists and non-string theorists, use to describe what is going on are not printable in a family-type blog like this one.

    Fundamentally, more and more people in the physics community are beginning to feel that they’ve been had. The theory groups at Harvard, Princeton, Stanford, etc., etc. are completely dominated by string theorists. String theorists have collected every reward academia has to offer (except the Nobel Prize) for work that Lawrence Krauss accurately describes as a “colossal failure”. This is driving a lot of the “disdain and resentment”. The incredible degree of over-hyping of the theory that has gone on is part of this. When you spend twenty years going on and on about how beautiful and wonderful your theory is, and then it becomes clear it doesn’t work, you shouldn’t be surprised that some people don’t think very highly of you.

    2. There’s a lot to be said about the generalized claims for successes of the theory that Sean makes, but to do this properly is a long story. I’ve done it elsewhere, and since I’m about to leave on a trip, I don’t have time to do this anyway. I hope a useful discussion will ensue here, I’ll try and write more later on my own blog and stop making long comments here.

  • Joe Bolte

    Dr. Carroll,

    You wrote: “If you scatter two particles together, string theory unambiguously predicts that the cross-section should look stringy, not like that of fundamental point particles.”

    Does this mean that you can use string theory to write an expression for the differential corss-sectoin, d-sigma/d-omega, which will differ from the one predicteed by the Standard Model? If so, please either tell me what this expression is, or direct me to a paper where one is derived.

    If no one can write such an equation, in what sense does “string theory unambiguously predict that the cross-section should look stringy”?

    Best Regards,

  • SM

    “If string theory turns out to be the same way, that’s life.”

    Not sure if that was an intentional pun..but it’s hilarious :-) If anthropic reasoning does turn out to be the crucial factor in selecting a string vacuum (the physicist in me does kind of recoil from this kind of approach, I’ll admit it) then perhaps biology really will turn out to be the science of the 21st century!

  • Peter Erwin


    Interesting post (and interesting blog!). I have a couple of brief comments from an outsider’s perspective (I’m an astronomer).

    I think part of the grumbling you’re concerned about stems from what some people see as the overselling of string theory — claims that it would solve all sorts of problems which it now looks like, perhaps, it won’t. It’s interesting that you seem rather sanguine about string theory maybe not being able to provide unique values of e.g. fundamental particle masses, etc., since that was one of the claims that people were making, at least at the popular level: that string theory could explain all the arbitrary values that had to be measured and put into the Standard Model by hand.

    Genuine predictions about Planck-scale effects would certainly be a nice start, but I’d argue that an unspoken part of the idea that “a scientific theory makes testable predictions” is that these really should be feasible predictions. Few astronomers would take (very) seriously an astrophysical theory whose tests required, say, 10-million-kilometer-diameter telescopes, or observations over a period of several thousand years.

    Brief reply to Philip Downey’s question: The first “experimental test” of GR was the solar eclipse observations of 1919, which is pretty good for a theory published in its complete form in 1915. The test could have been done with earlier eclipses, but World War I got in the way. (I think Einstein’s first prediction of a measurable gravitational lensing effect was published around 1912 or even a little earlier, and there was apparently an attempt to check this with a 1913 eclipse — but bad weather made it impossible). This is what got Einstein his international fame, though I don’t know how long it was before GR was generally accepted by the majority of the physics community.

  • Bob McNees

    Hi Peter,

    Please don’t take me not posting on your blog yet the wrong way. I will get around to it.

    I don’t think you’re responding to the argument I’m making. When you say that QFTs make lots of predictions about low energy physics you’re really talking about effective field theories. If you give me enough data I can reconcile it with an effective field theory. Furthermore, that effective field theory will provide me with a sophisticated framework for predicting the outcomes of other experiments. Inherent in its definition are the criteria for what I can and cannot predict.

    This is very different than what I mean when I ask you to pick out a quantum field theory using some finite set of low-energy measurements. Forget all the things you learned that led you to appreciate the notion of effective field theories. Write out the most general lagrangian you can. Put everything in it. Include all the interactions consistent with the most basic principles of field theory. Don’t exploit any internal symmetries to simplify things, and don’t forget all the interactions which, if you knew about renormalization, you would classify as unspeakably irrelevant. Now set up some experiment that you think should be described by quantum field theory. Make all the measurements you want but keep them below a certain scale, and tell me what QFT you think describes that.
    Whatever set of field theory parameters you give me, I know that there is going to be some very irrelevant (and probably very fine tuned) interaction that I can
    add that won’t be ruled out by your data. But I certainly don’t claim that QFT doesn’t explain your experiments. I just refine my original question and learn to ask about effective field theories.

    I know this example sounds contrived to you. We know a lot about QFT. Given an effective field theory we understand when it should be valid, and we know that my example of adding an ultra-fine-tuned irrelevant interaction isn’t relevant to the experiments you are doing (no pun intended). But there are a lot of deep things we must understand about QFT before we can think in those terms. They all represent clever and hard-fought answers to difficult problems. The scientific method isn’t just the part you mention (doing more experiments and getting more agreements). It also involves figuring out what the theory can and cannot tell you, why that is so, and how you go about asking the meaningful questions.

    Answering those last few questions is what allows effective field theories to make the predictions you refer to. I think we all appreciate how non-trivial this is. What would you have said to early field theorists when their calculations came up full of infinities? Would you have told them to give up? It seemed that their theories were making predictions which could not be meaningfully compared with experiment. Lots of people gave up on it for just this reason. A few people stuck with it, and it payed off. By your standards how could they ever have made progress? Nothing you’ve said about QFT is meaningful without their resolution to this problem.

    I don’t think string theory gets a free pass on most of the questions you raise. I think that some of them represent good, genuine criticisms that should interest all of us. But why immediately hold string theory, a young field, to the same standards you apply to a mature subject like quantum field theory? We understand much more about QFT than we do about string theory. There are lots of problems in QFT where we understand the resolution. Previously, one might have thought that those problems rendered QFT incapable of making predictions about the real world. Its *only* because we solved these problems that we understood precisely what predictability means in QFT. So, my comparison is not between string theory and effective field theory. Rather, I’m comparing the current state of string theory with QFT before all of the developments that proved to us that it provided a consistent, predictable framework. That seems like the more relevant comparison at this point.

  • Bob McNees

    When Sean mentions string theory predictions for scattering, I assume he’s referring to the different high-energy behavior one would expect. Field theory amplitudes are hard (power law) at high energies. String theory amplitudes are soft. If you ever do a high energy scattering experiment and see a soft amplitude, you have strings.

  • Haelfix

    I said this elsewhere, but in QFT there is a distinct notion of minimality. The correct effective field theory for our world is almost the simplest one you can write down.

    ST currently lacks this, but thats not to say such notions as the previous commentator wrote won’t appear (renormalization group flow, etc)

    The problem as I (naive physicist with little string theory background) see it is cosmological in origin. We observe that we live in one vacuum, and it seems therefore that we must demand the meta stable desitter vacua to not decay or tunnel to some other spot on the landscape, with error bars to at least the order of the age of the universe (observed as uor laws of physics seem to not have changed). This will require enormous, perhaps infinite amounts of fine tuning, not just in our vacua but in all the other infinitely close landscape modes.

    The anthropic principle (or principle of arrogance) does away with that, it just says it is that way because thats what we observe. However if such notions appeal to you, I claim its simpler to just say the world is the standard model with semiclassical nonrenormalizable gravity. Both have infinite amounts of fine tuning, and the latter is simpler.

  • Ann Nelson

    Hi Sean

    I am glad you posted this. As a phenomenologist/model builder who got her PhD in 1984 and found precious few jobs available which were not reserved for string theorists, I shared the disdain for and dismay about string theory that you are trying to counter. In the 80’s, many string theorists were clearly out of their minds–for instance I heard one famous theorist exult “now we don’t need experiment anymore”, and another made a bet that the elementary particle Yukawa couplings to the Higgs would be computed within 10 years.

    The second string revolution changed my mind. I still think most string theorists are way too arrogant and unscientific in their unshakable beliefs. Also the sociology of the field has many pathologies. That said, the field and its practitioners have redeemed themselves by discovering deeper and more interesting insights into Quantum Field Theory then ever would have been found by phenomenologists or by traditional formal quantum field theorists. Spin-offs of string theory including D-branes, AdS/CFT, orbifolds, dualities, supersymmetry, and new dimensions have all found their way into testable particle physics phenomenological model building, and particle theory would be would be completely moribund without them. Also some interesting experiments have been done in response to some of the string-inspired models, such as tests of the gravitational inverse square law at the sub-centimeter level, that would not otherwise have been conceived, and which very well might find something revolutionary. Other approaches to quantum gravity dont offer many such spin-offs, as far as I can tell, and neither do other formal approaches to quantum field theory.
    Finally, as to whether string theory and its predictive power should be thought of in an analogous way as quantum field theory—ADS/CFT has taught us that string theory IS quantum field theory and vice versa. For instance QCD, which is certainly a quantum field theory, can very likely also be formulated as a string theory and the ADS/QCD approach shows that the latter formulation can give interesting insights into phenomenona such as vector meson dominance and confinement.

    I certainly wouldn’t defend every paper in the field, including some of the ones Peter Woit uses to disparage the whole enterprise. But I now would hate to see the pendulum swing so far that string theorists have as hard a time getting jobs as phenomenologists did in the 80’s.


  • Scott

    So I have a question about the claim that ST predicts gravity that “you just say the word “strings,” and gravity leaps out at you whether you like it or not.” Does it simply predict the existance of a 2 spin, massless, chargless, ect particle or does it also predict that all particles emit and absorb this graviton based on their energy and momentum as expected? Also does the existance of a particle that is also predicted by all the failed quantizations of gravity really a good thing for a theory to predict?

  • Peter Woit


    First of all, I’m not talking about QFTs as effective field theories. Non-abelian gauge theories with a small enough number of fermions have perfectly consistent high energy behavior, and make sense at all energies. I understand perfectly the standard ideology these days about renormalizability not being a fundamental criterion and that QFTs are only effective theories valid at low energies for some more fundamental non-QFT. I just think it’s probably misguided.

    When QED was first formulated in the late 1920s, it immediately explained very well a huge array of experimental data. No one knew how to handle loops, but as any experimentalist can tell you, you don’t need to calculate loops to get highly accurate approximate answers for just about anything they can measure. So, from the time it was written down, QED was a theory with very strong grounding in experiment. You could worry about its problems of principle and people did, but nature had told us incontrovertibly that there was something very right about the theory. It took twenty years to figure out how to handle higher order calculations, but during much of that time the physicists involved were not working on this, but either trying to save their own skins or figure out how to incinerate cities. I think any analogy with string theory is ludicrous. After more than twenty years of full-time work by thousands of physicists, there is still not a single connection to experiment, and people are now devoting their efforts to more and more convoluted excuses for this situation.

    I know perfectly well what the tree-level amplitude in perturbative string theory looks like. I just don’t think this allows you to say anything about what amplitudes for the conjectural full non-perturbative M-theory look like. It seems to me you’re trying to tell me it’s a “prediction” of string theory that you can ignore non-perturbative effects in this case, although you hope you can’t ignore them in determining the vacuum state of the theory. This doesn’t make sense.

  • Peter Woit

    Hi Ann,

    I’ve always been careful to not “disparage the whole enterprise”, but to distinguish between those parts of string theory that have been fruitful (AdS/CFT, connections to QCD, mirror symmetry and other applications to mathematics), and those which have been complete failures (attempts to unify gravity and the standard model in terms of a 10 or 11 dimensional supersymmetric theory of strings, branes, etc.). The problem here is that the part of string theory Sean is promoting is the part that has failed.

    While string theory has led to new insights about QFT, there has been a huge opportunity cost. Do you really believe that if Witten and thousands of others had devoted the time they have spent thinking about string theory thinking instead about QFT they wouldn’t have come up with something interesting?

    I’m not interested in making it hard for string theorists to get jobs, but in making it possible for smart, young, mathematically sophisticated people who don’t believe in string theory to be able to get a job if they decide to pursue a non-string theory topic. While phenomenologists and string theorists can get jobs, I don’t see much place in the field these days for people with more formal, mathematical interests who don’t want to do string theory.

    You say you wouldn’t “defend every paper” I’m supposedly using to disparage the whole field. Presumably we’re talking about the Susskind et. al. landscape stuff. What about not defending every one of them, but acknowledging that these things are not science?

  • ohwilleke

    The big problem with string theory is that it predicts a whole lot of things (most notably supersymmetric particles, branes and extra dimensions) for which we have little or no evidence, while predicting few things which we can compare to practicable experiments.

    String theory does, as noted earlier, also suffer from being oversold. String theory has most definitely been sold as a means of determining fundamental particle masses in the standard model from first principles. Indeed, one of the main reasons people are driven to look for something deeper than the standard model at all is that abundence of constants that seem to have some sort of relationship, but precisely what that relationship is, isn’t terribly obvious. String theory doesn’t appear to have made meaningful progress on that front.

    In contrast, quantum gravity theories that are focusing on backgrounds — the loop quantum gravity people, the dynamic triangulations people — are making at least comparable progress without a lot of the ugly sides of string theory. These provide quantum gravity paradigms in which a 4D macrouniverse emerges naturally, without having to use supersymmetric particles, maybe even without having to have a graviton at all. Those theories are just starting to develop, just a string theory is still not a really mature theory, but they seem likely to have more testible predictions. LQG proposes that neither the Big Bang or Black Holes are true singularities, and most of the background oriented theories call for a non-abelian gauge field that for sub-Newtonian effective field strengths at short ranges, and super-Newtonian effective field strengths at very long ranges (in a bit of an analogy to QCD). These are things we can be tested macroscopically, and one would expect that a quantum gravity theory, be it stringy or not, ought to have so macroscopic manifestations. In the same way, while QED occurs at tiny scales, you can do simple inteferrence, oil slick rainbow and tunnelling experiments that tease out its distinctive properties.

    String theorists also come across, rightly or wrongly, as remarkably tone deaf to developments to other fields of physics that could be pointing us in the right direction. Where are the string theorists pointing us to places to look for CP violations? Where are the string theorists explaining in advance that neutrinos must have mass?

    We’ve had some of the brightest people in the world looking at the math for thirty years without much in the day of conclusive answers. This is the sort of thing that doesn’t happen unless a piece of the puzzle is missing.

    Extraordinary claims require extraordinary proof, and string theory definitely makes extraordinary claims while offering little by way of proof.

    String theories also, and the original post here is no exceptions, have done a poor job of pinning down what the theory says so far. One of the sociological reasons that the standard model has taken such prominence in the minds of educated laymen is that it has offered up a “small catechism” of what it does say, in the form of a big “periodic table” style document, the Feynman diagrams, and a fairly spare theoretical framework.

    If String theory is to get wider acceptance with the educated public, it needs to be able to offer more of a catcheism of central points that string theory has established, rather than simply saying “its great and beautiful and it does everything, trust me.”

  • Ann Nelson

    Hi Peter
    I am saying it now seems just as silly to say one “doesn’t believe in” string theory as it is to say one doesn’t believe in QFT. It is such a demonstrably useful tool that I would indeed be suspicious of a formally inclined theorist who refused to use it as some matter of principle. To say one doesn’t believe in all the nonsense ond hype of the 80’s about how one will use string theory to calculate all the parameters of the standard model and make predictions about what lies beyond is different, but practically noone does believe in that stuff anymore. Some people still think a supersymmetric extension of the standard model will be verified by LHC, but that is a perfectly testable and scientific theory. It does mesh well with how some people think string theory will unify gravity, and I view that model as a spinoff of string theory, not the most interesting but reasonably well motivated. Personally, I think “believing in” speculative ideas which do not yet have sound experimental evidence is nutty, but I think one can and should do research on them.
    ( I must admit I pretty much agree with you, (and Lubos!) about the papers on ‘predictions’ from the landscape. Such papers are not a big part of string theory however, despite the impression one might get from reading your blog, and in particular, hardly any young people are involved.)


  • Bob McNees

    “When QED was first formulated in the late 1920s, it immediately explained very well a huge array of experimental data. No one knew how to handle loops, but as any experimentalist can tell you, you don’t need to calculate loops to get highly accurate approximate answers for just about anything they can measure.”

    But you knew you *should* calculate loops, and that if the theory made sense then so would those calculations. There were plenty of examples in QFT where loop calculations seemed to not make sense. Should everyone have been willing to overlook that? Now, I know that QED also agreed with a lot of observations (all of them), and it was data and not an appeal to unifying gravity and quantum mechanics that buoyed the hopes of researchers trying to reconcile the obvious applicability of QED with the mathematical uncertainty of the techniques it entailed. But if all you had was tree level QED, would you be satisfied with that? I thought we were talking about quantum field theory.

    “It seems to me you’re trying to tell me it’s a “prediction” of string theory that you can ignore non-perturbative effects in this case, although you hope you can’t ignore them in determining the vacuum state of the theory. This doesn’t make sense.”

    I haven’t made any claims about predictions, with or without quotation marks.
    But I don’t understand your argument here. There are plenty of physical processes in field theory that are described perturbatively, and there are other phenomena in the same theory which are inherently non-perturbative. Are you saying that if there is any question for which non-perturbative physics is important, then I should be suspicious of all perturbative results in that theory?

    If your argument is that we should never do theory in any energy range beyond those we can probe in the immediate future, then fine. We can agree to disagree.


  • Peter Woit

    Hi Ann,

    I agree with you that it doesn’t make sense to just say one “doesn’t believe” in string theory. It certainly is a non-trivial and interesting structure, with important connections to QFT and mathematics. But it does make sense to say one doesn’t believe in the specific conjecture that launched the whole business in 1984 and that has been hyped continuously since then. This is the conjecture that, at a fundamental level, the world is decribed by a superstring in the critical dimension (10), or its extension to M-theory.

    I don’t think the people working on the landscape are fools, abandoning science for no good reason. I think they’re doing it because that’s the corner they’ve been backed into by the conjecture they refuse to give up on. Over the last couple years, I’ve found it incredibly shocking to see the lengths to which smart people will go rather than admit than something doesn’t work, one reason I’ve gone on about this quite a bit in the blog. String theorists desperately need to change the way they do business, starting with an honest evaluation of where the subject is, what works, and what doesn’t.

  • Peter Woit


    Of course I’m not saying one shouldn’t do speculative work on things which we can’t test experimentally in the near future. Hey, I work in a math department and happily spend lots of time thinking about stuff that has no possible connection to experimental physics. I’m not a phenomenologist like Ann, and I think what particle theory needs is to encourage people to spend time thinking more deeply about QFT, not worrying for a while about how to connect up to experiment. String theorists are encouraged to do this kind of thinking and have come up with interesting things because of it. But if the road to progress beyond the standard model requires deeper insights into QFT that don’t have anything to do with string theory, we’re not going to get there as long as only string theorists can get jobs doing speculative work.

    But when one chooses what speculative idea to work on, one needs to have something that tells one when one is on to something. In the case of QED, the fact that tree-level worked so well meant that speculating about the full theory was a very well-motivated thing to do. Speculative work in string/M-theory hasn’t been able to get any evidence from experiment that it’s the right direction to be going in. This seems to me reason to be skeptical that it’s the right direction, and seeing how things have gone the past twenty years appears to me convincing evidence that it’s a wrong direction.

    My comments about perturbative string amplitudes have to do with Sean’s original claim that they are a “prediction” of string theory. I just don’t think such a claim is consistent with the whole string theory research program, in which the idea that there is some unknown non-perturbative version of the theory that will explain low energy physics plays a large part.

  • Ann Nelson

    It makes no sense to say that one believes in or doesn’t believe in “the conjecture that at a fundamental level, the world is decribed by a superstring in the critical dimension (10), or its extension to M-theory.” There is no way to test that conjecture as it stands because the physics implied by that conjecture at experimentally accessible energies is nearly infinitely non-unique. Everyone admits that now.

    One can however, formulate more extended versions of the conjecture, with some specifications of what is to be done with the extra dimensions, the supersymmetry, the role played by branes, fluxes etc. Perhaps one can use very early cosmology to motivate some such extended version. In the end one has some effective field theory which one can test. I’m not a huge fan of such “string phenomenology” but I admit the resulting effective theories of physics beyond the standard model are sometimes interesting in their own right, particularly when they involve concepts like supersymmetry or D-branes that particle theorists otherwise would not have thought of. If some such model turns out to be verified at the LHC I would not view that as evidence for string theory, but rather as vindication of string theory.

    So one cannot test “the conjecture”, unless it turns out that the string scale is much lower than the Planck scale , or that some superstrings are observable cosmic strings, or that decoupling actually fails in some subtle but observable way. It is not a useless conjecture however, since it has inspired some good ideas. The fact that it also inspires some bad ones isn’t really the fault of string theory. There are many more bad ideas out there which were not string inspired.

    You could blog about the bad papers some people are writing in any subfield of theoretical physics.

  • Sean

    Man, you try to go do some work and people fill up your blog with comments. Most of which are great, by the way, thanks. Bob has done a good job of expressing the perspective of a real string theorist, Ann has put her finger on why a phenomenologist might be impressed with string theory, and Peter W. as always mounts a staunch critique with good points. The one sociological point of Peter’s with which I might completely disagree, which I don’t think has yet been challenged by anyone, is the idea that “many of the people doing it have become discouraged or even left the field.” That hasn’t been my experience at all — people will grumble good-naturedly that things have been relatively quiet and it’s about time for another “revolution,” but seem largely still just as excited and optimistic as ever. Probably we have been interacting with different string theorists.

    As far as substantive points are concerned, I guess I need to clarify my point about making predictions about scattering. As Bob mentions, string amplitudes differ in being “softer” in their high-energy behavior than point particles. (Joe, yes there are explicit expressions here; I’m not really the one to give you a relevant example, perhaps a real string theorist will help you out.) My point was that in principle you don’t need to go to “ten times the Planck scale” to see these differences; with arbitrarily precise experiments, you could see them at low energies. Such precision is completely out of reach, of course, but that’s not string theory’s fault.

    I don’t really understand the harping about non-perturbative effects. They might indeed be important in choosing which vacuum we live in, just as they are in field theory. That doesn’t mean we can’t make any predictions that are safely within the purview of perturbation theory. Scattering well below the string scale would seem to be one example.

    I think I disagree with Ann’s pespective in the last post, although perhaps our disagreement is purely semantic. The thing that everyone admits is that it will be hard/impossible to uniquely derive a low-energy effective field theory from string theory. (Probably “everyone” is an exaggeration.) But I think there is some conjecture that amounts to saying “string theory is the correct description of quantum gravity,” which is either right or wrong. (And the correctness of which can only ultimately be decided by experiment, whether or not those experiments are ones we know how to do.) It’s hard to formulate precisely what the conjecture should be (at least for me), since we don’t understand string theory perfectly. I’m guessing that it would involve the stringyness of perturbative processes in quantum gravity. Any experts want to take a stab?

  • Sean

    I hope that at least many people understood the connotations of the title of the post. It’s a variation on an essay by Steven Weinberg, “Two Cheers for Reductionism.” The point is that two cheers is a good number of cheers, but is not quite three cheers, which would be really good.

  • Peter Woit

    Hey Sean, first a user-interface complaint: I just wrote a comment, hit submit without first putting in my name, and e-mail. An error message then appears, and the comment vanishes irretrievably. Ah well, I’ll try again.


    I think there are two ways in which a conjecture can fail: it can predict something wrong or it can turn out to be vacuous and predict nothing. If you agree that the unification via string theory conjecture is consistent with just about any low energy physics, than my point of view is that is that it has failed for the second reason.

    Honestly I know of no other scientific field in which leading figures are engaged in anything half as unscientific and bizarre as the landscape studies. This isn’t garden variety bad science, and I think the reason it isn’t is important and worth examining.


    Here’s a more specific version of my argument. Maybe I’m wrong about this, and if so would love to know why.

    String theory perturbative scattering amplitudes are “softer” than field theory ones because they involve the exchange not of a single particle, but of an infinite tower of states with linearly rising mass. A field theory with a finite number of fields can’t reproduce this. But if you agree that the lowest energy state is determined by non-perturbative effects, shouldn’t this also be true of higher mass states? How do you know that higher mass states won’t turn into something inherently non-perturbative like black holes? But if this spectrum could be all sorts of different things, the scattering amplitudes also could be all sorts of different things. It seems to me that about all you can say is that they’ll probably somehow look different than field theory amplitudes. This isn’t much of a prediction….

    Again, maybe this argument is wrong, but I’d like to know why…

  • Pingback: Constructing Your Own Universe | Cosmic Variance()

  • Arun

    A few IMOs:

    If one can live with the landscape in string theory, if that is just the way nature is, then one can live with fine-tuning in QFTs – that too, may just be the way nature is.

    A good example to look at is a QFT that failed – SU(5) grand unification, I believe was quite pretty, predicted a rate for proton decay that wasn’t exactly experimentally accessible; but the experiments were done, no proton decay was found, and the theory was discarded. What is so perturbing about string theory is that I can’t see any way in which we can ever come to closure that it is wrong.

    I don’t know enough of the history of physics, but I think the trend towards finding ingenious ways of retaining ideas that theorists loved but were not evident in nature started with supersymmetry. After that we’ve had a piling on of superstructure. Who knows, it may not come crashing down.

  • Moshe Rozali


    I am not sure I really appreciate the problem, but let me try. String theory is calculable when the string coupling is small, and the volumes associated with compactification are large in string units. All recent models of flux compactification are of that category. When discussing the vacuum and its low energy excitations one is faced with the problem that all measurable quantities are extremely tiny compared to the fundamental scale (and furthermore they are model dependent). Therefore one is interested in tiny contributions such as non-perturbative effects (e.g to lift the moduli). This is very different from the vacuum structure in QCD where the coupling is large and non-perturbative effects are dominant.

    In this scenario, if we had a luxury of collisions with stringy center of mass energy (or as Sean points out, sufficient accuracy), perturbative string theory has a definite qualitative feature which is universal- scattering will fall off exponentially with the appropriate kinematical invariant, rather than power law. This is of course only an approximation- the stringy tower will be truncated at high enough energies, and will consistes of narrow resonances rather than stable particles. However, these (perturbative and non-perturbative) corrections will be small. In this (really hypothetical) scenario one can falsify the whole class of perturbative string vacua all at once. In the real world one has to work harder.



  • Bob McNees

    “Bob has done a good job of expressing the perspective of a real string theorist,”

    Actually, I just stayed at a Holiday Inn Express last night.

  • Peter Woit

    Hi Moshe,

    I understand that there is a regime of small coupling and large volume where you expect to be able to calculate reliably and thanks for the explanation, I see your point about the vacuum choice involving small energies. But given that you don’t know what the non-perturbative theory is, can you be sure that non-perturbative contributions are under control?

    In any case, I guess I still object to Sean’s claim of having a string theory prediction. It seems to me that if you only have a reliable calculational method for this very special regime, you can’t claim that the results you get from such calculations are predictions about the real world, where the string coupling or compactification volume may be such that non-perturbative effects are not small.

  • Jacques Distler

    “But given that you don’t know what the non-perturbative theory is, can you be sure that non-perturbative contributions are under control?”

    Because, just as in field theory, the strength of the leading nonperturbative effects can be estimated from the growth of large-orders in perturbation theory.

    In weakly-coupled field theory the h-loop amplitudes grow like g^{2h} h!, leading to the familiar (instantonic) e^{-1/g^2} effects.

    In string theory, the h-loop amplitudes grow like g^{2h} (2h)!, leading to stronger nonperturbative effects, of order e^{-1/g} . This is stronger than in field theory, but still controllably-weak at weak coupling.

    We even have a good (but not great) understanding of what those leading nonperturbative effects are in may weakly-coupled string theory backgrounds.

    There is, by now, a rather large literature on the subject…

  • John Chunko

    Hi Scott,

    So, I’ll throw my 2 cents in on the matters you posed above…

    >So I have a question about the claim that ST predicts gravity that “you just say the word “strings,” and gravity leaps out at you whether you like it or not.” Does it simply predict the existance of a 2 spin, massless, chargless, ect particle or does it also predict that all particles emit and absorb this graviton based on their energy and momentum as expected?Also does the existance of a particle that is also predicted by all the failed quantizations of gravity really a good thing for a theory to predict?

  • John Chunko

    Well, that’s a bit odd – the reply engine ate the majority of my post. Maybe bowing to the East three times will appease the computer gremlins… Okay, that done, let’s try this one more time…

    Hi Scott,

    So I’ll throw my 2 cents in on the matters you posed…

    “So I have a question about the claim that ST predicts gravity that “you just say the word “strings,” and gravity leaps out at you whether you like it or not.” Does it simply predict the existance of a 2 spin, massless, chargless, ect particle or does it also predict that all particles emit and absorb this graviton based on their energy and momentum as expected?”

    Simple bosonic strings can be described by a field theory defined on the string worldsheet where the action is dependent upon the metric h_{mn} on the worldsheet, the 2-curvature R_{2} of the worldsheet, and a dilaton phi (a scalar field on the worldsheet). If we demand that the worldsheet field theory is conformally invariant (scale invariant), then the so-called beta functions must vanish (this is necessary for us to gauge-fix the metric to the conformal gauge). This only happens in 26 dimensions for bosonic strings. Incorporating supersymmetry results in a dimensionality of 10 for the beta functions to vanish. Now, if you look at the spectrum of states for this bosonic string you’ll discover that one of the states allowed for a closed string is characterized by zero mass and spin 2 – one is very tempted to equate this state with the graviton. Also, open and closed strings can absorb and emit these spin 2 states – this implies that these states can behave like gravitons – and the way these spin 2 closed strings propagate is essentially identical with the way small gravitational distrubances propagate in GR.

    Now, if our theory seems to be predicting graviton states on a flat background spacetime, one would naturally ask what happens if the background spacetime is curved, since gravity in a theory implies a potentially curved space. If the background is curved, then it has a metric which enters the field theory on the worldsheet as a set of couplings between the string coordinates. When we demand our beta functions to vanish in this case, we discover that one of the requirements for this to occur is that the Ricci curvature of the background metric and the second derivatives of phi vanish, or R_{ab} + 2D_{a]D_{b}phi=0. This is Einstein’s equation for a spacetime with scalar field. So, the simplest version of string theory has Einstein’s equation jumping out at you almost immediately, yielding more evidence that gravity is hard-wired into the guts of string theory.

    Now, string theory also makes a prediction on how Einstein’s equation behaves when one gets close to the string scale. Above, a certain step was hidden when we went from beta functions vanishing to Ricci curvature popping out. This step was the expansion of the string coordinates in a perturbation series in the string scale. The lowest-order terms in the string scale expansion give rise to the Ricci curvature and scalar derivatives when the beta functions vanish, and leads us to Einstein’s equation. If we include the higher-order terms we find that Einstein’s equation picks up contributions of order alpha(R^(2n)) where alpha is the string scale and n=0,1,2,.. This description is only really valid in perturbative string theory when the spacetime curvature is small in relation to the string scale.

    Modified versions of gravity similar to this have been investigated for many reasons, one of those reasons being an attempt to explain the acceleration of cosmic expansion without appealing to dark energy or its cousins (see Mark and Sean’s papers atro-ph/0306438 or astro-ph/0410031). While the actions they use do not result in field equations of the form given above for the stringy corrections to Einstein’s equation, their results do illustrate that even small corrections to standard gravity can have pronounced effects.

    So, in short, string theory (both perturbative and non-perturbative) predicts Einstein gravity to lowest order, it predicts a particle state with all the properties and behaviours of the graviton, it predicts this graviton interacting with other particle states as well as with itself, and it predicts higer-order corrections to Einstein gravity that could be experimentally verified or rejected.

    “Also does the existance of a particle that is also predicted by all the failed quantizations of gravity really a good thing for a theory to predict?”

    As for whether or not the prediction of a particle predicted by other “failed” theories of quantum gravity is a good thing I’d have to say yes. From our past experiences with the Standard Model and quantum field theory we have very good reason to believe that gravity is carried by a massless spin-2 boson. The fact that string theory automatically predicts such a particle state is a Good Thing for any theory that strives to be a quantum theory of gravity. Also, no nails have been banged into the coffins of the other contenders for a quantum gravity theory, such as loop quantum gravity (LQG), dynamical triangulations, causal sets, twistors, etc.. These other theories are still young, just like string theory, and have their own problems, just like string theory. Some of the concepts from these other theories (holography for example) have cross-polinated with those of string theory, to the betterment of all the theories involved. Granted, theories like LQG don’t claim to be Theories of Everything. LQG began with the modest proposal of quantizing Einstein gravity with the fewest assumptions and guiding principles and has resulted in a theory that has made predictions that have been confirmed by string theory, such as black hole entropy. One of the challenges for LQG (and the other TOE contenders) is to work in matter and the Standard Model gauge groups better, as well as fix the dynamics of the theory via a working Hamiltonian constraint so that one can start making predictions as to how quantum gravity affects Standard Model physics, if at all.

  • agm

    It’s like meeting someone, the first impression never really dies, though it may eventually be overcome. The thought that a physicist could ever think that the mathematics of a theory is so beautiful that we don’t need to test it, it must be right… C’mon, we’re engaged in the project of describing the universe, not someone’s mathematico-aesthetic senses. Pretty much this claim is what permanently destroyed the street cred of string theory for the non-string theorists grinding an axe, and since the string PR machine hasn’t stopped, this caraciture hasn’t had a chance to go away. If someone could please pull Brian Greene off the lecture circuit, that right there will calm things down a lot. If you can’t tell me a way that we can someday test some version of string theory, which I can then turn around and explain to someone who is badgering me because that person has heard that I’m in physics so “What do you think of string theory?”, then you’re full of crap. It’s really as simple as that.

  • Haelfix

    I think SuSy would have been discovered without String theory. It seems to me its a rather logical refinement and generalization of the algebra used in field theory, as well as one of the few ways to escape the tight leash the Coleman Mandula theorem stuck phenomonlogists in. I have no idea if its ultimately relevant, but it is a relatively simple and elegant idea that im sure a smart phenomenologist would have eventually contrived.

    No one doubts ST’s usefulness in studying Gauge theories, or that it has produced very interesting spinoffs. For that reason alone it should always remain an active research program.

    But as a full theory of gravity, well, I don’t know anymore (and I would love if all those great claims came true one day). Its certainly worthy of having people work in it, but I partially see Peters point that it doesn’t deserve to have such a complete and utter stranglehold on the industry.

    The landscape stuff, if it remains as is, pretty much kills in my opinion the greatest purpoted benefit the theory offered, namely ‘no continous adjustable free parameters’.

    Thats precisely the reason why so few study qfts in curved spacetime anymore. The problem ultimately was intractable and unpredictive, at least in so far as what we currently understand with the available tools. Yet who among us doubts that such a thing exists?

  • Thomas Larsson

    John Chunko,

    Why do you demand that the worldsheet field theory is conformally invariant on the quantum level? It is not necessary for consistency, see the first line in section 2.4 of GSW:
    “Classical free string theory can be consistently formulated for any spacetime dimension, but quantization with a ghost-free spectrum requires D &lt= 26.”
    So the no-ghost theorem rules out D &gt 26, but not D &lt 26 (for the free string).

    In the presence of a conformal anomaly, the conformal symmetry is elevated from a gauge symmetry to a conventional symmetry quantum mechanically. The same thing almost certainly happens to general covariance in 4D gravity, see e.g. Roman Jackiw’s gr-qc/9511048, p 20:
    “What does any of this teach us about the physical four-dimensional model? We believe that an extension in the constraint algebra will arise for all physical, propagating degrees of freedom: for matter fields, as is seen already in two dimensions, and also for gravity fields, which in four dimensions (unlike in two) carry physical energy.”

  • Peter Woit

    Hi Jacques,
    Thanks for the argument, although before I completely buy it I’d have to look up exactly what sort of assumptions on the behavior of the full amplitudes are implicit in it.

    But this question about bounds you can get in the limit of small coupling doesn’t really have anything to do with my original objection to Sean’s “prediction”. If the real world is governed by a string theory, you have no particular reason to believe that the coupling is arbitrarily small. It’s some finite number, and for that finite number, non-perturbative effects may dominate. By “predicting” that the world is in a region of small enough coupling to be able to control non-perturbative effects aren’t you like the guy searching for his keys in the dark who “predicts” that they’ll be under the lamppost. Do you think it is legitimate to call this a “prediction” of string theory?

    Hi John,

    You say string theory predicts higher-order terms that can be experimentally verified or rejected. You seem to be claiming that one can falsify the theory this way. What exactly is the prediction for a higher order term, such that one can do a specific experiment and see if it is there? Don’t these higher order terms depend on the details of your compactification, etc., so that it practice you don’t have a specific prediction, but could get pretty much any result you want?

  • Pingback: Friday Random Ten: iPod Tells the Future of String Theory? | Cosmic Variance()

  • Thomas Larsson

    Let me clarify why I think that the subcritical free string is so important, and why it gives a very strong reason why both string theory and LQG must be wrong, at least as theories of quantum gravity. As is well known, the free string is 2D gravity coupled to other fields, and as such it is natural to try to learn about quantization of general-covariant theories from it. The most striking thing is what happens to the constraint algebra:

    * In 1D gravity (point particle), there are no gauge anomalies.
    * In 2D gravity (free string), there are some anomalies. They can be shifted between the Weyl and diffeomorphism sectors, but they cannot be removed.
    * In 4D gravity things become even more complicated, and one strongly expects there to be many anomalies. This is how I read Jackiw’s paper.

    Now, when you quantize the free, subcritical string, you automatically construct a non-trivial, unitary rep of the Virasoro algebra. Conversely, if you for some reason could not construct such a rep, you would be unable to quantize the subcritical string. I think everyone can agree about that.

    The situation in 4D gravity is completely analogous. Successful quantization will lead to diffeomorphism anomalies. If you cannot construct non-trivial, anomalous reps of the diffeomorphism group, you cannot construct the correct Hilbert space of quantum gravity, because it must carry such a rep. No such reps appear neither in string theory nor in LQG, and hence these are not the correct theories of quantum gravity.

    What about the theorem which states that there are no pure gravitational anomalies in 4D? Such anomalies cannot arise if you quantize the fields alone; one must also add an explicit representation of the process of observation, and quantize the observer’s trajectory in spacetime along with the fields. This is mandatory, because the relevant anomalies (“multi-dimensional Virasoro algebra”) are functionals of this trajectory. If you don’t introduce the observer’s trajectory, you cannot write down the relevant anomalies, and you are in the same position as people who try to quantize the subcritical string without conformal anomalies.

    This is the technical reason why I don’t believe in string theory.

  • Jacques Distler

    “But this question about bounds you can get in the limit of small coupling doesn’t really have anything to do with my original objection to Sean’s “prediction”. If the real world is governed by a string theory, you have no particular reason to believe that the coupling is arbitrarily small. It’s some finite number, and for that finite number, non-perturbative effects may dominate.”

    The coupling doesn’t have to be arbitrarily small for e^{-1/g} effects to be numerically utterly negligible compared to perturbative g^{2h} effects. Please entertain yourself by plugging in some sample values for g which are “small, but finite”.

    There are, however, two things to be said about this.

    1) Even if g is small, there are some effects to which the perturbative contributions vanish and the dominant contribution goes like e^{-1/g}. You can see discussions of some of those effects (and how they affect the vacuum structure of the theory) up-thread.

    2) There are reasons to expect that, depending on what class of string backgrounds you are considering, the string coupling may not be small if that background is to describe the real world. There is, roughly-speaking, a map from the fundamental parameters (the string scale, the string coupling, and the compactification scale) to the observable 4D parameters (the 4D Planck mass, the GUT scale and the GUT gauge coupling).

    Some view that as a problem. I view it as an opportunity.

  • Juan R.

    Dear Sean, the problem with current unpleasant status of string theory into the community was built by own string theory community. Let me first remember to you some basic points —the list, of course, is not exhaustive—, which will help to you to rewrite your “cheers”.

    – String theory did born like a failure to explain strong force and since it has been always a complete failure. Nothing predicted and all past claims shown to be false. String theory is a theory without laws or postulates because they are modified with time. Please, let me remember to you the history of dimensions: 4D -> 5D -> 26D -> 10D -> 11D -> 12D (some people is working in more than a time dimension) -> 4D (Segal has claimed that we may find a 4D version for solving compactification problems), etc. The claim of string theory is “open” is, of course, a complete nonsense when claim is properly interpreted on both epistemological and ontological terms.

    – It is well known that string theorists have manipulated public opinion about string theory. In fact, no popular string theory writer has still convincingly explained to public that string theory failed like a TOE since is being substituted by still unknown M-theory. The popular dissemination of string theory to non-experts violates the most basic ethic guidelines.

    – The arrogant attitude of many string theorists is also very well known. Please talk with some critics of them like Peter Woit or Glashow and learn the true sense of the word “pressure”.

    – String theory is a mathematical goulash and a dishonest copy of formalisms developed by others. For example, after of decades of very wrong claims about the supposed TOE, now string theorist recognized that were wrong since usual quantization of string was not exact. Now they are launching the TFD version of string theory BUT TFD was previously developed outside of string theory. Again, non-string theorists were correct and “smart” string theorists (e.g. Witen Greene, Vafa, Schwartz, etc.) completely wrong. In fact, none string theorist did contributions to TFD. Even some string theorist has recently recognized that string theorists usually copy the work of others and after “rename” it like string theory.

    – All past claims by string theorists were shown to be false, absolute all! In fact, the popular idea of that pointilike particles would be substituted by one-dimensional strings has been superseded by recent M(atrix) formulation by Banks and others, which is basically a quantum mechanics of pointlike particles (D0-branes). People again are ignorant of that, since that, like openly admitted by some theorists, the old name “string theory” is maintained by marketing purposes.

    – It is also well knonw and denunciated in several occasions that string theorists have ignored other approaches to quantum gravity. It is very hard for a loop theorist to hear in a popular talk —given by a string theorist— that string theory is the only approach to quantum gravity. The only game into the city!

    – It is also well known that young researches were forced to research into string theory because financial support of other theories was stopped in departments, funding agencies, and others due to aggressive string marketing activities. Many young physicists begin a PhD on string theory, discovered that string theory was a waste of time (real string theory is not the same that popularised version of string theory), and leaved the field. Some of them feel…

    – Let me take a simple example from chemistry. According to “ignorant” and very arrogant people like Ed Witten, string theory is a promising TOE and reduces all of others sciences, e.g. chemistry. A moment, chemists know that is false, the reduction of chemistry to physics is a myth, as brilliantly explained in innumerable papers in Foundations of chemistry and others journals. Interestingly, 30 years ago some chemists were working in advanced formalisms for explaining behaviour that cannot be studied with usual methods. If you compare the very advanced theoretical work developed in the 60 and 70s with corresponding string theory status you found that string theory was wrong like a TOE even a joke. String theorists, arrogant as they are, ignored all of that and claimed that all of chemistry was an application of string theory. String theory was so advanced that no one other theory could provide to us an explanation of nature more profound, they said. Of course, chemists smiled, like they smile in the 20th century, when physicists (including Nobel laureates like Stark) attempt to convince to them that chemical bond was modelled by classical electrodynamics and that Lewis bond theory was, in simple words, nonsense. Now in the last part of 90, some string theorist discover that all past claims were wrong and are developing a new version of string theory called non-critical one. It is interesting that all past quantum methods and basic stuff is abandoned whereas work developed in the 60s by chemist Ilya Prigogine (see for example his Nobel lecture) used for a radical generalization of old string theory. But the ideas used NOW in string theory were developed in the 60s by other people! Prigogine and others were correct, string theorists again wrong. Interestingly, the ideas of the 60s have been updated in the 90s by the Prigogine and co-workers. Therefore, the current “radical” generalization of string theory by string theorists is, again, an outdated theory. This is real status of string theory; an authentic revolution if one read that masterful piece of marketing called the Elegant Universe (by Brian Greene) and focused to laymen, but claimed to be “very conservative” and outdated in the recent conference Quantum future by expertises that know stuff. Said I again once more? The first step for any serious theorists is to read previously published literature and then develop a better theory, but crackpots are specialist in ignoring the scientific method. If string theorists continue to develop a really outdated theory at one hand and arrogantly claim that are doing (they believe that in their infinite ignorance) the most important, the most powerful, the most fundamental theory at the other, then they would feel comfortable with the mocking of their colleagues. If Brian Greene, offensively claim in his talks that we may quantize everything, and Dyson convincingly reply him saying that Greene is providing no solid arguments in his belief, the problem is not with Dyson, the problem is with Brian Greene, that would first study serious stuff before doing irrelevant claims surrounded by a halo of pomposity.

    – Etc.

    Sincerely, I believe that non-string theories have been very generous with string theory community. String theorist would please to us our kindly attitude.

    Once “refreshed” your memory, let me now comment some of your points.

    The idea of that string theory is the most promising way to reconcile gravity and quantum mechanics, is one of well-established myths of literature. In an absolute sense, loop quantum gravity is so “successful” like the strings but in a relative sense (successes / total number of researchers), the loop approach has been around 10 times more satisfactory. Please, let me remember to you again that string theory has been substituted by M-theory.

    It is false that “string theory” is based in one-dimensional loops. In fact, you appear to unknown the current joke on Internet that say that “string theory is now a theory without strings”. Yes, you obtain a remarkably rich structure, but just at mathematical level.

    It is false that string theory predicts or explains gravity (this is another myth). “To predict” a massless spin-two particle is not the same that quantum gravity. In fact, causality is defined on a flat fixed metric with graviton modes arising in the perturbation, which violate GR basic idea of that full causality is defined in the full metric. This has been the main criticism of general relativists and loop theoreticians during decades. Now, string theorists are recognizing that great mistake (in the past they claimed that one would not take GR “too seriously”) and are unsatisfactorily searching for a background independent version of the old (outdated) string theory.

    “In string theory, you just say the word “strings,” and gravity leaps out at you whether you like it or not.” This is not true, in fact one use previous ideas from GR, like to leave “freedom” to the metric into the string action. Somewhat like we need know previously that universe looks 4D and then introduce an arbitrary (that is by hand) compactification 10D -> 4D x 6D.
    “At this point it’s a little unclear what the fundamental building blocks of string theory are” It is clear that string is NOT the fundamental entity into the non-perturbative regime.

    “One often hears that string theory simply makes no predictions, but that’s clearly false. If you scatter two particles together, string theory unambiguously predicts that the cross-section should look stringy, not like that of fundamental point particles.”

    Humm, even ignoring that prediction really mean, this is another myth. Let me simply quote to D. Friedan:
    “Even if some particular macroscopic background spacetime is chosen arbitrarily, by hand or by ‘initial conditions,’ string theory still fails to be realistic at large distance. The large distance limit of string theory consists of the perturbative scattering amplitudes of the low energy string modes, which are particle-like. But the particle masses are exactly zero, and the low energy scattering amplitudes are exactly supersymmetric. String theory fails to provide any mechanism to generate the very small nonzero masses that are observed in nature, or to remove the exact spacetime supersymmetry, which is not observed in nature. More broadly, string theory is incapable of generating the variety of large characteristic spacetime distances seen in the real world. At best, for each macroscopic background spacetime in the manifold of possibilities, string theory gives large distance scattering amplitudes that form a caricature of the scattering amplitudes of the standard model of particle physics.”
    The problem, of course, is not the popular statement of the difficulties for testing Planck-scale physics as you incorrectly argue; the problem is that nobody has obtained the successful standard model from string theory. A first step of any new theory is obtain that is already known before predict any new (including Planck-scale physics). So string theory is not compatible with available experimental data. “it’s just that we are as yet unable to test them.” As explained above, this is wrong.

    “Recently there has arisen another sense in which string theory purportedly makes no predictions, associated with the ‘landscape’ of possible string vacuum states.”

    “Well, too bad. It would have been great to make such predictions, but the inability to do so doesn’t render string theory non-scientific.”

    No comment!

    “… and possesses a mathematical beauty that is so compelling that the theory simply must be correct.”

    The world is as it is, no that we like we want that it was. Mathematical beauty is a guide newer a justification. Moreover, string theory is rather ugly, at least for me. About his supposed “beauty”, sceptics suggest that string theorists try to colourfully camouflage the well-known theory’s flaws, like “a 50-year-old woman wearing way too much lipstick”.

    “These kinds of arguments just don’t carry that much weight with the non-converted.”

    Yes, I believe that “converted” is the correct word to use here. String community looks like a sect or, in the words of one of its most famous members, “a kind of a church”. It is not science.

    “it’s the most promising way we know to quantize gravity. If there were multiple very successful ways to quantize gravity, it would be important to distinguish between them experimentally; but so long as the number of successful models is less than or equal to one, it makes perfect sense to make every effort to understand that model.”

    This marvellous piece of promotion just emphasize the myth of string theory is the only game in the city. Please read literature in semi phenomenological approaches to quantum gravity and recent advances in other theories like LQG and predictions for the future LHC.

    Yes, the comparison between Microsoft and Apple Linux is correct!! Microsoft is a layman-oriented business, whereas Apple or Linux are more specific but more serious. Moreover, it is well knonw that windows OS is a copy of graphical Apple OS and certain kernel properties of Linux/Unix. The success of Windows is in marketing and layman orientation, somewhat like string theory. It is interesting remark that when Microsoft presented the revolution of the trash icon, graphical copy and paste, multitasks, and others features in his first versions of windows, all of that was already known for decades for Mac users. Remember the famous Windows blue display. Yes, your comparison is really good!!

    “It didn’t have to work out that the entropy of a black hole calculated from semiclassical gravity ala Hawking would be equal to the entropy of a corresponding gas of strings and branes, but it is.”

    Another myth!! Loop quantum gravity obtains the entropy of Schwarzschild black holes. In “string” (really brane) theory, one traditionally has worked with BPS and idealized models of black holes. Strictly speaking, the “traditional” results in string theory do not concern, precisely, black holes, as they are found in a limit in which the gravitational constant is turned off. But they concern systems with the same quantum numbers as certain black holes. There is a kind of analogy instead of an identity with GR black holes.

  • Vish Subramanian

    The problem with string theory is stated by Philip Anderson:

    “string theory is the first science in hundreds of years to be pursued in pre-Baconian fashion, without any adequate experimental guidance. It proposes that Nature is the way we would like it to be rather than the way we see it to be; and it is improbable that Nature thinks the same way we do.”

    The results of string theory – impressive as they are, are not close to dispelling the scepticism.

  • Moshe Rozali


    I think the analogy with QFT is good to reply to your point- QFT does not make general predictions (or rather it makes very mild ones), you have to choose a model to make predictions. In a certain class of string models there is a robust feature which can falsify them, albeit in a very science-fictional way (no such model-independent feature exists at accessible energies, the double edge sword of universality…). This feature does not exist in other corners (e.g models based on 11dim SUGRA), and for many corners we don’t know. To be precise I would call this feature a predicition of perturbative string theory. I agree with you that one ought to use precise language, though this is a non-technical forum.

    Sure that assuming we are in the calculable regime has no apriori justification. At any point in time one has to choose between looking under the lamppost (where else would you look?), or developing the theory further. This is a personal choice, and it is good to have people doing both. The latter possibility is intimately related to further developing the language of QFT and quantum gravity (not sure there is a clear distinction), or as WL phrased it above, figuring out how things work.



  • Levi

    A short question. If, as rumored, experiments show that gravity weakens at small distances, would this be a serious blow to string theory?

  • Sean

    I think I finally understand Peter’s point about the non-perturbative regime, and it’s a fair criticism of my claim that string theory makes unambiguous predictions — it’s only true at weak coupling. (Of course, I’m not the one to be talking about these things, but I understand that the blogosphere is self-correcting.) In the uncompactified regime, the low-energy limit seems to contain five critical string theories at weak coupling, and at strong coupling looks like eleven-dimensional supergravity. It’s possible that our four-dimensional world is best described as a compatification of 11d supergravity rather than one of the string theories. In which case, I think it’s fair to say that you wouldn’t be able to see unambiguously “stringy” effects in perturbation theory. Of course, you might see them, and claim such an observation as evidence for weakly-coupled string theory, but the general framework becomes harder to falsify.

    (You might wonder whether the same problem would exist for strongly-coupled field theories, where we’re not very good at calculating things. But strongly-coupled field theories tend to rearrange themselves into weakly-coupled field theories with a different set of low-energy degrees of freedom — e.g., QCD becomes a theory of pions. When strongly-coupled string theory rearranges itself into supergravity, the essential stringiness is obscured.)

  • Chad Orzel

    I wrote a long set of comments on this issue over in my own blog, because I thought it would be rude to post a comment of that length. Given some of the things that have been posted since, I may have been wrong, but consider this a manual TrackBack:

  • Chad Orzel

    (And, of course, the long URL got broken by the comment system… I hate computers.)

  • AS

    I disagree with the sentence “So string theory makes predictions, it’s just that we are as yet unable to test them”.

    Quantum gravity effects are expected to be observable in the context of large extra dimensions, so in the past years many physicists studied their experimental signatures. Basically all results have been obtained from Einstein general relativity, while unfortunately nobody could tell what a “stringy” prediction really is.

    (I think this can be considered as a fact. To be precise computations of stringy cross sections have been done in string models that at low energy have contain super-symmetries and no protons. Since LHC will collide non-supersymmetric protons, this kind of string results is far from relevant).

    This lack of concrete results in the scenario where having a theory of quantum gravity would have been of practical relevance is probably another factor that contributed to deteriorate the way string theorists are perceived by high-energy physicists.

  • Clifford

    Sean, actually I’m not sure I agree with your final comment above. I think that 11 dimensional supergravity might be a nice *positive* example of the entire non-perturbative completion changing the theory from perturbative strings to something that does not look stringy. And it is not so different at the core from things that can happen in field theory properties you mentioned. It is positive because supergravity is still a recognisable way that the (*low energy*) strongly coupled physics has organised itself. So we have not lost control and all wierd hell has broken loose, as is the concern that Peter Woit was expressing about strong coupling.

    Note that we do know of examples where strongly coupled field theories actually don’t just reaarange themselves into familiar-looking field theories. Take the example of several five dimensional theories that actually become six-dimensional at strong coupling, and for which we cannot write a proper Lagrangian because the natural degrees of freedom couple to a two-form potential. The theories are just wierd at first sight, and aren’t in your typical textbook, but we come to grips with them eventually and see that they still might be under control.

    The moral of the story I think that that Peter is right that we should worry about what the final shape of the theory and dynamics are once we’ve allowed for fully non-perturbative physics to take over, but that is not reason to abandon the program altogether because it might not look too much like it does perturbatively. I need only point to QCD as a good example. The most common phase in which we find it is a strongly coupled mess where the basic degrees of freedom in terms of which we formulate the theory (quarks and gluons) are not at all apparent. Someone might say that it is *not* a good example becasue we can get access to at least some of the perturbative regime by doing high energy scattering. Good, and in return I would say that perhaps when we understand strings a lot better, we will find that there are regimes where the strongly coupled -maybe the most common- phase is not recognisably string theory and that there are no stringy features left to observe under most circumstances, but perhaps there will be extreme processes analogous to high energy scattering in QCD that we will conceive of (perhaps in early universe cosmology -goodness knows?) which do probe the weakly coupled regimes (perhaps via duality) where we see the “stringiness” directly in the physics. I just made that up, but I don’t see it as unlikely in principle.

    I just think that it is *way* too early to say one way or the other. My feeling is that the theory is still in its infancy. It is unfortunate that the Stringevangelists have promised so much so soon. But their exuberance was/is understandable, if somewhat problematic for having produced a bit of a backlash. I still am excited about string theory and its related theories and endeavours. It is undeniably rich and promising. (And it is useful for other things besides just figuring out “theories of everything”.) We just should make sure to qualify our enthusiasm with caution about how long it might take to turn the framework into a predictive theory -or familiy of theories- that can then be confronted with Nature. Also, while I have a gut feeling that it is premature to look for stringiness in Nature in so much detail that will rule in or out the whole enterprise, it does not hurt to keep a dialogue going between string theory and phenomnology in case we get lucky. There’s a lot of intersting effort out there should *should* continue. Maybe some of the broad principles underlying a possible *class* of things that string theory is part of might be already there for the harvesting.



  • Moshe Rozali


    Some strongly coupled field theories rearrange themselves as string theories, and vice versa. It is therefore just inconsistent to talk about string theory and QFT as separate frameworks. Falsify the one and you falsified the other.

    Rather, the stringy language seems like a more efficient way (perhaps not the final story) to organize the beast, at least sometimes. As a framework and a language, string theory is not some future promise, it is already a very successful enterprise.

    But I’ll stop here, lest I find myself over-hyping…



  • Pingback: Musings()

  • Eli Shkrob

    It is all very well but where is the promised answer to “What is the gravitational field of an electron?”

  • Peter Woit

    Very interesting comments, I think this blog is setting new standards for the level of discussion. All hail!

    After beating up on Sean about predictivity, I should point out that I agree with Clifford and others that coming up with real predictions that go beyond the standard model is probably too much to hope for right now. Another way to say it is that it’s pretty clear our tools now aren’t good enough, so we need better tools before we can build something better than the standard model. I’m all in favor of string theorists working on building better tools involving string theory, but feel the field needs to encourage people to also build new tools that don’t involve string theory.

    I’m tempted to argue with Moshe about his claims that QFT and string theory are all the same thing, but maybe I should just agree with him. Maybe space will open up in theory groups for young theorists who want to do pure QFT once everybody agrees that QFT=string theory.

  • Pingback: Ars Mathematica » Blog Archive » Two Cheers for String Theory()

  • Clifford

    Oh, Peter, I should say one thing on that point. Don’t forget that “QFT=StringTheory” is not another of those later annexations that our field organised in its often-claimed “assimilation” of all of physics. (We’ll get around to the rest of science eventually, he says provocatively.) At least for the case of gauge theories, (and in a superficially different sence from what Moshe and I were talking about above) this was something taught to us by ‘t Hooft in the early 70’s, right? And this is an example of what I was saying about strings being good for other things. It is a window into the nature of Quantum Field Theory, which we *definitely know* is a powerful tool for describing Nature. I’m looking forward to the day when we get really powerful insights or tools from string theory into calculating in useful dynamical regimes of field theory that we probe with real experiments. I would consider 30 or 40 years of effort in string theory worthwhile if we acheived just that alone.

    Yes, there should be jobs for young people working on hard problems in field theory. Some of those should have powerful string theory technology in their toolboxes so as to serve their employers well.


  • Dan Piponi (SIGFPE)

    I also think that whatever happens, we need to be able to do String Theory. String Theory isn’t really a new theory – it’s just quantum mechanics (and a bit of supersymmetry). It seems to me that we really need to understand the ramifications of that theory before we can move on. The richness of String Theory is stunning giving its simple origins from quantising the classical string.

    The idea of giving up that richness to study another area before we’ve fully explored the consequences of Strings seems insane to me. Before the String theory programme really got going nobody could have predicted the connections to fields like classical elliptic functions and modular forms, group theory via Monstrous Moonshine, the unification of the 5 string theories, mirror symmetries and so on (I’m a few years out of date myself). Vast amounts of fascinating mathematics have been pouring our from the work of String theorists. And yet these are all just consequences of quantizing a simple classical system with no extra hypotheses, no extra laws of physics. Until we understand these it can hardly be said that we understand quantum mechanics, and yet QM is the foundation of all modern theoretical physics. If we’re stil surprised by the consequences of quantising this simple little toy 1D system I don’t think we can’t be ready to move on at all.

  • Frederik Denef


    You made some excellent points, and I think constructive criticism is much needed in string theory, or in any other science for that matter. You’re doing a great job in that sense, and I certainly agree with your general point that one should avoid at all costs that a single hypothetical theory starts monopolizing all activity in a field.

    However, I (of course 😉 don’t quite understand why you express so much undifferentiated disdain towards the “landscape studies” sector of our field. The working hypothesis for about anyone working in string theory is still that at least in principle it is a framework capable of describing the real world in a way compatible with all known principles of nature. How concrete this will ever become, or how useful insights from string theory (or any other theory) in predicting parameters of nature will turn out to be is still not known at this point. We certainly by far don’t know enough yet about the structure and dynamics of the theory to draw any definitive conclusions at this point.

    Getting more insight in this obviously very important question is what has motivated many of us “landscape loonies” to start studying these things in a much more systematic way than people did before. The goal of this work, at least for me and the people I worked with, is not to try to keep on “pushing” string theory as the theory that will ultimately explain everything, but rather to find out, with what we know now, up to what extent one could actually expect string theory to be predictive in this setting. This is exactly the question you have been interested in for many years now, and if anything thus far, landscape studies have put a number of your objections on a firm footing — so you should be encouraging us rather then tell us to stop doing this stuff; we’re on your side! 😉 I personally would be quite happy if we could actually prove string theory has zero predictivity no matter how accurate measurements become, or if we could prove the other extreme, that no vacuum of string theory reproduces our universe. Either way, that would be progress.

    But I want to stress that all this is still in its infancy, and that it is really premature to conclude, as you have suggested, that all hope is lost, and that therefore this line of research must be abandoned. Sure, the number of vacua is infinite, but we knew that all along — for example AdS_5 x S^5 with flux N gives you an infinite set of nonperturbatively defined vacua of string theory, since N can take any integral value. What we do not know at all is how many vacua resemble our own, and in particular not even how many lead to metastable de Sitter — from what we know now, it is still possible that there are in fact none of the latter! (There are a number of plausible constructions of metastable dS in string theory, but these are not nearly as firmly established as superymmetric AdS vacua.) Much more good work needs to be done to get a better grip on the set of vacua of string theory. I just hope young people entering the field now don’t get discouraged by the sometimes vitriolic comments towards those trying to think about the deep questions and ambitions of theoretical physics, and as a result would flee back in the safe confines of subjects that are as far from these questions as possible.

    I agree with you and others in this discussion thread that it is probably premature to expect predictions beyond the standard model from any theory at this point. But landscape research has nevertheless caused what I would call a mini-revolution in our collective thinking about string theory: after the euphoria of the mid-nineties, it has brought new (and much needed) humbleness to the field, and it clearly re-exposed the holes in our knowledge that will have to be addressed to answer many of the fundamental questions that were the original motivation for the theory. This has lead to some revived interest in quantum cosmology and vacuum selection principles, topics which before were hardly addressed at all in the context of string theory. It also has led us to rethink the notion of naturalness and the rock solid expectation this used to lead to, of finding new physics just beyond the electroweak scale. And yes, it has re-stirred the debate about the extraordinary finetuning of the observable universe and the extent to which environmental selection principles may play a role in that — needless to say (I hope!), such selection principles can only account for a very small subset of parameters, assuming other parameters fixed, but it is a logical possibility, and logical possibilities should never be excluded in any science just on the basis of aesthetic prejudices (especially if the logical possibility in question is the only one that has met with some quantitative predictive success!).

    In a way the greatest achievement thus far of this research is that the emergence of the landscape to the forefront of discussions has caused a new sense of crisis. Throughout history, crisis has proven to be necessary to be able to abandon old beliefs that obstruct progress; who wants radical new ideas if everybody is cozy with the old ones? Quantum mechanics would never have been born without the preceding period of utter confusion and crisis. So this “blow”, I honestly believe, is really the best thing that could happen to us at this point.

    Finally, about the claims that people get depressed by all this — well, I don’t get that. Science always progresses in a rollercoaster way. Ideas get killed, enthusiasm waxes and wanes, promising approaches turn obsolete so formerly obscure ones can jump in the spotlight. It’s all progress. If it makes you feel sad, it means you either were holding on to a scientifically unhealthy illusion before, or you are in dire need of a vacation to realize again that there is more to life than physics.

    Frederik (off to Mozambique :)

  • Moshe Rozali


    I also vote for 3 cheers for Sean and the rest…

    I am not sure I agree with the concise form you summarized my viewpoint. Let me just say that if you dream of a day where people study questions in field theory (say vacuum structure of SUSY gauge theories, anomalous dimensions of operators, helicity amplitudes,…), using both stringy techniques (SUSY, D-branes, topological strings etc.) and non-stringy techniques (matrix models, Bethe ansatz, twistor space, unitarity,…), you should consider yourself very lucky…

    I also don’t agree with your definition of “real” prediction. In my mind a predicition is always done in the context of a specific model. I see that Frederic has joined in, so I’ll leave the discussion of such points in his capable hands (unless he is having way too much fun in Mozambique already).



  • Pingback: The Blog as a Sharp Tool for Research | Cosmic Variance()

  • Torbjorn Larsson

    Congrats on an informative and balanced blog!

    A short comment from yet another outsider. Feasible falsiability makes us trust theories and can debunk junk. It would remove (probably all) anthropic principles which have been critizised in all their applications. The landscape seems to need more selection principles anyway so it seems like a small sacrifice; again from the outside of course.

  • Peter Woit

    Hi Frederik,

    Mozambique? wow. I’m writing from a lousy internet connection in Crescent City, California, on my way to Seattle. Much less exotic, and it’s difficult to write much this way. Do you still live in NY? If so we should get together sometime and discuss this over a few beers.

    I understand your point that in some sense we’re on the same side. What you’re doing has gone a long way towards torpedoing long-held hopes for getting predictions out of string theory. If I thought you had a chance of conclusively showing that string theory was not falsifiable, and that this would cause string theorists to agree that it was a wrong idea (as an idea about unification), and do something else, I might even work on this stuff too. If that really is the direction you are going, more power to you. But when I last heard Michael Douglas speak, he seemed completely unwilling to even consider the idea that string theory might be wrong, much less say that this was something his research might be leading to. Maybe I misinterpreted him, or maybe he is changing his mind. But I found the experience of seeing him unable to come up with any plausible hope for predicting anything, and simultaneously still promoting string theory unification to be profoundly disturbing.

    The problem is, I don’t think you can kill off string theory by studying the landscape, and the danger is that it is so huge that studying it could take up the whole careers of an exponentially large number of theorists. If the initial ideas about getting predictions out of statistics fail, people can always claim that finding the real M-theory will save things, perhaps by giving a framework in which there is a cosmological selection principle.

    I realize that my comments about this have sometimes been vitriolic. But I really do see a significant number of string theorists reacting to the fatal problems of the theory by abandoning basic principles of science, instead of admitting failure. Ann Nelson asked why I was so tough on you landscapologists, that there were always lots of bad papers around. It’s true that there always have been and will be lots of people who write bad papers. But you and Michael Douglas are among the smartest people in the field. Seeing the best people in the business involved in a project that is being used by many people as a justification for giving up on science seems to me really worrying, thus the harsh commentary.

    Have a good trip!

  • Peter Woit


    Sure, predictions are made in the context of a specific model, and these specific models are what you falsify if the predictions fail. So, you need to have a specific model or restricted class of models in mind when you say you have a theory that makes predictions. String theory involves such a huge class of models that in and of itself it doesn’t make predictions. You can imagine that there’s a small class of these consistent with some basic observational facts, then these would make predictions. People have been working hard on trying to do this for more than twenty years with zero success. I think there’s overwhelming evidence this won’t work. Every model I know of that has so far been investigated is already falsified by the fact that the model can’t reproduce the standard model. Maybe someday finally someone will come up with a string theory model that agrees with the Standard Model, but makes other predictions. Then you can say string theory predicts something, but not until then.

  • Peter Woit


    You act as if particle theory can only investigate one thing at a time. While sociologically this sometimes seems to be true, there are several thousand particle theorists out there, and your argument implies that some should work on string theory, but there’s no reason all of them should.

  • LuboÅ¡ Motl

    Some comments of mine about the topics in this thread are here:

  • D R Lunsford

    Somewhere in the above mass of comments, the owner of this blog made some basic comments about gauge invariance without mentioning charge conservation. This seems very odd to me.


  • Lee Smolin

    Dear Sean,

    Thanks for your comments and thanks also to everyone for the high level of the discussion. Thanks also for your supportive comments about alternative approaches to quantum gravity. But I believe you miss the key point, which is that there is good reason to believe that any acceptable quantum theory of gravity must be background independent. Those of us who work on “alternatives” do so because we find the arguments for background independence compelling, whatever the approach.

    You mention that we don’t understand what string theory is. I find it compelling that any complete formulation of string theory must be background independent, and it does indeed seem to be possible to make progress in this direction. For years it has puzzled me to no end why it is not obvious to everyone in string theory-as it is obvious to those who pursue virtually every other approach to quantum gravity-that any quantum theory of gravity must be background independent. On top of this, there is an obvious reason special to string theory, which is that we know different solutions correspond to different backgrounds, so any theory they are solutions of must make sense prior to the specification of a background. The fact that, after huge investments of effort, both AdS/CFT and the matrix models have been only partly successful supports this. Very recently it occurred to me that we in quantum gravity base our conviction about background independence on arguments that may never have been spelled out in a context where string theorists could encounter them. So I’ve written a paper making the case for background independence that I’m about to post on the hep-th. Since I regard the arguments I’ve reviewed in the paper as compelling, I would be very interested to hear if some readers don’t.

    So I would insist that the reason to support alternatives, is not just because it is the ethical thing to do. The only hope for string theory itself is to go in the direction that LQG, CDT and other background independent approaches have pioneered. In fact, contrary to your implication that these directions “soon get more or less stuck”, there are important recent developments in both LQG and CDT regarding hard problems including the emergence of classical spacetime. Even given the far fewer people and resources available for work in this area, progress since 1986 has been steady and cumulative, and never has been as rapid as now.

    As for predictability, in spite of the good work being done on the landscape, it must be emphasized that it is a BIG DEAL to give up on the main claim and main goal of a theory, in this case uniqueness leading to a unique explanation of the standard model and unique predictions for future experiments. One can rationalize it, but I would hope only after serious reflection on the possibility that a theory that fails to deliver the goal for which it was pursued is just simply the wrong theory. The point of the formulation of the landscape that I gave many years ago was that a theory with many vacuum states could still be explanatory and predictive, but only under certain conditions, which I spelled out in my book and papers on the subject. Let’s be clear that the burden is on us, when we advocate a theory, to propose falsifiable and doable experiments, and to take them seriously, so we are prepared to abandon the theory if the experiments go against it. Otherwise, there is no possibility to settle the truth of the theory by rational argument from evidence, and nothing to prevent the scientific community from splitting into quasi-religious communities of true believers.

    On Lubos’s comments, there are new proposals for unification within LQG. Crane et al have made some speculations in this direction. Look for new work coming soon, about this (already announced at several conferences) by Markopoulou and myself.

    Finally, there are experiments in progress that probe the Planck scale, such as AUGER and GLAST. Does string theory make precise predictions for what these experiments will see, when they report in the not too distant future? I know of no paper that claims so, although naively one would expect string theory to predict that Poincare invariance is unmodified and unbroken. Do the experts agree that this is the prediction?

    Lee Smolin

  • Moshe Rozali


    Sorry for mispelling your name earlier (I went with the french version), feel free to mutilate mine next time we meet. Nice post above, but I have to admit I for one am lacking a sense of crisis, maybe I have my head in the sand. If we had a very large number of realistic vacua (to start, non-supersymmetric ones) I might start getting worried…

    (reminds me of a Woddy Allen line, on his death bed , to his wife, “we could have made love more often, or maybe once…”)

    Of course, no way of knowing without looking, as you say.


    My comments about predictions were in the context of soft scattering etc., it has been a while since this thread started…



  • CapitalistImperialistPig

    If you are trying to make friends and influence string theorists, Sean, two cheers is not nearly enough. Check out our friend Lubos, for example. String theory demands complete dedication – sort of like being in Stalin’s entourage. The first guy to stop clapping is the next guy on the train to Siberia.

    It’s nice that string theory “predicts” gravity, but in the old days only stuff we didn’t already know counted as predictions.

  • James Graber

    Background Independence considered harmful (in obvious homage to Edsgar Dijkstra)

    In response to Lee Smolin’s post emphasizing the fundamental importance of Background Independence, let me suggest the contradictory proposition, not directly contrary, that all empirical evidence better supports the idea that a special flat class of background is realized in Nature. That is, despite the observational confirmations of General Relativity predictions and the success of Inflationary ideas, it still appears that we live in a flat universe. The fact that quantum mechanics is so hard to formulate in curved spacetime may be telling us that we live in a universe that is necesarily exactly flat, not just approximately flat, or accidentally flat. Perhaps the universe is not background independent, but rather requires a flat background. I think current astronomical evidence supports this position, or at least does not contradict it.
    I leave it to the experts to determine what this implies for String Theory vs. Quantum Gravity, but to me it appears to favor the String Theory approach.
    Jim Graber

  • Clifford

    Nice quip, CIP, but note that it predicts not just gravity’s existence, but the fact that gravity is *quantum mechanical*, and a manner in which this can come about. This solution to one of the biggest problems of the 20th Century is one we did *not* know already.



  • Thomas Larsson

    The key reason why quantizing gravity is hard is that people don’t understand the projective representation theory of the relevant diffeomorphism symmetry. There are many analogies: to quantize spin, you need to understand the reps of SU(2), in special relativity you need Wigner’s classification of Poincare irreps, in 2D phase transitions you need the irreps of the Virasoro algebra, etc.

    This was a key motivation for my discovery of the generalization of the Virasoro algebra to several dimensions, and its lowest-energy representations. The reason why such reps had not previously been constructed is that one needs to resolve essentially the same problems as in quantum gravity; how to avoid infinities arising from ordering effects, how to single out a privileged energy direction in a diff-invariant way, etc.

    What has prevented the acceptance of this discovery is the widespread belief that since the diffeomorphism symmetry is gauge in general relativity, only the trivial rep matters. People like Lubos Motl and Jacques Distler charmingly expressed this as “A gauge symmetry is a redundancy of the description, you idiot”, or something to that effect. But of course this is not true in the presence of anomalies. What is really funny is that the simplest counterexample can be found in the most elementary chapter of the string theory bible; it is explicitly stated in chapter 2 of GSW that the free, subcritical string is consistent despite the conformal anomaly c = 26 – D. Check it out for yourself.

    The free string is of course very relevant to gravity, since it is nothing but 2D gravity coupled to scalar fields.

    I wonder if many string theorists are unaware that the no-ghost theorem explicitly allows for consistent theories with conformal anomalies.

  • Haelfix

    James, current experimental evidence implies that our universe is NOT flat at all, in fact quite the opposite, it looks DeSitter.

    As for String theory, correct me if im wrong, but I always got the impression it was said the formalism was background dependant not necessarily b/c it was that way deep down, but rather it was merely the fact that it was perturbative at this time, read an approximation around a classical saddle point. If it didn’t admit such a thing, then its game over. Other than that, it is perfectly background independant in the sense that the metric is a dynamically varied variable in the action (indeed that has to follow since it is simply Einstein’s gravity + scalar fields + corrections)

    The good thing about string theory is that upon quantization hard consistency checks seem to come out -ok- when different metrics are inserted, and the same fundamental forms of the solutions appear. In fact it might be surprising that those checks *do* come out correct at all, even when the nonperturbative regime is obscured.

    The idea I guess would be similar to such a such a field theory being gauge fixed, where lorentz symmetry is not manifestly apparent (even if its there deep down), one has to work hard to show it.

  • Shantanu

    Thanks for the interesting discussion. I am a niave experimentalist. I hope
    someone can help me answer the following two questions:

    1) What experimental or observation will invalidate string theory?

    2) What are the unsolved problems in string theory (please answer this for
    non-experts) and how will string theorists know that they have finished their
    job of making a unified theory of all forces?

  • Sean

    Lee– Thanks for commenting. The relative merits of LQG is another thing that I have non-expert opinions about, but it deserves its own discussion rather than to be subsumed by the string theory thread. My desire to support alternatives was not based on ethics, but on the desire to get the right answer, and the appreciation that string theory may not be it. I also am sympathetic to the need for background independence, but (as I mentioned in the original post) every road to quantum gravity has problems, and how we prioritize them is a matter of judgment. Hopefully we’ll find time to get to it properly soon.

  • Dan Piponi (SIGFPE)


    I don’t mean to imply that everyone should be studying String Theory. Just that even without immediate experimental tests it’s a subject worthy of in-depth study and it seems reasonable to devote a non-trivial amount of resource to it. It seems to me that whatever mathematical equipment we need to do Strings is likely to be similar to what we need to do all kinds of other things with quantum systems, and that even if no successful physical theory comes of it we’ll be in a better position to solve other problems of value.

  • Wolfgang

    I would like to repeat the question asked by Levi:

    > A short question. If, as rumored, experiments show that gravity weakens
    > at small distances, would this be a serious blow to string theory?

    Perhaps Lubos, Jacques or some other superstring expert could answer it ?

    If superstring theory is the only consistent quantum theory of gravitation,
    then the effective potential should be known at least qualitatively.

  • Robert

    Holy cow, been away for the weekend and I will need the whole week to catch up…

    Two comments: Regarding strongly coupled strings: Unfortunately, 11D sugra is not enough as this is only the low energy limit like 10D sugra is the low energy limit to string theory. Doesn’t tell you anything about high energy scattering (for two reasons: It misses the higher modes, whatever they will be and it is not renormalizable, two sides of the same model). And this is worse than in low energy QCD, where one in priciple can compute on the lattice which captures non-perturbative effects. But there is no lattice string field theory (preferably for closed superstrings 😉

    Regarding gravity getting weaker: This is bad since naive extra dimensions predict that at short scales Newton’s law goes like 1/r^(2+d) for d extra dimensions and thus gets stron faster for r->0. So, if the effect is real, it cannot be plain vanilla extra dimensions.

  • Plato

    Lee Smolin:Finally, there are experiments in progress that probe the Planck scale, such as AUGER and GLAST. Does string theory make precise predictions for what these experiments will see, when they report in the not too distant future?

    While I had notice these points also, I would like to say, such thoughts linked to my website would be one in which bulk consideration might warrent inspection to perspective views about that bulk?

    Leading to graviton production seems a inevitable consequence, and Gia and those that would test the mirrors on the moon, help to exemplify these issues would they not?

    Model consumption in string/M theory would have benefited those who saw more this way?

  • Doug

    Wow Sean,

    If tenure were to be granted based on one’s ability to generate meaningful discussion of mankind’s efforts to understand nature, you might have fared far better at the UoC! But thanks for making it possible for us amateurs to observe, as you professionals discuss the merits, and lack thereof, of the most advanced thinking on the planet.

    I can almost imagine the reaction of an extra terrestrial type, from a more advanced civilization that long ago discovered how to explain the existence and properties of radiation, matter, and energy: “You guys are making this thing way more complicated than it needs to be!” he would probably exclaim. “Holy cow, the first thing all this confusion should tell you is that you are never going to get there from here. Go back to the beginning and start over.”

    I also imagine that he would think that Smolin’s comment here is the most relevant, Sean, not the least relevant. His comment doesn’t have so much to do with LQG per se; it has more to do with the conjecture “that amounts to saying “string theory is the correct description of quantum gravity.” The relevant point is that Smolin understands that the background dependence of string theory is the issue that trumps all others. All this fuss about the landscape of possible vacuum states on the one hand, and the cheering for the “prediction” of “a [theoretical] massless two-spin particle, whose properties turn out to be that of a graviton,” on the other hand, become tempests in a teapot, when one understands that no background of space and time exists upon which it all must be based.

    ET’s admonition to go back to the beginning and start over may be impossible advice for you erudite professionals, but for all of us uncommitted researchers, it’s most compelling. A space-time background is as essential to the formulation of classical and quantum physics as a floor is to a dancer. Take it away and we might as well sit down or go home. Perhaps, the time has come to realize that we can’t dance without Fock space and, therefore, must either find a new, background-independent, quantization, as Smolin et. al. have done, that comes with its own state space, or just sit down (or, alternatively, give Larsson’s ideas a try and seek a covariant way out.)

    However, IMO, the background-independence issue is only symptomatic of the real problem with the quantum theory of spacetime. Afterall, the reason the current formulation is background-dependent is fundamentally grounded in the goal of Newton’s original research program: focus on the forces in order to describe nature in terms of a few interactions among a few particles. This goal itself assumes that the universe is best described as a container of matter, which has naturally lead to an exhaustive analysis of the container and its contents: the container of matter is filled with an aether, it’s a continuum of locations, it has 3+1 dimensions, it has 26 dimensions, it has 11 dimensions, etc. (BTW, has anybody read Chen’s 3+3 papers, “Equations of Motion with Multiple Proper Time,” (quant-ph/0501034,0505104,0506176))?

    Chris W. recently wrote that he would find it interesting to read a “thoughtful exposition on the philosophical presuppositions underlying the string theory program,” which he found unlikely to emerge any time soon. However, going back to the beginning means going back to examine the “philosophical presuppostions” underlying Newton’s program, which is even less likely to emerge at this point, but Chris points to the crux of the matter at any rate:

    “This intersects in my mind with the issue of how to evaluate and interpret quantization methods. Here again is a subject for which thoughtful philosophical discussion could be most illuminating. One might wonder why the formulation of a quantum theory should depend on “quantizing” a classical prototype at all. A key notion appears (to me) to be that in our currently accepted understanding of quantum theories, discreteness is a feature of certain solutions, and does not really reside in the fundamental assumptions shared by quantum theories in general. Indeed, quantum field theory is supposed to largely explain the origin of the discrete substructure of matter that was taken as a given* in 19th century and early 20th century physics. (* ..with notable exceptions, of course.)”

    According to what ET tells me, Chris is right on. The idea of discreteness MUST reside in our fundamental assumptions of quantum theory, but the trick is, he says, don’t try to force this assumption into a substructure of space-time in the context of quantum gravity, but be smart and recognize what is staring you in the face: space and time don’t exist as a container (background); they are simply the reciprocal aspects of motion, nothing else.

    What’s more, he tells me, you don’t need a background to define motion, you only need to understand that space progresses in the same sense that time progresses, and that the joint progression of these two are only the reciprocal aspects of what actually does exist, a universal motion. Start with this, he says, and you will find an absolute reference system of motion that will provide the background-independent state space you need.

    Honestly, if we had been in a bar and I were a woman, I would have asked him to buy me a drink, and if he had demanded to know first, if I were going to have sex with him, I would have sworn that it was Feynman. I thought to bring up the subject of harmonic oscillators, but I was just too tired. This stuff plain wears me out, you know?

  • Levi

    Well, with that, I guess the discussion is over. But 80 or so good comments is a heck of a run. It’s the best discussion of string theory I’ve seen so far on the net.

  • Wolfgang


    I agree with your statement of course:

    > naive extra dimensions predict that at short scales Newton’s law
    > goes like 1/r^(2+d) for d extra dimensions and thus gets strong
    > faster for r->0.

    So do we have a test of string theory here ?
    By the way, if M-theory/string theory is indeed a consistent quantum
    theory of gravitation, then I would expect a stronger statement than the
    naive argument. At least you should be able to state how much
    (and at which scale) the effective potential deviates from Newton.
    Again, I would expect that Jacques, Lubos or somebody else could
    answer this question …

  • Juan R.

    Levi, Wolfgang, and Robert

    About If those experim experimental showing that gravity weakens at small distances. I will say next is not rigorous but is semiintuitive argument

    Ignoring possible dependence on relative velocity, one obtains strong effective gravitational interaction to shorter distances.

    Taking like good the rule 1/r^(2+d) for d extra dimensions (some recent RS brane model introduces Yukawa like exponential correction from extra 5th dimension), we can observe that smooth behavior is obtained formally with

    d -2 for r –> 0 imply formally elimination of divergencies on (1/r^2) force strengh.

  • Juan R.

    Message truncated by use of ilegal characters!!!!!

    I repeat the basis of my reasoning

    If experimental experimental showing that gravity weakens at small distances is correct we can derive it at least formaly from.

    d negative on 1/r^(2+d) for d extra dimensions. It is interesting the chossing d = -2 for short scales (dimensionality in string M theory is fixed but is not in other approaches) because

    i) It is compatible with recent advances in triangulations quantum gravity (hep-th/0505154).

    ii) d -2 for r —> 0 imply formally elimination of divergencies on (1/r^2) force strengh since (1/r^2) —-> (1/r^0).

  • Doug

    Smolin’s new paper on background-independence is available here: The case for background independence

  • Pingback: evolgen()

  • Levi

    Since you seem to be counting, just keep in mind that you have 97, not 102…

    (I just wanted to bring this thread back to the Recent Comments line)

  • Juan Maldacena

    I agree with Fredrik that studying the Landscape is very important.
    It would be nice to find a smaller number of vacua, so that
    predictions are possible. But we would only find out if this number is large or
    small if people
    study models carefully and find more and more controlled constructions.

    A large number of vacua is
    the only explanation we have today for Lambda in string
    theory. Hopefully we will have a better one soon, but since this is
    the best we have, it should be studied and I am glad that excellent people
    are studying it.

  • Ijon Tichy

    Because I have no original thought of my own, but still wanted to contribute a comment to this post (if only to push the count closer to the magic 100), I thought the following quote by a master from the past, Emilio Segrè, might be considered relevant and not entirely worthless. I suppose you could think of it as one part of the explanation of why a quantum theory of gravity is proving such a tough nut to crack:

    “The ultimate goal of physics is to describe nature and predict phenomena. It is impossible to do this starting with a priori theories; we would be stymied after a few steps, and every error would be compounded and would send us further from the right path. On the other hand, using experiments alone, we would soon be lost in a bewildering array of disconnected facts without any hope of making sense of them. It is the combination of theory and experiment, brought about by the use of mathematics as a language, that permits the astounding results physics has attained. It is Galileo’s immortal accomplishment to have clearly understood the power of this alliance and to have indicated ways of achieving it.”

    [p. 61, Chpt. 4, “From X-Rays to Quarks: Modern Physicists and their Discoveries”, Emilio Segrè, (1980)]

  • Clifford

    Excellent comment. A quotation to guide us all. Thanks Ijon Tichy! -cvj

  • Brian C

    I am of the opinion that string theory is complete and utter garbage. It shouldn’t even be considered a part of physics. Rather, it’s an incredibly complex and convoluted mathematical system with no connection to the physical world.

    The truth is, string theory is an embarrassment to the theoretical physics community. I can’t believe so many physicists actually buy into the mathematical gobbledy gook of branes and 10-dimensional strings.

  • Pingback: Approaches to Quantum Gravity | Cosmic Variance()

  • Pingback: The String Theory Backlash | Cosmic Variance()

  • David

    Why is it that there is no alternative to light being either a particle and/or a wave? Does the logical impossibility of the third have any relation to string theory?

  • Plato

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

About Sean Carroll

Sean Carroll is a Senior Research Associate in the Department of Physics at the California Institute of Technology. His research interests include theoretical aspects of cosmology, field theory, and gravitation. His most recent book is The Particle at the End of the Universe, about the Large Hadron Collider and the search for the Higgs boson. Here are some of his favorite blog posts, home page, and email: carroll [at] .


See More

Collapse bottom bar