More Links

For some insight into the difficulties associated with being an untenured professor and trying to get an NSF grant in astronomy, see this post (and comments) by Rob Knop. For those innocent of the ways of academia, there’s a post by Doug Natelson on what a well-organized faculty search is like (both of these via Angry Physics).

This week Witten will be giving two series of talks on gauge theory and geometric Langlands, one at Berkeley, and one at Harvard. Unfortunately I won’t be able to make it to either, perhaps someone who does attend some of the talks can report on what Witten has to say, in particular on what might be new since his long paper with Kapustin on the subject earlier this year.

Some of the talks from last month’s workshop on axions at the Institute for Advanced Study are now on-line.

Honeywell is sponsoring an educational program involving Nobel Laureates in chemistry and physics, called the Honeywell-Nobel Initiative. They already have material on-line from Leon Lederman and Horst Stormer, and from one of their promotional ads it looks like Frank Wilczek will also participate.

The Perimeter Institute has a new, improved web-site, with many online talks, both scientific seminars and talks for the general public.

The Templeton Foundation also has a new web-site. This month the featured “Life’s Big Question” is Do we live in a multiverse?

The European Mathematical Society has a newsletter.

Several people have noticed that the cover of the UK edition of my book is being widely emulated, see here and here.

The folks at Axes and Alleys have a review of my book in their latest issue. It seems that the members of the Royal Tractor Repair and Maintenance Society of Outer Mongolia found the book quite confusing. I guess this is only fair, since I’ve always found them quite confusing…

In case anyone is worried about the supply of unadulterated hype about string/M-theory drying up, there’s a new book out by John Gribbin which I took a look at yesterday in the bookstore.

Since the KITP does such a great job of providing a video record of the activities there, informal talks at the string theory phenomenology program have provided a fascinating public record of the way in which well-known string theorists are struggling with the all-too-obvious collapse of any reasonable hopes for getting predictions out of string theory. Last week there was an informal discussion with Michael Douglas of the String Vacuum Project, which was quite fascinating to listen to. I had trouble hearing what David Gross had to say, which was a shame, maybe someone with better ears who listens to this can report what they hear. Douglas encountered objections from the audience when he claimed that if one could get virtually any low energy physics out of one string vacuum or another, there was not any point to the whole project, with someone making the now standard claim that this situation is no worse than that of QFT.

This entry was posted in Uncategorized. Bookmark the permalink.

43 Responses to More Links

  1. anon. says:

    John Gribbin’s expertise in scientific predictions was first confirmed when at Nature in 1974, he came up with the money-spinning ‘Jupiter Effect’, which bears an uncanny similarity in some respects to biblical prophecy, alas without the confirmed predictions.

    His book with Plagemann, ‘The Jupiter Effect’ (1977 edition paperback), stated: ‘in 1982, when the Moon is in the seventh house, and Jupiter aligns with Mars and with the other seven planets of the solar system, Los Angeles will be destroyed.’

    Obviously, Dr Gribbin merely omitted to mention that this prophecy was fulfilled in an alternative parallel universe of the cosmic landscape of string theory.


  2. Bee says:

    Hi Peter,

    Thanks for mentioning the Honeywell-Initiative. Coincidentally, I came across their site today, but didn’t realize it’s new (I just assumed I had never seen it before.) I found it a bit disappointing though, both in design as in content, but then I guess they’ll stock up their database.

    […]a global education initiative designed to connect students across the globe with Nobel Prize winners in Chemistry and Physics… […] The Honeywell – Nobel Initiative establishes a forum for students worldwide to learn directly from Nobel Laureates in Chemistry and Physics

    Does somebody know if their concept of ‘global’ and ‘worldwide’ goes beyond those who have an high-speed internet access?

    Thanks also for the link to my blog, I hope you don’t mind the joke.



  3. anon. says:

    some other anon. wrote:

    “His book with Plagemann, ‘The Jupiter Effect’ (1977 edition paperback), stated: ‘in 1982, when the Moon is in the seventh house, and Jupiter aligns with Mars and with the other seven planets of the solar system, Los Angeles will be destroyed.’”

    Funny, I thought that was when peace would guide the planets and love would steer the stars. I guess that prediction got superseded.

  4. anonymous says:

    it is highly surprising that the “now standard claim that this situation is no worse than that of QFT” is standard. If the situation is no better than in QFT, what is the purpose of studying strings?

  5. Thomas Larsson says:

    with someone making the now standard claim that this situation is no worse than that of QFT.

    I think I finally understand what is wrong with this statement. It is true that there is an infinite landscape of QFTs, but it is the simplest ones that are realized in nature. Already free theories give a qualitatively correct picture of photons and phonons, and the simplest interacting theories are also realized in nature. phi^4 theory (Ising model) is the generic critical point and phi^6 theory (tricritical Ising) is the generic tricritical point, and among gauge theories, nature has selected the ones with the smallest gauge groups.

    You can in principle realize models deep inside the QFT landscape in nature. E.g., phi^300 theory would be the generic 150-critical point, which can be realized experimentally by fine-tuning 150 parameters, and the 150th RSOS model exhibits such behaviour. But although such complicated models exist in theory, they don’t exist in reality.

    In contrast, the simplest string theories are deeply unphysical: neither 26 flat dimensions and tachyons, nor 10 flat dimensions, unbroken SUSY and 496 gauge bosons are compatible with observation. There might be complicated models in the ST landscape which agree with experiments (or rather, have not yet been ruled out), but the big difference compared to QFT is that the simple string theories are wrong.

  6. spacepig says:

    Anonymous and TL:

    It is implicit that by going to string theory one has solved also the problem of quantum gravity. And, the people making this claim, all have in mind that (as with QFT) one should look for some new “stringy” phenomenon that determines which class of vacua is correct. Having observed this phenomenon, string theory fits the data (including gravity), QFT does not, and one used experiment instead of magical divination to find the correct description of nature.

    Many people try to get you to falsely see the issue as: “strings have a landscape, therefore we’ll never derive the correct vacuum without unforeseen breakthroughs.” That is not the issue. We never derived the correct vacuum in QFT either, we found it by experiment.

    The only obvious tuning that is required is for the CC. And that is NOT just a feature of string theory, it is a feature of ANY theory we currently have for the CC. So to say it is analogous to requiring a highly tuned phi to the 150th QFT is just wrong. Strings do not add any tuning that isn’t there in all approaches we have to the current data. And it solves problems that other approaches don’t.

    The whole landscape discussion in this sense has been deeply misleading. A feature (that one can make more realistic models) has been turned into a bug (how do we ever derive which of the much better models we now have is correct??). Of course that has been done on purpose, both by Peter and Lee (who dislike string theory),and by those string theorists whose dreams of changing science into an exercise in mathematics (“derive the world from thought alone!”) has been seemingly stymied.

  7. woit says:


    You’re engaging in pure sophistry, not science. A scientific theory is supposed to predict the result of an experiment. String theory doesn’t do this, and the nature of the landscape is the main reason why.

    Here and at Clifford’s blog you spend your time anonymously attacking not only me, but those string theorists (Dine, Taylor, Douglas) who are actually trying to do science and get a real prediction out of string theory. They’re not dreaming of “changing science into an exercise in mathematics”, they’re trying to do what scientists are supposed to do, come up with a prediction of what an experiment will see so the theory can be tested. What they’re finding is more and more evidence that this is impossible.

  8. John Baez says:

    Peter writes:

    A scientific theory is supposed to predict the result of an experiment. String theory doesn’t do this, and the nature of the landscape is the main reason why.

    I think it’s wiser to take spacepig at his word. He says the problem of determining the correct vacuum in string theory is comparable to the problem of finding the right Lagrangian in quantum field theory: we should give up on “deriving” it; it should be determined by experiment. That’s a consistent position, though a strong retreat from the long-held hope that string theory would relieve us of this need.

    If we take this position, we should ask whether it’s possible to find a string theory vacuum that matches existing experiments.

    As far as I can tell, the vacua most people like to study don’t work: they are manifestly supersymmetric, so they predict that for each boson (e.g. the photon) there should be a fermion (e.g. the “photino”) of the same mass, and vice versa. None of these “superpartners” have been seen. So, for string theory to have a chance to match existing experiment, people need to find vacua that aren’t supersymmetric.

    One approach is to come up with a theory of “spontaneous supersymmetry breaking” which could save the vacua that people like from being irrelevant to physics – but as far as I can tell, this is very hard, so not much progress is happening in this direction.

    It seems the hard-core theorists mainly leave this crucial problem to “phenomenologists”, who solve it by throw in ad hoc “soft supersymmetry breaking terms” into the field theories that can be derived from string theory vacua. To get theories similar to the Standard Model, this requires choosing over 100 new numbers. I don’t like all those free parameters. And what I like less is that nobody knows where these “soft supersymmetry breaking terms” should come from. As far as I can tell, they’re just stuck in by hand, out of desperation at being unable to solve the supersymmetry breaking problem in a principled manner.

    In short: if string theory now demands that we choose a vacuum to match existing experiments, it would be really nice if someone could find a vacuum that match existing experiments!

    Until then, Michael Douglas’ claim that string theory is “no worse” than ordinary quantum field theory seems overoptimistic. Yes, we can get gravity, but no, we can’t get matter that matches what we observe – not without extra ad hoc assumptions that nobody is able to derive from string theory.

    Saying we can get “virtually any low energy physics out of one string vacuum or another” also seems overoptimistic. We can get lots of different choices of low energy physics, but none of them match our universe – until one throws in extra “soft supersymmetry breaking terms” by hand.

    This at least is my impression. If progress has been made on this since I last checked, I’d love to hear about it.

  9. Bert Schroer says:

    The view of spacepig about QFT is a caricature seen through stringy eyes. It has nothing to do with what QFT really is. It starts with his metaphoric inflationary use of the terminology “vacua”.
    In QFT you have a dichotomy of algebraic structure and states. A theory is given by the spacetime-indexed net of observable algebras and its “states of physical interests” can be classified and hence belong all to the same theory. If there is a spacetime symmetry group acting on the algebraic net one can define a distinguished invariant state, the vacuum state, if there is no global symmetry (as in the generic QFT in CST situation) there is no distinguished reference state at all (in particular no vacuum). The total state space of the theory is a direct sum of “superselection sectors” (no coherent superposition between sectors) and the vacuum (if present at all) only belongs to one sector.
    In a reasonable way of counting there is only a finite number of perturbative renormalizable theories and renormalizability in the standard sense does not go beyond spin 1. There are quadrilinear couplings between scalar fields and Yukawa as well as gauge couplings up to spin 1, i.e. 3 types of couplings. The only remaining feature which one can play around with are multiplicities (different SU(n) or O(n) symmetry groups) but nobody with a rest of physical reasoning left would compare this to the different physical principles which underly the different string vacuua. QFT (even outside that perturbative setting) is bound together by testable physical principles (e.g. dispersion relations for scattering amplitudes); what bind string vacua together is metaphoric garbage.
    In order to save QFT from such a metaphoric sellout by ST, I posted a paper in todays hep-th.

  10. woit says:


    As Thomas pointed out, and I’ve argued repeatedly, the difference between the string vacuum case and the QFT case is that the simplest of a beautiful class of QFTs (gauge theories) works incredibly well and makes a huge number of confirmed non-trivial predictions, while the simplest string vacua are ruled out, and one has to go to ever more complex constructions, just in order to match some of the crudest features of the standard model.

    The problem with supersymmetry breaking terms is that you don’t even know what they are, just that they have to be big enough to avoid conflict with experiment. People seem to have string vacua that do have large enough amounts of supersymmetry breaking. It’s true that the state of current technology is that no one can do accurate calculations of most things in these vacua, but as far as I know there is no known reason that these things can’t give the SM at low energy. And if one vacuum gives the SM, all indications are that there will be essentially an infinity of others, ruining the hope of making any predictions.

    One could argue that one should pursue these studies as a way of potentially falsifying string theory. If one could understand all string vacua and see that none agreed with the SM at low energies, that would be something. Unfortunately, understanding in that detail “all string vacua” looks hopeless, and no one has seen any kind of argument that would allow a general conclusion that some aspect of the SM can not be matched by a sufficiently complicated construction.

  11. spacepig says:


    Thanks, you got my point exactly. I do not mean to be attacking anyone in a personal sense, I really think the quest to “derive” predictions from string theory would be rather like the quest to “derive” the correct Lagrangian for nature. (It is a bit different, in that the different Lagrangians are a superselection issue while in string theory the vacua are part of one theory, but this is irrelevant given the cosmological issues involved in probing one vacuum from another).

    It is not true any longer that string theorists study only supersymmetric vacua. Much of the fuss about the KKLT paper was that they proposed ways to break supersymmetry while nominally retaining computational control (and keeping the scale of breaking quite low, perhaps low enough to explain the hierarchy). This has been expanded upon in many works which explain how to break supersymmetry, in an apparently controlled way, in the framework of string theory. By now there is a huge literature on this, and most of the landscape papers that Peter and others deride, are actually working steadily to improve the technology for supersymmetry breaking in string theory. The most recent burst of activity in this regard has been tying the Intriliator-Seiberg-Shih models to string constructions, but there are many other approaches too.

  12. Anon says:


    What do you make of QFT’s like pure N=2 supersymmetric SU(2) Yang-Mills, which have continuous families of inequivalent choices of vacuum state (superselection sectors with inequivalent physics)?

  13. Bert Schroer says:

    John, it is simply not true that hard-core theorists do not look at the question at spontaneous SUSY breaking. The result is that it cannot be spontaneously broken in the sense as e.g. the Lorentz symmetry is broken in a thermal medium. They found such under such circumstances it “collapses” (i.e. there is no averaging which can save the symmetry in a highly mixed state, see my hep-th paper for more details and references).
    Again I emphasize that QFT models are bound together by common model-independent testable (and tested!!) common principles. This is a hell of a difference to the “vacuum problem of ST”.

  14. spacepig says:


    I don’t know what Bert is talking about, but standard references which
    discuss the issue of supersymmetry breaking include reviews like:

    Poppitz and Trivedi, hep-th/9803107
    Shadmi and Shirman, hep-th/9907225

    The subject developed after Edward Witten described the possibility of
    dynamical breaking of supersymmetry in the early 1980s, in the very famous
    paper Nucl Phys B 188 p. 513.

    As I said a significant recent effort has been devoted to making such models in string theory in conjunction with other moderately realistic features, and it seems pretty clear that one can do this, by now.

  15. Bert Schroer says:

    The word “spontaneous breaking”has a very precise conceptual and mathematical meaning. Its consequence is that by forming mixed avaerage states one still can recover symmetry (albeit in an artificial mixed state which violates cluster properties). Such an averaging does not exist for supersymmetry (the only known exception).

  16. spacepig says:

    Well, Bert, the references I gave are for the standard usage. These theories are thought (by e.g. the thousands of people who work on string theory and also go to SUSY 200X conferences each year) to be theories whose Hamiltonians are supersymmetric, but which do admit admit a supersymmetric ground state. Or in some cases (those with metastable susy breaking vacua), theories which possess both supersymmetric ground states and long lived metastable quasi-ground states.

    It is quite possible you have some “rigorous” meaning in mind, which I do not know about, but which is also not thought by the many phenomenologists who work on the MSSM, to be all that relevant. I suspect when they discover SUSY and some structure of soft terms at LHC, no one will care much about the axiomatic QFT existence of the definition of breaking which you are worried about.

  17. spacepig says:


    I am sorry, I meant “…but which do not admit a supersymmetric ground state.” Somehow I typed admit twice.

  18. Bert Schroer says:

    There is nothing “axiomatic” about a clear conceptual distinction between spontaneous symmetry breaking and no symmetry (=broken symmetry), you should not try to push common sense into some axiomatic or high-brow corner.

  19. spacepig says:

    The standard usage of “spontaneously broken” that I know of, states that a theory exhibits spontaneous breaking if the hamiltonian exhibits a symmetry that is broken by the ground state.

    That is the sense in which I was using the term.

    If you were using it in a different or more precise sense, the theories I mention may not exhibit spontaneous breaking by your definition. They do by the definition I gave.

    This is important: it means that in the theories I mentioned, the divergence structure is not sensitive to very high energies for certain processes. E.g. corrections to the vacuum energy, are cut off at the scale of supersymmetry breaking. So in principle it is finite and calculable.

  20. Bert Schroer says:

    This is not quite the correct definition. The diffeomorphism group of the circle e.g. does not leave the conformal vacuum invariant (only the Moebius subgroup has this property) but this group is by no means broken in chiral theories.
    Your definition does not cover the breaking in thermal states. The problem there is that the ground state Hamiltonian is not the same as the one which leaves the thermal average state invariant (one cannot argue here with quantization boxed and Gibbs states because they break the supersymmetry in an explicit manner).
    Coming back to the statement of John, there is no mathematical theorem that supersymmetry can never be spontaneously broken in vacuum state (presumably you want it in such a way that the Poincare-group is conserved) but that special “collapse” aspect in a heat bath (where any other symmetry is at most spontaneously broken) makes the whole issue of spontaneous SUSY breaking extremely suspicious. Fermions and Bosons are separated by the strongest superselection rule in this world: the univalence superselection rule. My physical explanation for the thermal collapse phenomenon: you only get SUSY by an extremely fine tuning of coupling strength and such a fine tuning is extremely unstable (more unstable than the fine-tuning problems which SUSY is supposed to solve).

  21. spacepig says:

    OK. Well, supersymmetry can be spontaneously broken by my definition; that is the definition the phenomenologists who work on the MSSM and its extensions use; and theories with that kind of breaking of SUSY lead to the successful prediction of unification of couplings, good dark matter candidates, and a solution of the hierarchy problem. The recent landscape constructions incorporate more or less this kind of supersymmetry breaking into string theory, quite successfully. (Though by no means producing a fully realistic model of the world — who knows if that is possible).

    I have no knowledge of, or intelligent comment to make, about your definition.

  22. Bert Schroer says:

    No, the situation remains undecided. Whatever our definitions are, we don’t spontaneously break a symmetry rather the question is whether nature permits such phases (symmetric and phases with order parameters).
    I think that your picture of breaking “by hand” is caused by a too formal reading of the Schwinger-Higgs mechanism which is not a symmetry-breaking (even though people use this misleading terminology). Local gauge transformations have nothing to do with symmetry and hence cannot be broken. There is a hell of conceptual distinction between the Goldstone spontaneous broken phase and the Schwinger-Higgs description of interacting massive vectormesons even though formally (in the Lagrangian approach) they look very similar.

  23. spacepig says:

    At this point, I can only make the sociological comment: the rest of the field
    has already moved on, so in practice, the issue is decided. Spontaneously broken susy plays a central role in many papers that appear each day on the arXiv. I think this is well justified, but other readers will have to judge for themselves.

    Just as the field has moved on (correctly or not), I will too. I have nothing else to say about this subject.

  24. Bert Schroer says:

    o.k. I know that there is a difference between science and sociology and the only thing what happened in this long discussion (which is now finished) is that you confirmed this.

  25. Peter Woit says:

    As far as I can tell, this discussion of the formal status of supersymmetry breaking is irrelevant. Partly because you don’t actually have a real underlying theory that does what you want to discuss (no non-perturbative version of string theory), partly because even if it does what you want (break supersymmetry), you end up with all sorts of different classes of models. The simplest of these models that doesn’t violate what you already know (the MSSM) has 105 extra parameters in it, and no known reason these cannot have arbitrary values.

    Using the word “success” in this context is a bit absurd. The bottom line is that you can’t even tell us if the LHC will see anything at all. If it does happen to see exactly the superpartners, you can extract predictions about some of their behavior, but if these predictions don’t work out, all you’ve done is falsified the simplest model, and you have an infinite array of other models you can use to match the data. In that case you’re not testing a theory or predicting anything, just parametrizing observations in a complicated way.

  26. Bert Schroer says:

    Fine, so there is after all a scientific conclusion.
    Spontaneous supersymmetry-breaking is conceptually extremely suspicious and phenomenologically (in the form of superstrings) totally futile.

  27. JC says:

    Years ago I first bought into the SUSY thing largely from the standard arguments in favor of it, such as dealing with the hierarchy problem, unification of coupling constants, etc … When I came to the realization it would take around 100 free parameters to incorporate SUSY into the Standard Model, that’s when I was turned off from SUSY initially. (Before that time, I thought the Standard Model had too many free parameters).

    When string theory became popular, I initially thought it could find an easy way to reduce the number of free parameters in the MSSM. But so far that has not happened yet in a convincing manner. At times I wonder why some theorists are tolerant of the idea of zillions of free parameters. Do any string theorists today believe in the idea that string phenomenology can completely determine the free parameters of the Standard Model?

  28. Anon says:

    Finite temperature breaks supersymmetry, just as it break Lorentz invariance (there’s a preferred reference frame, in which you are at rest, with respect to the thermal bath).

    This is not what people are talking about, when they talk about the spontaneous breaking of supersymmetry (at zero temperature). As spacepig points out, there are plenty of theories, in which the Hamiltonian (or Lagrangian) is supersymmetric, but there is no supersymmetric ground state. Instead, there’s a massless Goldstone fermion, which transforms inhomogeneously under supersymmetry, etc…

    The real setting, for phenomenology, is one in which one is looking at supergravity, rather than global supersymmetry. The massless Goldstino is eaten by the gravitino, in an analog of the Higgs mechanism.

    Bert Schroer will complain that this statement is not solidly grounded in quantum field theory. He is correct. Supergravity (or any kind of quantum gravity theory) is not a QFT, as he understands the term. And, to really define supergravity, it needs to be embedded in some UV-complete theory (like string theory) that he would consider completely beyond the pale.

    But I think he misunderstands even the case of globally supersymmetric QFTs, in which supersymmetry is spontaneously broken.

  29. spacepig says:


    Wow, the opinions here, about SUSY and its breaking, are totally nonstandard. Anyone trying to “learn” about physics from blogs, beware!
    Mr. Schroer has lightly dismissed the work of thousands of papers over a period of 20 or more years, finding theories that exhibit spontaneous and dynamical (spontaneous, nonperturbative) SUSY breaking.

    JC: The MSSM does have approximately 125 parameters. However relatively trivial symmetries, which are also required by data (things like R-parity, needed to avoid proton decay), cut down the number. There are a few well motivated scenarios with just a handful of parameters.

    It is however a problem of some interest, usually now called the LHC inverse problem, to try and figure out which version of the MSSM with soft breaking parameters is the correct one, given some hypothetical LHC data that confirms SUSY. It is not an easy problem, and indeed because of the large number of parameters, one might have to wait for ILC or some other machine to really nail down the model.

    Most string theorists do not believe the use of the theory will be to determine these (kinds of) parameters. It will be useful for phenomenology if and when some stringy signature, like cosmic strings or moduli, shows up in experiments. This is something peter continually refuses to acknowledge, but is a far more likely scenario for contact, than classifying string vacua and then finding the right one by brute force. It has been useful for more conceptual issues (like singularity resolution and black hole entropy), solving gauge theories, motivating new classes of models, etc. already. In fact it would be hard to gauge the strength of its impact on these subjects over the past 20 years, it has been enormous.
    And, from the viewpoint of many established, contributing theorists in
    many top departments, it has been very positive.

  30. Bert Schroer says:

    In my case it was not lightly; it would be much simpler for me if I permit myself to use what Feynman called “excuses”.

  31. Peter Woit says:


    I agree that the most likely scenario for making contact between string theory and phenomenology would be if astronomers find a cosmic string, study its properties, and find that it has the properties elementary strings of cosmic size are suppose to have distinguishing them from cosmic strings coming from a Higgs field.

    But there is not the slightest evidence for this, and there is no reason to expect it other than pure wishful thinking. It also would be strong evidence for string theory if, when LHC collisions begin, an angel appears out of the interaction region with gold tablets explaining that the multiverse exists and is governed by the laws of string theory. I just don’t see any reason to expect this to happen, but you’re welcome to put your hopes in it. Just doesn’t seem like science to me.

  32. spacepig says:

    The cosmic string scenario seems more likely than the one with angels,
    to me. As do scenarios with 5th forces (which is why several groups are doing experiments to detect them). Or scenarios with low string scale (which is why Lisa Randall wrote a popular book that recently won her the Lilienfeld prize, about string-inspired scenarios which solve the hierarchy problem).

    Particle physics has always progressed by “model building,” not by epiphanies which predict the structure of the whole theory. You can make fun of this and say “there is no guarantee any interesting model will be selected by the data,” and you’re right. But there was no such guarantee for the past 100 years, and yet, we did fine. The problem for the last 25 has been the lack of experiments that push the energy frontier, the SSC cancellation didn’t help, and with LHC that will finally be corrected. I expect the bottom up approach will continue to succeed, with no need for angels, and interesting phenomena will continue to turn up.

  33. Peter Woit says:

    If we see a 5th force it will be quite interesting, but far from evidence for string theory.

    I certainly hope and expect that interesting phenomena will turn up at the LHC or elsewhere, I just see no reason at all to expect that they will have anything to do with a 10/11d string/M-theory with something very complicated done to the 6/7 extra dimensions.

    Particle physics has not progressed due to model-building itself, it has progressed due to people coming up with models that actually have to do with the real world, based on finding testable, compelling models that both explain things already seen, and predict things not yet seen, which are then confirmed.

    That Lisa Randall’s popular book has received a prize is absolutely irrelevant to the question of the validity of the models she discusses. You seem very concerned with sociological rather than scientific evidence for models. If someone decides to award a prize to my book, will that prove that string theory has failed as an idea about unification?

  34. spacepig says:

    Peter, you fail to acknowledge that in fact string theory has been an inspiring source of models that have to do with the real world, have testable and compelling features, and predict things not yet seen. None of these models have yet been confirmed. If one is, it will not be proof of string theory. But to pretend string theory has not had this role, is really strange.

  35. Peter Woit says:


    String theory has inspired all sorts of things over the last 35 years, some of them very important, some of them worthless. One thing it has definitely inspired is some great mathematics. I don’t happen to be a big fan of the kind of phenomenological models I think you have in mind; they appear to me to be complicated, and don’t actually seem to really explain anything that we observe. Maybe I’m wrong and the LHC will provide experimental evidence for them, we’ll see. If so, that would be a positive thing to attribute to string theory research. It wouldn’t mean though that the world is 10/11 dimensional, or that string theory ideas about unification have anything to do with reality.

  36. JC says:


    How would you feel if the LHC and all other near-term future colliders (ie. built and running before 2100), only found the standard model Higgs particle and nothing else?

  37. anon. says:

    “… the opinions here, about SUSY and its breaking, are totally nonstandard. Anyone trying to “learn” about physics from blogs, beware!” – Spacepig (comment number 19,194)

    Spacepig, you mean to write, surely, “… Anyone trying to ‘learn’ about orthodox string theory which we claim is physics, from blogs, beware!”

    “String theory has inspired all sorts of things over the last 35 years, some of them very important, some of them worthless. One thing it has definitely inspired is some great mathematics.” – Peter Woit (second to last comment).

    Can you actually substantiate this claim, please? Please give a great equation from string, and show how to solve it. Kaku in a New Scientist essay a year or two ago wrote that he cried when he first saw the elegant simplicity and predictive power of Dirac’s equation. Is there anything like that in the stringy Calabi-Yau’s? I thought that the whole point is that the failure of string theory is a mathematical failure to actually solve the extradimensional equations because they’re so inelegant? Surely there is nothing useful in equations which can have so many solutions?

  38. Bert Schroer says:

    this is an issue on which I agree with you; perhaps one can bring it more to the point by the following remarks.
    It is not mathematical physics which is helpful for unifying mathematics but rather “physical” mathematics i.e. metaphoric ideas about particle physics which are produced most abundandly in the present crisis of particle physics. This is pretty evident in the way physics ideas enter the Langlands program. It is not the world (say the SM) as it is in nature, but rather the way Atiyah, Seiberg and Witten… want it to jump over their e-m duality stick which is useful for mathematics. The reason why this is successful is that mathematicians need once in a while (especially for unification of their field) to get away from the rigor and activate their free-floating imagination; metaphorical ideas from particle physics are the ideal catalyzer for this.
    What is fruitful for mathematics is a dead end for particle physics.

  39. anonymous says:


    you abandon the hope of predicting the vacuum. You abandon the hope of making statistical predictions. You suggest that what theorists did so far is enough. I agree that some scenarios would smell more stringy than others and that finding just the SM at LHC would smell anthropic. But selecting a QFT means selecting a QFT and does not mean proofing that string theory is the theory of quantum gravity.

    If you wish to abandon the scientific method, we can close now the issue about string theory with the following pool:

    a) it is 26th-century physics that fell by chance into the 21th century;

    b) it is 21th-century archeology;

    c) it is a fart in 11d space-time.

    If instead you want to follow the scientific method, you must provide testable predictions. In my opinion, if this cannot be done, string theory has to dissolve as in c).

  40. Tony Smith says:

    Now that I have a copy of Freeman Dyson’s book “The Scientist as Rebel”, I see in it some passages that, like its cover, remind me of Peter’s description of superstring theory.

    Dyson said:
    “… At the beginning of the seventeenth century, the birth of modern science had been proclaimed by … Bacon in England and … Descartes in France.

    According to Bacon, scientists should experiment … and collect facts … until the accumulation of facts would make clear the way nature behaves.

    According to Descartes, scientists should deduce the laws of nature by pure reason … Cartesian vortices were supposed to fill space … pushing celestial objects along their orbits.

    At the time when Newton made his discoveries, the learned men of England were mostly doing science in the empirical styple of Bacon, but … believed in the Cartesian theory of vortices because it was the only theory available.

    In 1667 Newton … resumed his solitary existence in Cambridge. … He spoke to nobody about his alchemical studies, and to almost nobody about his discoveries in physics.
    For him, alchemy and physics and theology were parts of a single enterprise, three aspects of a single search for knowledge that God had placed within his grasp. Since he was not free to talk about his theology, he saw no reason why he should talk about his alchemy or his physics.
    He might never have talked about his physics, if his friend Halley had not come to Cambridge in 1684 begging him to publish what he knew. Then, once he had started writing … he did not stop until he had finished the three volumes of the Principia. …
    In the first two volumes …[of] the Principia …[Newton]… built a grand edifice of mathematices … and then in the third volume he … demonstrat[ed] with an abundance of observational facts that nature danced to his tune.
    As soon as the Principia was published and widely circulated,
    the Cartesian vortices were dead. …”.

    If Halley had not convinced Newton to publish the Principia, what would have been the course of science ?
    Would others (using Hooke gravity, Leibnitz calculus, etc) have produced a Principia,
    would Hooke gravity have remained as empirical physics and Leibnitz calculus have remained as mathematics reasoning, with the Newtonian synthesis never occurring ?

    Tony Smith

    PS – I guess I should say explicitly that the similarity that I see is that:
    Baconian accumulation of facts = Standard Model plus gravity
    Cartesian vortices = superstring theory
    “the only theory available” = “the only game in town”
    maybe a Newtonian synthesis ( a model using new or unconventional math techniques that enables calculation of particle masses, force strengths, etc) might be necessary to unify Baconian standard model plus gravity physicists with Cartesian beautiful-math physicists.

    PPS – As Dyson’s essay also said:
    “… Newton himself was at heart a Cartesian …”.
    That is borne out by the fact that Newton lived his life in accord with an Ovid quotation on Descartes’s tombstone:
    “Bene qui latuit, bene vixit.”
    there exist today pressures against unconventional physics similar to the pressures against unconventional theology in Newton’s time, and it may be that even today some feel that they have a better life by hiding their true feelings.

  41. I’ve been trying to look at books that might have something to say about strings and say to myself, “What would Peter think?” to evaluate it. I’m not very far into it, but I was actually thinking that Gribbin’s book, though undoubtedly pro-strings, was more skeptical than I’d thought it would be.

  42. Jeremy says:

    If you want to read the Axes & Alleys review online and not just in the PDF, you can find it here:

  43. Bert Schroer says:

    Dear friends
    I stop and say good by to all who engaged me in interesting discussions. It is not only the spam-filter which irritated me but I find this whole enterprise unworthy, regimented to rules which never were spelled out, and at the end and plainly ridiculous.

Comments are closed.