Not Even Wrong

By Sean Carroll | August 23, 2005 1:17 pm

Peter Woit, noted blogger and string-theory gadfly, has written a book about his objections to string theory: Not Even Wrong, to be published next year by Jonathan Cape.

Good. I completely disagree with Peter’s opinions about string theory, and think that his accusations that the Landscape is non-scientific are completely off the mark. But his objections are not crazy, and his dislike for the theory is grounded in an informed scientific judgement. (Sometimes more than others, but that’s a matter of personal opinion.)

The whole discussion is a nice contrast with the Intelligent Design mess. The fact is, we don’t know what is the correct theory that unifies particle physics with gravitation. String theory is far and away the leading candidate, but its status as leader is a reflection of the educated judgement of the experts, not any airtight evidence. This judgement comes from looking at various pieces of information — what we know about gravitation, and quantum mechanics, and particle physics, and the history of ideas in physics, and the mathematical structures underlying gauge theory and general relativity, as well as an intuitive feeling for what principles are most important and what clues most worth pursuing — and deciding which path toward progress is likely to be fruitful. When people like Peter (or Lee Smolin) read these tea leaves, they come to a different conclusion than most scientists in the field. But it’s healthy disagreement among professionals working at the edge of what we know and don’t know — not politically-motivated intervention from people who have no clue, just an agenda, and operate completely apart from the scientific mainstream. To people looking in from the outside, I hope an accurate picture comes across: there is a widespread feeling that string theory is the best hope for a quantum theory of gravity, but it’s not a settled issue, and we’re working in good faith on moving forward.

So I’m happy to see this side of the argument represented in the popular press, even if I disagree — we shouldn’t be afraid of the free market of ideas. If people don’t agree, they should explain the sources of their disagreement rationally. There is always the danger of misprepresentation of course, and in this case there is an obvious worry — that a spate of stories will appear about how string theory is in trouble, and a house built on sand, and so forth. That might be true, but certainly isn’t the impression I have from talking to string theorists. In any event, I hope that we defenders of the theory can stick to the high road, and welcome this intervention in the discussion of these important ideas.

CATEGORIZED UNDER: Science and the Media, Words
  • Wolfgang

    Since the main text is posted twice, I figure I should post my comment also twice ๐Ÿ˜Ž


    since you mention string theory and Intelligent Design in one blog entry,
    I have to say that ID in the form of “Flying Spaghetti Monsterism” (yes, I am a Pastafari) finally found a way to unite religion and science.
    The FSM obviously created the world in His image as a collection of vibrating spaghetti (= strings).
    You say there are also branes ? Well, to every meal of Spaghetti there is a plate …
    So, Peter’s book is clearly blasphemous, but spaghetti theorists knew this all along .

  • Peter Woit

    Hi Sean,

    Thanks for your comments. I’d like to remark that the objections to string theory are only one part of the book, although the part that will undoubtedly get all the attention.

    I agree with you that string theory and Intelligent Design are two very different stories. Intelligent Designers don’t even try to make real science out of it, and have an explicitly religious and political agenda they are pushing, whether they admit it or not. String theorists are operating within the realm of standard speculative scientific practice, even though I very much disagree with their take on the scientific evidence for the validity of this speculation that has piled up over the last twenty years.

    But thinking about string theory and the landscape has convinced me that the scientific method is something a bit more subtle than we sometimes think. There’s lots to be said about the landscape, but the kind of “anthropic” and related research programs that some people are promoting seem to me to cross over the boundaries of what can legitmately be called the scientific method. When I see leading physicists promoting research which doesn’t seem to have any plausible way of ever leading to the making of falsifiable predictions, I see real dangers for the field, including the danger of not being able to effectively answer the challenge to legitimate science from ID.

    Fundamentally, the most attractive thing about science to me has always been the fact that it doesn’t rely on faith or on appeals to authority, the way religion does. If one has the time and energy, one can look into the evidence for any scientific theory, and decide for oneself on its validity. I hope that my book, by laying out the opposite side of the argument from the pro-string theory one that has had wide distribution, will allow many people to make up their own minds about this issue.

  • Gordon Chalmers

    I still dont see why string theory is frowned upon. I get mail all the time telling exactly the opposite. But I have to admit that people are entitled to their opinions.

  • DrMax

    EXACTLY! What is a theory worth if it does not allow for questions?

    Now if you string-theory guys were the current administration, you’d be digging in to this Woit’s guy’s background so you could have your smear campaign ready by the publication date.

  • Amis

    Sean hits the mark when he said that string theory is currently the leading candidate to unite particle physics and gravitation. Although no one knows if the theory is true, studying it is a worthwhile enterprise that the theorists take on. What may be worrisome is the scenario that for the next 200 years, reputable institutes hire half of their faculties on a giant theory that will be proven utterly true or utterly false, at the expense of other fields in particle physics.

  • hack

    “Now if you string-theory guys were the current administration, you’d be digging in to this Woit’s guy’s background so you could have your smear campaign ready by the publication date.”

    Don’t worry, Lubos is already on the case!

  • Lubos Motl

    Dear Peter,

    let me repost a reaction to your ideas about the “wide spectrum of ideas” and your focus on Chern-Simons theory.

    You seem to misunderstand the difference between mathematics and physics completely. Chern-Simons theory is simply not a theory that is designed as a competitor of string theory to unify the known physics.

    Chern-Simons theory is a theory that admits a similar type of (quantum-field-theoretical) description as some physical theories, but that apparently lacks the physical strength to have anything to do with the observed particle physics and that can only bring us interesting mathematical results, not testable physical results.

    Chern-Simons theory is just an effective description of D-branes in topological string theory.

    The main problem of yours is that you have completely lost your knowledge of physics and especially the idea which mathematical ideas may be relevant for which physics. When you talk about the “wide range of ideas” that physicists should be talking when they try to go beyond GR+SM, you obviously don’t know what you’re talking about.

    There exist no general conceptual frameworks to surpass the existing theory except for string theory, and if someone tries to force people to work on these non-existent ideas, he is doing the same job as the Intelligent Designers. It just can’t work. There exist “small” ideas how new phenomena behind the Standard Model could look like, and this is what phenomenologists work on. But there is no unifying deep structure except for string theory. It’s not a theorem yet but it may well become one next year.

    Most likely, you will never be capable to understand why these alternatives to string theory can’t work – because you’re probably just too old for these things and you have not learned these important technical things in time. But you should at least try to understand that there is a crucial gap in your knowledge that makes all your “big conclusions” totally worthless.

    You just can’t judge string theory without knowing anything about its math, its physical implications, and its uniqueness, and if you try to make big conclusions anyway, then you’re a crackpot.

    Best wishes

  • Quantoken


    Super string theory and ID are certainly very different. But they differ in exactly the opposite way from what you described. If you take off the religious agendas that are actively pushing the idea, and just look at the Intelligent Design Hypothesis by its own merit. Then it is a perfectly scientific hypothesis: It makes certain assumptions about the nature, and those assumptions can be checked against evidences. Thus such a hypothesis is perfectly falsifiable, and indeed it has already been falsified, by compeling evidences supporting the Theory of Evolution. The ID theory, just like the Ptolemy theory, is shown to be a wrong theory because they are eventually falsified by evidences. Any thing that is falsifiable and is willing to be subject to the falsification of nature, is within the domain of legitimate science inquiry, regardless of the final outcome.

    On the other hand, there is so far no evidence that super string theory ever makes any prediction or is ever falsifiable at all. If something does not make a connection to the nature and can not subject itself to the falsification of nature, it really can not be called science. And, for the benefit of Sean, the kind of firm belief you saw in super string theorists, without any any experimental backing whatsoever, is the kind of faith based belief, versus the other kind which is evidence based.

    In a broad definition, any faith based belief IS a religion. Whether that religion is called Christianity or other names is an un-important detail.

    To be fair, SSTers do believe they need to find material support, and they are struggling towards that goal. I am not going to say that the odd they eventually find experimental support is zero. Good luck searching. Mean while, before that happens, your faith is still a religious one.


  • The Anti-Lubos

    Has Lubos ever said anything, ever, that was 100% free of rudeness? Is he capable of it?

  • Gordon Chalmers

    Is there somebody else in Harvard besides a Motl person?

  • Wolfgang

    > Has Lubos ever said anything, ever, that was 100% free of rudeness?

    Lubos always adds the ๐Ÿ˜Ž and ๐Ÿ˜‰ and so it is not really rude ๐Ÿ˜Ž

  • Lubos Motl

    Thanks, Wolfgang, for your understanding of my true gentle heart, and greetings to your daughter! ๐Ÿ˜€

  • spyder

    there is something reassuring here when Sean writes: “as well as an intuitive feeling for what principles are most important and what clues most worth pursuing รขโ‚ฌ” and deciding which path toward progress is likely to be fruitful.”

    Intuitive feelings and making choices from those is a wonderful human characteristic. No matter how expansive a mathematical modelling becomes in multi-dimensions or how infinitely complex the probabilities must be for SST to be manifest, science and faith really come down to our acknowledgement of our inability to not be human. I like that about it all. The path through the maze of discovery about what is true about the constructs of the universe is filled with deadends, cul de sacs, huge monsterous obstacles, and tiny glimmers of light needing our very human intuitive awareness to guide us on. As my rocket scientist father often told me: As long as we are human, we can never really know!

  • Anonymous

    Questions for Quantoken:

    1) Exactly how has ID been falsified?

    2) GR and the Standard Model are incompatible but separately successful. This issue needs to be reconciled. String theorists are trying to come up with a falsifiable theory that will do this, but they admit they haven’t found one yet. Do you know another way to go about doing things?

  • Fyodor

    I think Peter Woit’s book should be used as a text in every physics department. Let’s teach the controversy, folks.

    Oops. Where have I heard that before? :-)

    By the way, it is a bit unfair of Sean to bracket PW with Lee Smolin. Smolin at least has an alternative to offer.

  • Quantoken


    (1)Buildings are by intelligent design, for example, and trees are by evolution. The same abundance of fossil evidences that supports the Theory of Evolution also falsifies the Intelligent Design.

    (2)The way of unifying GR and QM is neither to quantize GR, nor to derive a quantum theory that contains GR. Neither way works. The correct approach is a theory of quantum information conservation, which naturally leads to both GR and QM, as necessary ingredients to ensure quantum information conservation. Put it in this way: GQ and QM are two branches grown from the same tree, NOT two sides of one coin.

    Many people have begun to realize the importance of the concept of information, or entropy, in building a TOE, but I guess I am the first one to realize that the key is the conservation law of quantum information. The math details still need to fill out. But it surely leads to both curved spacetime, and uncertain principle, in a very natural, and mandatory way!!!

    Put it this way, given the requirement that only a finite amount of information can be embedded, and you can pick and choose the rest of physics laws any way you want, even the God can NOT create a world that is classical, deterministic, with space and time extends to infinity and no gravity exists, and still be compatible with the requirement of finite information.


  • Anonymous


    1) I disagree, in that evidence for one theory does not necessarily imply evidence against another theory (oops – called ID a theory just then!). A smart IDer (?) agrees that the tree fossil record is consistent with descent with modification, but he claims that there are certain structures on trees which are too complex to have evolved. So I don’t see how ID can ever be falsified. I can always find something that seems too complex for evolution to make – that is, until biologists explain every single solitary detail about life on earth.

    2) “Many people have begun to realize the importance of the concept of information, or entropy, in building a TOE, but I guess I am the first one to realize that the key is the conservation law of quantum information.”

    You seem to have many non-mainstream ideas. What is your science background?

  • Richard

    “There exist no general conceptual frameworks to surpass the existing theory except for string theory, and if someone tries to force people to work on these non-existent ideas, he is doing the same job as the Intelligent Designers. It just can’t work. There exist “small” ideas how new phenomena behind the Standard Model could look like, and this is what phenomenologists work on. But there is no unifying deep structure except for string theory. It’s not a theorem yet but it may well become one next year.”

    This is an astonishingly pompous statement. Until I read the last sentence my immediate reaction was where is the proof? Proof by assertion or intimidation? “There exist no …” requires rigorous proof. Proof of this statement would require showing that any conceptual framework that “works” can be faithfully represented in some mathematically rigorous way within the string theory framework, i.e., string theory must provide a universal object. I’d love to see this attempted without circular arguments.

    As for the last sentence, a statement for which there is no known proof is not a theorem or even a proto-theorem. It is called a conjecture. Spell it … c o n j e c t u r e.

    Do we have a crackpot in our midsts?

  • Plato

    Let me put it plainly.

    Let religion to those who want to believe their religion. Each has that right under your constitution, and let me also say that this was well thought of when Benjamin Franklin went over Thomas Jefferson’s words.

    Some might argue these things are ancient and lost touch with reality, but they seemed to have seen farther then those who converse string theory and idealization to ID.

    Everyone can believe in their own God if they like, but the essence of the debate about science should be about science and not associative trademarks that people like to give them with ID.

    Like I said, I have watched this debate from the sideline and watched it fuelled by connotation given from those in the know. Would such racial profiling be accepted, that you find each others position, “satanic verse”?:)

    Let ID die it’s own death, and let science spread forth it’s wings to dailogue and reason. I’d put some words from Benjamin here for others to examine, but I am having computer problems right now

  • Scott

    I see Richard has already pointed this out but, Lubos said,

    There exist no general conceptual frameworks to surpass the existing theory except for string theory.


    and you know this how? It hasn’t even ever been shown that the string framework is a way let alone the only way. But apparently this will all become clear next year…


    on a related note, there are many speculative ideas one could explore to make a conclusive theory for which GR and QM are different limits, there is a list of some of the more popular ideas at wiki,

    A lot of the more popular methods try to make gravity and the feild theories different types of the same thing something which seems very unlikely to some people like myself and apparently quantoken, who think that instead there is a encompassing framework of which they are completely different aspects.

  • Thomas Larsson

    Gordon #3: The main reason why people frown upon string theory was succinctly formulated by string theory pioneer Dan Friedan in hep-th/0204131, subsection 1.6:

    “Recognizing failure is an essential part of the scientific ethos. A complete scientific failure must be recognized eventually.”

    I think that there is a quite widespread feeling that the string community is acting unethically in this respect.

  • Clifford

    Sigh. Here we go again. Please see the discussion on the Landscape thread .


  • Haelfix

    Im with Sean on this. ST is promising, but certainly not ironclad. Its not even ironclad mathematically, in a way mathematicians like Peter would feel comfortable with. Nm the lack of experimental evidence

    But thats ok in my book, string *theory* isn’t really a *theory* yet. Its more like a rather general program to understand quantum gravity. In 20-30 years or however long it takes them to start making things really rigorous along with a good idea of which phenomological model to use, then perhaps it might be fair to criticize it as a failure or success.

    But until that time, I still think there is enough tantalizing promises inherent within the structure to make it the primary research endeavour.

  • Wolfgang


    > ST is promising, but certainly not ironclad

    I agree that superstring theories are “promising” and indeed M-theory is a very good candidate for a consistent quantum theory of gravitation.
    Unfortunately, there is no empirical evidence for a main ingredient: supersymmetry. The LHC data (expected for 2008) may change that.
    Until then, physicists will investigate other ideas as well, this is what physicists do, whether Lubos likes it or not.

  • Anonymous


    I was actually more interested in Quantoken’s idea that string theorists are being “religious” in that that their work is not yet testable. I wanted him to put forth a non-religious way of proceeding – something that was testable or for which there was evidence, and he didn’t.

    He usually makes very strong statements about physics, and I have wondered about his science background. I think I’ve figured it out from his blog.

  • seebee

    It seems that quantoken has single handedly destroyed any progress that Sean has been so assiduously and eloquently been making in this blog in defining the issue with ID, namely, there is no issue. Intelligent design is, in my humble opinion, nothing but a religious agenda. While one might arguably consider a building to be a product of “intelligent design”, whereas one must have faith for the brand of ID the scientific community is concerned about, I don’t think much faith is required to recognize the existence of the architect who’d designed that building, and if you still do doubt that, look in the yellow pages and give her/him a ring. Let’s not try to be dodgy about this subject by trying to be kind to the theory of ID and considering how it just might contain a valid point. As has already been stated, exclamed, cried and lamented in this blog, ID does not enjoy the priveleges nor rights of a full-blooded scientific theory.

  • Clifford

    Dear Peter,

    As a result of several discussions on other comment threads on this blog, I was under the impression that we’d all made some progress in sorting out what were well-posed disgreements you have with some approaches research in string theory, what were “gut-feelings” that you have (over which we can simply agree to disagree), what were misconceptions based on not being an active researcher in the field, and –very importantly– what were simply your misattributions of a minority view to that of the whole field. Recall that I spent a fair amount of time trying to clear these up. I refer you to the comment thread of the Landscape post, for example. I thought we arrived at some agreement that your views about what is actually going on in the field need a bit of re-balancing. If so, will these refinements be incorporated into the book before it is published? Or will your pre-cosmicvariance position be published? I do hope that these “finishing touches” might involve significant rebalancing some of your emphasis to reflect the outcome of the enlightening discussions that have taken place here. Otherwise, it will be a missed opportunity for you to put out a book that is a useful alternative view, and not just a view based on an exaggerated chariacature of research in string theory.

    I’d like to ask you to pleaase make the effort. It probably won’t delay publication at all, and even if it did, it will be worthwhile: It will improve your book, and thereby enhance your reputation. If it comes across as an uninformed rant, however, you’ll do service to nobody’s cause at all, which would be sad, at the very least.



  • Clifford

    Dear Peter,

    I refer particularly to your comment # 67 in that thread, although it is worth reminding yourself about the discussion that led up to that point. Quoting you entirely:

    Peter Woit on Aug 15th, 2005 at 8:42 pm


    Sorry for harassing you into stating the obvious that once one has shown a theory is unpredictive, it’s wrong (or not even wrong…) and one has to give up on it, but I think this discussion was worthwhile, it certainly helped me clarify some things for myself. And it’s helpful to see that we share fundamental criteria for evaluating science. I’m afraid that I sometimes share what I take to be Lee’s perception that for some string theorists, the possibility that the idea of string-based unification is just wrong seems to be something they won’t even admit to be a possibility.

    No, I’m not going to take you up on your suggestion and devote myself to working on string theory. There are already many, many smart people doing this, and they appear to me to be doing a good job of slowly accumulating evidence that the string theory unification idea doesn’t work. I don’t think I could significantly speed that process up. I’ll stick to pointing out what other people have already found, and trying to develop what seem to me to be more promising ideas.

    The main thing this clarified for me is the whole issue of falsifiability. You and Sean are right that it’s a good idea to think about the analogy between the gauge theory and string theory frameworks, although I draw different conclusions from this analogy. I guess I do think that the difference is one of degree, but that differences of degree are crucial. Whatever theoretical framework one has, one can generally find some way of making it fit the facts. If it’s a good theoretical framework it’s easy, if it’s not you have to engage in all sorts of ugly contortions. Thus, in evaluating theoretical frameworks, a sense of aesthetics is crucial, and claims like those that Susskind is making that it doesn’t matter if things are really ugly are dangerous. I’ve been thinking a lot in recent years about this kind of “aesthetic” issue, and the connection to falsifiability is something I hadn’t thought about before.

    So, will the refinements of your views mentioned by you in the above be reflected in the book? (Not to mention other points I mentioned which you agreed with elsewhere on the thread?)



  • Peter Woit

    Hi Clifford,

    Yes, the discussion here has had an effect on some of the changes I’m in the middle of making, specifically the new insight into the falsifiability issue that discussion here helped me with is one of those changes.

    As for the other issues you mention, I should point out that I have a somewhat different point of view about parts of our discussion. In some cases what to you may have appeared to be a clearing up of misconceptions on my part to me seemed to be just my clarifying some things that I hadn’t written carefully enough, allowing them to be too easily misunderstood or misconstrued. In any case, the book manuscript is written more carefully and at greater length than my web comments, so it shouldn’t have so much of this kind of problem.

    One thing you’ve properly taken me to task for is sometimes attributing to all string theorists views held only by a minority, or at least appearing to do so. To some extent this is hard to avoid. The sheer complexity of the range of different opinions is hard to do justice to in any piece of expository writing about these issues, so one has to oversimplify to some degree. I’m well aware that many if not most string theorists are eminently reasonable people I can agree with about most things, who don’t hold unreasonable or indefensible views. Some of my best friends are string theorists, and I never have trouble talking about the subject with them.

    On the other hand, there are a significant number of string theory partisans out there who seem to me to be unwilling to engage in rational discussion of the issues surrounding string theory, and often engage in the offensive behavior of assuming anyone skeptical about the theory is just stupid and ignorant. I’ve had a lot of this to put up with in the last day or so since publicly announcing my book project. These people are presumably overrepresented in internet forums, and range from fools hiding behind pseudonyms like F. Uckoff, to Harvard junior faculty, to respected senior faculty at major research institutions. I’ve just wasted some of my time trying to respond on Dave Bacon’s blog to Greg Kupferberg, a mathematician string partisan who holds the unshakeable belief that my objections to string theory are of the same sort as Intelligent Designers’ objections to the theory of evolution and that my only motivation is unwillingness to do the hard work necessary to learn string theory. I think Lubos Motl’s comments here and elsewhere speak for themselves.

    So, while I’m willing to believe that the majority of string theorists are reasonable sorts, that’s not so clear from what goes on on the internet, and some of my experiences somedays leave me feeling not especially charitable. While there are certainly some stupid comments left on my weblog by people bashing string theory, I’d like to think that if any of these were coming from serious people in respected positions (e.g. Harvard faculty members), I’d be taking them to task for their behavior and I can’t help noticing that this doesn’t seem to be something any string theorists are willing to do.

    About the landscape: my own view of the issue is extremely simple. Any theorist working on a theory who ends up deciding the theory leads to something that ugly and that unpredictive has to just acknowledge failure and do something different. I understand that there’s a wide range of opinions about this among string theorists, but don’t think this is a subtle issue. The book was largely written in 2002 before the landscape controversy got going, so material about it is kind of added on, and given the way I see this, I haven’t had the interest or energy to go into too much detail about the various issues that people often get into when talking about this.

    Finally, I don’t want to put you on the spot in public, but will soon contact you privately with a proposition about the issues you raise. Maybe you can help me out…

  • Clifford

    Ok Peter, we’ll talk. Just at least try to do a good job of giving a balanced view of the types of activity going on in the field, and the types of people you can meet too! Books last longer than blog comments.



  • Pingback: Not Even Wrong » Blog Archive » Not Even Wrong: The Book()

  • Lee Smolin

    Dear Clifford,

    Can I add that these discussions have been extremely helpful as my own writings on the subject have progressed, especially with regard to the range of views among workers in string theory? So many thanks. I don’t know what Peter will propose but, if I may, could I make two suggestions of ways in which you and others could help myself and others obtain a more balanced view of the range of views among string theorists?

    -If you have the time, some recountings of private discussion illustrating the range of views on crucial issues would be very helpful.

    -Any documentation of the range of views would also be helpful. This could be pointers to papers, talks on line, posts etc by string theorists disagreeing with expressions of views by leading string theorists on crucial issues.

    This would be helpful as one of my projects is a kind of intellectual history, and I am very aware I could have, as you correctly suggested before, a skewed view. I should also emphasize that recently I have sensed things changing as there is strong disagreement among string theorists about what to do concerning the landscape. So I am talking about the past, say before 2002.

    Let me give some examples.

    A number of string theorists have told me privately in the last year that they are happy to believe in the landscape because they never believed in the idea that there was a unique unification leading to unique predictions. But to my knowledge, none of them published their views, at the time. If you can find papers or talks by string theorists prior to 2000 favorable to the now current views on the landscape or expressing doubt that there would be a unique and predictive string theory, I would be most grateful.

    In this case I have some relevant experience. Beginning in 92 I wrote several papers (and a book in 97) proposing that there was a landscape of string theories, that there would be no unique predictability, and that environmental or statistical reasoning would be required to get predictions from string theory. I also gave reasonably many talks about this, some at major string places, a few at conferences attended by some string theorists. So if there was then a spectrum of views on this, I would think I would have heard, and I didn’t. What I recall hearing was many assurances by string theorists not to worry, that there would be a unique vacuum in the end.

    There are other issues on which it would be very good to have evidence of a range of views, such as the status of the major conjectures such as finiteness, S-duality, and AdS/CFT duality. I have done some literature searches, and I am afraid that you would be unhappy with my characterization of what I’ve found. But I would much prefer to believe that you are right, and that there was a range of views, which I somehow miss finding the evidence for. So I would be most grateful to be pointed to papers and talks that illustrate a range of views on these issues. Also useful would be pointers to reviews that give careful, critical and correct statements about what was known and not known at the time about these and other key issues.

    Finally, here is what I would really like to see: we drop completely the distinction between who is “a string theorist” and who is not, and stop worrying about the range of views among “string theorists” and instead just think about all of us as theoretical physicists working on quantum gravity and unification. That way, the views of someone like myself, who has written actually fairly many technical papers about string theory (17) but for some reason is not considered “a string theorist” could be counted when you consider how wide is the range of views in your community. How about it? If all people who worked on quantum gravity thought of themselves as one community, who had, say, common conferences, and research groups, in which people from different research programs felt equally at home, then we would automatically get to be part of a community with a broad range of views. Wouldn’t that be the best thing for all of us, scientifically?

    Again, let me express my thanks for the high quality and openness of the discussion here.



  • fyodor

    cvj said:
    “Ok Peter, we’ll talk. Just at least try to do a good job of giving a balanced view of the types of activity going on in the field, and the types of people you can meet too! Books last longer than blog comments”

    We live in the age of the artificially generated “controversy”, as followers of the ID campaign know all too well. And the favoured weapon of the artificial controversy exponent is to get himself into a debate with a real scientist. Even if he gets clobbered, he can still claim to have been taken seriously enough to be debated, and it is a short step from there to “teaching the controversy”. The notion that string theory is “not even wrong” is no more controversial than the idea that eyes have to be “designed”, and it would be regrettable if the public were given any other impression. However, I’m confident that cvj won’t fall for this, though it may require him to be a bit more blunt than he would like to be.

  • Clifford

    Hi Lee,

    Thanks for the kind comments about the kind of discussion that’s been encouraged and maintained here, so far. I hope it’s worth it.

    I would like to say that I have not elected myself as some sort of barometer of the field’s opinion. I have no special knowledge, and I’m certainly not leading an anti-establishment movement within the field. Further, my knowledge of the literature is not encyclopedic, so I won’t claim to be able to give you long lists of supporters of one view or another. Actually, I don’t think anyone can. All I’ve been saying is that there are a lot of string theorists out there, and a wide range of motivations for working on the subject (an impression you can get by just talking to people in the field at conferences, etc). Alternatively, just pick a random day on hep-th and read the introductory paragraphs. They are not all talking about the landscape. They’re not all talking about uniqueness either. They’re just getting on with working on trying to understand the theory, since most people know that we can stand around talking about which of the two it is all day long, but it won’t actually determine the outcome. We have to do the research, which is the point I keep trying to make, but nobody seems to hear this as they want to see a fight. Nobody knows the answer, so why have a live or die controversy over the result of a guess? This puzzles me, but anyway, let’s move on. So all I’ve been saying to Peter is that it is an exaggerated impression that he often gives by talking about the program of string theory entirely in terms of the landscape. Just as I spoke strongly against your post on the Landscape thread which again gave a very one-dimensional view of what motivates people working in the field.

    I’m amused by your well-motivated (and I hope you don’t mind me saying, ironic) comment about wanting discussions to stop talking in terms of “string theorists” vs everyone else. I agree, it would be good if we could all move beyond that. So let’s try.


  • Jacques Distler

    I’m amused by your well-motivated (and I hope you don’t mind me saying, ironic) comment about wanting discussions to stop talking in terms of “string theorists” vs everyone else. I agree, it would be good if we could all move beyond that. So let’s try.

    Personally, I’ve always referred to myself as a “high energy theorist,” rather than as a “string theorist.” I don’t think it makes sense to label oneself by the particular research program one happens to be pursuing.

    (On those grounds, I might prefer “theoretical physicist,” but that’s a little too vague for most purposes.)

  • Clifford

    A good point Jacques. Actually, I rather like to just think of myself as a physicist. (I have a hope to get a chance to do a useful experiment again one day!).


  • Quantoken

    How about “fundamental physicist“, or maybe “physics fundamentalist“? :-)

    What some considered the biggest success of super string theory, I considered the biggest failure of super string theory, i.e., the derivation of a spin-two zero-rest-mass particle which you call graviton.

    The existance of graviton is very troublesome and that’s exactly where GR and QM are incompatible, or quantum gravity is none-renormalizable. Equivalence Principle requires any mass or energy, must gravitate with each other, no matter how small a quantity they are or how far they separate. If the gravity is exchanged by graviton, it requires every single particle in the universe exchanges lots of gravitons with every other particle in the universe, and of course gravitons themselves have energy and exchange gravitons amounst themselves, too. If you tally up the number of gravitons flying around you get an infinite integer. And certainly nature does not allow anything in infinite quantity.

    Even worse, all those gravitons occupy certain quantum states, if you sum up entropy represented by these quantum states, you easily surpasses the Hawking-Bekenstein entropy bound.

    My opinion, like those of Einstein, is that gravity should simply be explained as a geometric effect of spacetime curvature, NOT a force mediated by bosons.

    Any string theoretist has any thought how to resolve the paradox, and rescue graviton from quantoken? I especially want to hear Lee Smolin’s comment since he has been meantioning the word “information” a lot more in recent years, an indication he may be exploring towards the right direction, though not quite there yet.


  • Richard

    “And certainly nature does not allow anything in infinite quantity.” — Quantoken



  • Peter Woit

    Hi Fyodor Uckoff,

    I’m a bit curious about who it is that is hiding behind anonymity to accuse me of not being a “real scientist”. Whatever you think of my views, I always attach my real name to them and am prepared to take personal responsibility for them. If you want to evaluate my qualifications to comment on string theory or other issues in physics, there are several ways you can easily check up on these. Could you tell us who you are, or at least what your qualifications are to be saying what you are saying?

  • Lee Smolin

    Dear Clifford,

    Thanks, I appreciate of course that the priority is to work and that there is a range of directions and motivations. But just so you understand what I meant when I suggested that the range of viewpoints was too narrow in the past, and don’t take offence, what I was referring to was the range of views that could be found on key unsolved conjectures such as the AdS/CFT conjecture, finiteness, S-duality, the existence of a unique vacuum, the existence and form of M theory, etc. One reason why some people got the impression that there was a narrow range of views is that it was easy to find many papers and reviews just asserting or assuming that these different conjectures are true, while papers that carefully and critically discuss the actual detailed status of the conjectures are rather rare (at least I have found very few.) So someone who wants to think carefully about the status of the different claims has to do quite a lot of searching just to find precise, reliable statements about what is known.

    For example, the different versions of the AdS/CFT conjectures made by Maldacena, Witten, Polyakov et al and others have very different implications for what string theory is and hence what should be worked on. It would then be very helpful if there was somewhere a critical review that explained carefully how the different conjectures differ and discussed which version of the conjecture is best supported by present evidence.

    If some readers of your blog can help to find such papers or talks, on any of these key issues, I think it would be useful for all of us.

    By the way, my proposal is not ironic. I encourage students, postdocs and anyone who asks to work on more than one approach to quantum gravity. I find for myself that the best way to keep an objective view of the field is to work on problems in a variety of approaches.



  • Clifford

    Hi Lee,

    I agree that it is difficult for someone working outside the field to get a good view of what the status of these things might be (or even opinion about the status), by looking through papers which are more concerned with chipping away at their chosen part of the coalface. It is even difficult to do if you’re working in the field too, since you’re occupied with your own chipping away… Now recall that “absence of evidence is not evidence of absence”, and so I say -again- that this does not mean that everyone in the field has one (or a few) monolithic opinion or motivation.

    You have to kind of get steeped into the area of choice, reading the papers, the footnotes, and also listening to the conversations…. This is where you learn that opinion is alive and well, and very varied. From the outside, reporters (and other commentators) just say “String Theorists are thinking…” because it is convenient, and lazy oftentimes.

    More particularly though. What different versions of AdS/CFT? I’m already confused there. Most people I know of in the field have only one version in mind. Maldacena made a suggestion, and it was refined and put on a firm computational footing by Witten, and by Guber, Klebanov and Polyakov (which is what I presume you mean by Polyakov et al). They’re all talking about the same thing, Lee. Am I missing something deep here?


  • Lee Smolin

    Dear Clifford,

    Thanks. I believe that if one reads the actual conjectures stated in the papers of Maldacena, Witten, Polyakov et al one sees right away that the conjectures are different. I disucssed this in detail in my comparative review for the Wheeler conference, hep-th/0303185, section 6.4.3. Rather than repeat the arguments here, let me refer you there. Please keep in mind that that paper is more than 2 years old, and so will be a bit out of date. But that’s not relevent for whether the original conjectures are distinct.

    The key point is that Witten’s conjecture, which I called there conformal induction, falls short of full equivalence, and is hence weaker than Maldacena’s conjecture. I claimed in the paper that as of then (03), Witten’s version explained most of the calculational results, except those that concern relations between BPS states. It was then possible to assert that the evidence was explained by Witten’s conjecture plus the possibility that stronger results hold for BPS states on account of the extended SUSY algebra imposing relations between the spectra of two theories in their BPS sectors that are not true in the full Hilbert spaces. If true, this would mean that there is only a partial relation between string theory on AdS^5 X S^5 annd N=4 SYM. This is still very important, and many results follow from it, but not the claim that “N=4 SYM is proven to give a non-perturbative formulation of a string theory.”

    I would be interested in whether recent results have changed this situation. Certainly there is still no proof of the strong (Maldacena) conjecture.

    This is an example of what I mean by a range of views. It seems to me that given the fact that neither string theory on AdS^5 X S^5 or N=4 SYM have been precisely defined, it is a reasonable possibility that they are not equivalent but that there is a weaker relation between them, such as Witten’s conjecture. This may or may not be true, but I would hope that experts consider it as a possiiblity, untill proven wrong. If it has been proven wrong I would be very happy to be updated and corrected.



  • Clifford

    Thanks Lee. This looks like a rather long conversation to have, because there is a large literature….and this points up why several of us are puzzled by such harsh partisan criticism from your camp (and those of others) when it is clear that the criticisms may well be based on out of date or just wrong information. I will try, but this is going to be brief, as I have to work on several other things.

    Now I could be just naive here, and you’re getting at something deeper, so I apologize for the above remark if I am wrong. But it seems to me that you’re just talking about different strengths of the conjecture, one being stringy and one being more restricted to supergravity. As for BPS vs non-BPS, I suggest you have another look at Ed’s paper. That beautiful work very much applies to a much larger setting than just BPS! …. Anyway lot of work has been done of very very stringy tests of the conjecture, beyond just BPS states. For explicitly stringy stuff and non-BPS stuff I mention just for starters Berenstien, Maladacena, Nastase, Gubser, Klebanov and Polyakov, and all the numerous papers that cite them (go to SPIRES to get that). But even without that, several papers have been written (years ago now) which strongly suggest that the full stringy completion of the correspondence is true (the key role of branes, and very stringy effects -such as the giant graviton story, the myers effect, etc…yes, myers – your colleague down the hall?…etc).

    There is no proof, but I don’t think you need to get hung up on that. There simply will be no rigourous proofs of any strong-weak coupling dualities or others in that spirit untill we find techniques that go well beyond the current formulations of the theory (no need to start on about background independence here…I get it….). But that’s ok. One point of view is that these dualities themselves are strong hints as to the shape of what we’re looking for.

    There are no firm signs (a.f.a.i.k.) of anything being wrong with the various dualities you mentioned. This is not a proof, I’m just informing you as to the state of affairs.


  • Wolfgang

    am I confused here ?

    Lee wrote: [..] Witten’s version explained most of the calculational results, except those that concern relations between BPS states. [..]

    cvj wrote: [..] I suggest you have another look at Ed’s paper. That beautiful work very much applies to a much larger setting than just BPS! [..]

  • Wolfgang

    Sorry for posting twice in a row …

    cvj wrote: [..] There simply will be no rigourous proofs of any strong-weak coupling dualities or others [..]

    Since there is not much empirical evidence to keep one on the right track, there is a real danger that this will get you into some sloppy reasoning. Not everybody submitting to hep-th is Ed Witten.

    As Lubos Motl once wrote on his blog: “if you cannot disprove the conjecture in 20 minutes, it is probably correct …”

  • Clifford

    No, I’m confused. I freely admit that I misread Lee’s message on the BPS part. Thanks Wolfgang. Sorry Lee.

    Most of my message is not about that, though, and still stands.



  • Clifford

    Wolfgang, you wrote:

    Since there is not much empirical evidence to keep one on the right track, there is a real danger that this will get you into some sloppy reasoning. Not everybody submitting to hep-th is Ed Witten.

    Several people in the field are aware of this valuable insight, and mindful of it.
    You can at this point start the tedious discussion we’ve had several times on several threads about how this sort of thing means we are doomed because there is no experiment to test the idea….. but I do not think it will help us to repeat all of that (interested parties can go and read those threads). We are aware, as a field, of the limitations of what we are doing. The only other option is to do nothing further in this area, which as I have said countless times, would be at best neglectful. There are others working on other approaches, and I wish them well.



  • Wolfgang


    > You can at this point start the tedious discussion we’ve had several times on several threads about how this sort of thing means we are doomed

    This was not my point and I am sorry if my comment made you angry.
    My point is that more mathematical rigor is necessary if the experimental guide is missing. I do not think that M-theory/string theory is doomed. At least I hope not.

    As fas as I understand, this is also Lee Smolin’s point, but I cannot and do not want to speak for him.

  • Clifford


    You did not make me angry. “You wouldn’t like me when I’m angry”… to quote a famous fictional scientist … ๐Ÿ˜‰

    I know what you’re saying and it has been discussed before. My response is that much more research is needed on what we are actually dealing with here as a theory before we can rigourously prove anything mathematically. See my penultimate paragraph to Lee…I said more research is needed….as I do in every thread on this blog when this issue comes up. But people seem to forget…..that’s ok. That’s why we’re here, I suppose.

    And while we are on the subject, there are several powerful pieces of physics that have no rigourous mathematical proofs. Rather, new interesting mathematics often flows from the physics – less so the other way around, at least in this field. (Others in the field probably have a different take on that, by the way…I speak only for cvj, not the whole field….let’s not start another “string theorists say…” controversy here…. :-) )

    Off to the beach to do some computations of ironic experimentally unverifiable stuff now….



  • Clifford

    In case people are wondering what I’m talking about in the third paragraph of my comment above, I’ll just mention the whole framework of Quantum Field Theory, and mention that there are several examples within that framework alone of what I’m talking about. Several results from that field were proposed and studied in the looser physics setting and seen to spectacularly relate to Nature -eventually..these things take time- without us sitting around worrying about whether the Mathematicians had vetted them yet. I see the same things aas possible here… will take time…..but the payoffs are worth the wait, in my opinion.

    I hope we’re all agreed that QFT is a useful box of physics tricks that is hard to motivate from rigourous mathematical perspectives alone.


  • Levi

    Very interesting discussion here…

    Clifford: a.f.a.i.k.? As far as I know? You may be carrying shorthand too far, but t.i.a.v.n.b.a. (this is a very nice blog anyway).

  • Wolfgang

    > I hope we’re all agreed that QFT is a useful box of physics tricks that is hard to motivate from rigourous mathematical perspectives alone.

    Yes, therefore we are so happy that many high-precision experiments support it. ๐Ÿ˜Ž

  • Lee Smolin

    Hi Clifford,

    Thanks very much for your answer, but it does not address the claims made in the paper I referred you to, hep-th/0303185. My claim there 1) is mostly not about holding at the supergravity vrs the stringy level, 2) is not contradicted by the Berenstien, Maladacena, Nastase and followup results on pp waves, which were discussed and cited in my paper (ref 164) and 3) is also not contradicted by Gupser, Klebanov and Polyakov, which was likewise cited (ref 155) and discussed. It does concern “different strengths of the conjecture” which are allowed by present evidence, even at the “stringy level”. Given that only the strongest requires complete equivalence of the string and gauge theory, I hope you will agree that it matters very much which of them is true.

    So before you get excited and call my remarks “harsh partisan criticism” I would very kindly ask you to please read what I actually wrote. And please, I am not making the claim that there is no AdS/CFT correspondence, and I am not being “partisan”. Nor is my understanding of this field “out of date or [based on] just wrong information.” I read the literature carefully (something not everyone does), I go to talks and conferences to keep up and I aim when I can to contribute constructively to research in string theory. I would hope that carefully parsing out the different possible conjectures consistent with present evidence would be seen as a useful contribution to research. In this case, I would insist that the point is not whether there is or is not piles of evideence for some form of a string/gauge duality. The point is precisely what is true about that duality.

    In fact you seem to agree with my main point, as you say that there “will be no rigorous proofs of any strong-weak coupling dualities or others in that spirit until we find techniques that go well beyond the current formulations of the theory”. How can you assert this and not be interested in the question of which precise conjecture will turn out to be true? And, not to attack you personally, but to try to make here constructively a point I’ve been trying to make, when you call someone who studies the literature carefully and makes a constructive contribution based on it a “harsh partisan critic”, you are not showing evidence of an openness to a range of viewpoints within the field.

    I also agree with your main message that people should not criticize ignorantly and that we should get on with research as that is the only way to find out what is true. But I do think that carefully and critically setting out the range of possible outcomes of that research is an important part of succeeding at that research. In a case like this, where there are several distinct conjectures all consistent with present evidence, it helps to be aware of that fact, as different strategies may be required to attempt to prove different versions of the conjecture. Furthermore logic dictates that only the weakest conjecture consistent with the present evidence can be considered to be supported by that evidence.

    Here is one reason it matters. There are as well different forms of the holographic conjecture (see hep-th/000305 for a review of them). It matters for which version of holography is realized in string theory which version of the AdS/CFT conjecture is true. Some people use the strong form of the AdS/CFT conjecture to argue for a strong form of holography, one that would exclude bouncing black hole singularities and baby universes. Since there is evidence in non-perturbative quantum gravity calculations for elimination of black hole singularities there is a possible contradiction here with the strong form of AdS/CFT. But there is no contradiction with a weaker form, such as Witten’s conformal induction conjecture, as that only requires that certain observables (those that can be measured from infinity) of the string theory are represented in expectation values of the gauge theory. This is, by the way, why the talk of Liu at the Toronto string conference was so interesting, because it was an indication that the gauge theory might be able to probe what happens to the singularity.

    If someone just believes that the strong form of AdS/CFT is true, they may miss the possibility that string theory could lead to baby universes. So what you believe about open problems does determine the direcction of research you are willing to pursue. This is why I believe it is important to have careful, critical discussion within a field, and to have a field be friendly to the widest range of views consistent with the actual results.



  • Plato

    Clifford, you continue to provide a nice framework for discussion here. Thanks

    While one might not want to be the barometer, it does allow for a point of view to materialize(nurturing the creation of ideas). For others, to come through and watch this process.

    How would one move constructively from what has been offered? The next step?

  • Greg Kuperberg

    Since Peter Woit mentioned me, I might as well run my impressions by the string theorists and others here. As I said on David Bacon’s blog, I’m not a string theorist, in fact I will probably never be a string theorist, but string theory does look interesting to me.

    I also haven’t read Peter Woit’s book, of course, but I did read physics/0102051. An excerpt:

    To the extent that the conceptual structure of string theory is understood, the Dirac operator and gauge fields are not fundamental, but are artifacts of the low energy limit. The Standard Model is dramatically more “elegant” and “beautiful” than string theory in that its crucial concepts are among the deepest and most powerful in modern mathematics. String theorists are asking mathematicians to believe in the existence of some wonderful new mathematics completely unknown to them involving concepts deeper than that of a connection or a Dirac operator.

    The way that I read this is “I view gauge field theory as an intellectual pinnacle, therefore I don’t want string theory to surpass it.” I realize that this is harsh, but am I really wrong?

    Another excerpt:

    Graduate students, postdocs and untenured junior faculty interested in physics beyond the Standard Model are under tremendous pressures in a brutal job market to work on the latest fad in string theory, especially if they are interested in speculative and mathematical research. For them, the idea of starting to work on an untested new idea that may very well fail looks a lot like a quick route to professional suicide. Many physics researchers do not believe in string theory but work on it anyway. They are often intimidated intellectually by the fact that some leading string theorists are undeniably geniuses, and professionally by the desire to have a job, get grants, go to conferences and generally have an intellectual community in which to participate.

    It smacks of professional jealousy, if you ask me. Moreover, the idea that we need to save the children from string theory is surely offensive. Somehow I doubt that the string theory job market is so easy compared to the rest of physics. In light of passages like this, I don’t really know what Sean means when he says that it’s a healthy disagreement.

  • Clifford


    I was talking about partisan criticism in other contexts, not this particular discussion of AdS/CFT. Anyway, I’d like to point out that I did say in a preface to my remarks that if I misunderstood what you were getting at, then I apologize for any negative characterization. So, I seem to have misunderstood what you’re getting at, and so I apologize.

    So since the points I made seem to be irrelevant to what you’re getting at, I’ll step back and let someone else have a go at addressing your concerns. I’m not seeing the two sharp forms of the conjecture that you’re seeing. Moreover, you may well be as up to date on things as I am, possibly more, in which case I may be of no use to you whatsoever.



  • Peter Woit


    From our previous discussion you clearly have no interest at all in my actual views, instead you adopt the pathetic tactic I’ve become all too familiar with from string theory fanatics of making up idiotic interpretations of what I write instead of even bothering to try and understand what I have to say. No, I don’t think anything like “I view gauge field theory as an intellectual pinnacle, therefore I don’t want string theory to surpass it” and never have said anything like that. I do think gauge field theory is very deep mathematics and physics. It is not a pinnacle but something we need to better understand and ultimately surpass. I recommended to you to take a look at hep-th/0206135 but I doubt you’ve bothered. It contains a detailed speculative outline of mathematical ideas that I think surpass those of gauge field theory and which I personally think are a promising way forwards to better understand the standard model QFT and find a way to improve on it. Maybe I’m wrong, but if you want to argue with my views on mathematics and physics, that paper is what you have to argue with, not some stupid statement you make up.

    I can explain a bit more about the paragraph you quote since you’re a trained mathematician and should more easily than most physicists be able to see what I’m getting at if you’re willing to pay attention. I just spent a year teaching our graduate geometry class here at Columbia. There wasn’t a textbook, but during the first semester I was largely following Kobayshi and Nomizu, volume I. I share their point of view that perhaps the most fundamental construction in geometry is that of a connection on a principal bundle (aka a gauge field), and the implications of this idea takes up most of the book.

    During the second semester I first covered aspects of Riemannian, symplectic, Kahler and spin geometry. In some sense spin geometry is the most fundamental, since you can build arbitrary sorts of tensors out of spinors, while you can’t construct spinors out of tensors. Finally, in the last part of the course, I was discussing Hodge theory and the various sorts of geometrically defined elliptic complexes that make up the basic examples that motivated the Atiyah-Singer index theorem. In this story, the fundamental role is played by the Dirac operator.

    You may have a different point of view on differential geometry, but I think the one I took in the course is a very modern and very powerful one. It followed excellent texts by mathematicians who developed these ideas for purely mathematical reasons, especially Kobayashi and Nomizu’s book, as well as the more recent book on Spin Geometry by Lawson and Michelson. This point of view puts connections, spinors and the Dirac operator in central roles in modern geometry, and I happen to think it is a very deep and amazing fact that the same constructs are among the fundamental constructs of the standard model.

    The story of the geometry behind string theory is a complicated one, but for the part of string theory which is well-understood the main parts of the construction involve Riemann surfaces, which are truly fundamental mathematical objects, but also complex Calabi-Yau 3-folds, which really aren’t. Calabi-Yaus are mathematically quite specialized and complicated gadgets, which before string theory few mathematicians took an interest in. No one was teaching about them in first year graduate geometry classes (and if they are now, it would be a bit perverse). The perturbative string theory philosophy for getting the standard model is that the things I was talking about in my graduate geometry course are not fundamental, but what is fundamental is the superstring (a mathematically quite complicated and non-obvious construction) and the Calabi-Yau (also a mathematically rather complicated construction). The connections=gauge fields of my graduate course are not fundamental in this picture, but only dominate in the low-energy limit.

    We suffer from having no experimental guidance at all about how to get beyond the standard model. Personally my point of view is that in such a circumstance the best thing to do is to remember that deep mathematics and deep physics have traditionally gone hand in hand. The mathematical structures of the standard model are about the deepest ones we know of in the modern approach to geometry. Those of the superstring and Calabi-Yau aren’t. This seems to me to be evidence that the superstring idea of how to get beyond the standard model is misguided from the point of view of mathematical aesthetics. The fact that it has utterly failed to lead to a framework in which one can make predictions about the real world seems to me to confirm the argument from aesthetics that it is the wrong way to go.

    String theorists like Clifford will reasonably object that, for such an aesthetic argument, what is really important is not the approximate perturbative string theory, but the unknown non-perturbative theory they suspect exists, and for which they have unearthed all sorts of tantalizing evidence, including relations to the gauge field theories I am taking as fundamental. Maybe he’s right and there is some more wonderful underlying thing to be found which involves depths of fundamental physical and mathematical beauty still unkonwn, with the geometry from my graduate course only coming out in some limit. No one has actually produced this yet, and I’m skeptical of its existence. Or more accurately, I think there are still unknown very deep mathematical and physical structures to be found, I just believe they will be closer to the ones of the standard model than to the ones of string theory found so far.

  • Peter Woit

    About the second part of your comments, again you’re making up something I didn’t say. I never said “the string theory job market is so easy compared to the rest of physics”. What I said is that the particle theory job market in general is a brutal one, and this is very true. String theorist or not, it is very difficult to get a permanent job in this business. If you want to have any real chance at it at all, you need to be working in what is perceived as a really active area of research where a lot is happening. In practice these days, that means your chances, while slim, are best if you’re doing string theory, phenomenology, or cosmology (best if you can do all at once!). If you happen to believe that string theory is misguided, but also believe that sophisticated mathematics is the thing needed for progress in particle theory, you’re pretty much completely out of luck. I don’t know what physics department these days is going to hire you. If, like me, you believe that our best hope for progress is smart, ambitious young people working on mathematically sophisticated ideas about how to extend the standard model, you would find this current job market problematic.

    As for your idea that I’m suffering from professional jealousy, let me point out you don’t know the first thing about me personally, and as far as I know we have never met. If you knew anything about me, you would know that, far from being bitter and disappointed in my career, I feel quite the opposite. When I went into particle theory I assumed that I’d most likely end up struggling to get a job involving too much teaching at a not so great educational institution in some place I really didn’t want to live. Instead I’ve ended up very happily with a permanent position living in my favorite city in the world, part of a great mathematics department with wonderful colleagues I enjoy learning from and interacting with and who have treated me exceedingly well. In recent years I’ve been able to teach pretty much whatever I felt like and have learned a great deal from this. I’ll also point out that due to a clever choice of parents and dumb luck in the real estate market I am financially extremely well off. I could quit my job tomorrow and never work again if I felt like it. I do what I do every day because I love it, not because I have to. Few people in life have been blessed with the good luck I have.

    Your comparison of my criticism of string theory to the Intelligent Design criticism of evolution and conviction that my only motivation is unwillingness to learn string theory are both massively stupid and deeply disgraceful. You should be ashamed of yourself. I really don’t understand what it is about string theory that leads people to behave in this way.

  • Greg Kuperberg

    No, I did look at hep-th/0206135. Sections 1-6 are a competent summary of a grab bag of interesting mathematics. Sections 8 and 9 are a competent (I suppose) outline of quantum field theory in 1 and 2 dimensions. Sections 7 and 10 are some speculative remarks that I don’t understand. (Maybe someone else here can help me out?) Finally section 11 is a denunciation of string theory and supersymmetry, a short version of physics/0102051.

    In regard to the above remarks, the words “deep” and “beautiful” are often euphemisms for “math that I learned”, while words like “complicated” and “gadgets” are often euphemisms for “math that I don’t want to learn”. A lot of people talk this way. It is a plausible interpretation of the above comments, because otherwise I would have no idea why Riemann surfaces are “truly fundamental” while superstrings are “quite complicated and non-obvious”. I don’t see that passing from Riemann surfaces to super-Riemann surfaces is a great leap, much less a downward plunge.

  • Peter Woit


    Let’s see, you don’t understand string theory and you don’t understand my speculative ideas, but you feel quite comfortable attacking me anyway. Isn’t there something wrong with this picture?

    You still seem to suffer from this obsession that my objection to superstring theory is that “I don’t want to learn” it. I suspect I know far more about the construction of the superstring and Calabi-Yaus than you ever will. Unless you have some actual evidence that my views are based on not wanting to learn something, you should stop making this kind of offensive personal argument.

    Any discussion of what is “deep” or “beautiful” mathematics suffers from the problem of “de gustibus…”, but your comments lead me to strongly suspect you don’t understand the construction of the superstring (there’s much more to it than passing from Riemann to super-Riemann).

  • Greg Kuperberg

    I spent some time with Polchinski (or rather, his textbook) to better learn what a superstring is. I am not sure that discussions in blog comments are intellectually healthy, but reading Polchinski certainly is.

    In the simplest definition, the ordinary bosonic string is not even as complicated as a Riemann surface. It’s really just a topological surface, and its action functional is just its area as a submanifold of spacetime. This is equivalent to the Polyakov functional for a conformal surface. As I said, whether or not it is “deep” depends on whether or not you want to learn it. Certainly some mathematicians have seen before, because in mathematics, the same two functionals are familiar as the Plateau problem. In some treatments of the Plateau problem, the Nambu-Goto action is called “area” (duh), while the Polyakov action is called “energy”.

    Is a superstring anything more than a superconformal surface? If it’s a Type II superstring, not really. Admittedly consistent Type I superstrings and heterotic strings have some decorations that seem ad hoc to naive little me. (For example, that Type I superstrings have to be open and have SO(32) Chan-Paton decorations.) But the action for the Type II superstring (either IIA or IIB) seems like a natural superization of the action for the bosonic string, which is either nothing more than or not even as much as a conformal surface.

    So if you think that conformal surfaces are beautiful but Type II superstrings are ugly, that’s like saying that Audrey Hepburn is beautiful but Eliza Doolittle is ugly. You might say that if you don’t want to learn about Eliza Doolittle.

  • Lee Smolin

    Hi Clifford,

    Thank you for kind remarks, they are very much appreciated. They illustrates one reason you are respected widely not only inside the string community, but outside of it as well. I am happy also to acknowledge that the main points you have been arguing for are correct: there are critics of string theory who are not as informed as they should be, and they do under estimate the breadth and variety of views and approaches within the string community.

    But I would ask to impose just a bit more on your hospitality here to say something I think is terribly important, that exchanges on this and other blogs have illustrated. Let us call the range of views that can be heard from people who live in the string community R. My observation is that, however wide this is, it is a proper subset of another set P, which is the possible range of views that could reasonably be taken on various issues, which are as well supported by the evidence from calculations and from nature as are the views in R. I will call the views in P but outside R, O. I also observe that among those who are considered outsiders by string theorists, there are some who are in fact familiar with many of the technical details, and who could, and in some cases do, contribute research to string theory. Let me call such people the “competent outsiders”. Now here is what I think is so important:

    -One draws different conclusions about the possible futures for string theory, given views in O than in R. Thus, it matters a lot for physics, which views are correct. The importance of the competent outsiders is that they sometimes hold views in O.

    -Occasionally a view is moved from O to R. One illustration of this was the significance of 11 dimensional supergravity and the 11d supermembrane for string theory, which was moved from O to R in 95. Another example is the view that the connection between string theory and nature will involve a landscape of equally possible theories rather than a unique, single theory, which was moved from O to R in the last few years. In these and other cases there were competent outsiders who had been insisting on the importance of these views, that were not listened to by insiders.

    -These examples illustrate the importance of O, as it can have a big effect on the direction of research when a view is moved from O to R.

    -Nevertheless, at any one time, insiders are often not aware of the views currently in O or, if they are, they do not accord them much interest. Even when a view is moved from O to R, some insiders think of it as a new invention and do not appreciate that the view has been held for some time by competent outsiders.

    -There remain views in O that may still turn out to be important. Among these I would put the fact that there are a range of possible versions of the AdS/CFT conjecture allowed by the evidence and the view that string theory cannot succeed unless a truly background independent formulation is found.

    So, while I agree that the breadth of views in R, held by insiders is sometimes unappreciated, I hope you can appreciate that some outsiders have a valid point when they remark that the range of views allowed by the evidence is wider than that usually heard within the string community.

    These stories raise several questions for me, that I am thinking about as I try to write a kind of intellectcual history of the subject. First, wouldn’t it have been better if the views I mentioned that moved from O to R had been all the time included within the range of views discussed and considered by insiders? Does this imply that progress would be faster if the range of views considered seriously by string theorists were broadened?

    There are also some general questions about how science works. Is the situation I’ve described common, or is it special to string theory? Is R always narrower than P, and why? Is there commonly a class of competent outsiders? Related to this, why is R at any one time narrower than P? And why are insiders sometimes not aware of, or dismissive of views in O? Are strong divisions between research programs, such as we see between the different approaches to quantum gravity, generally good or bad for the progress of science?



  • Peter


    “the ordinary bosonic string is not even as complicated as a Riemann surface. It’s really just a topological surface…”

    “Is a superstring anything more than a superconformal surface? If it’s a Type II superstring, not really.”

    Well, I no longer suspect that you don’t know what a superstring is, now I’m sure. From what you wrote, you don’t even seem to know what a bosonic string is. What you’ve written is pretty confused, but for one thing you don’t seem to be aware that the subject is about quantized, not classical strings, and there’s a lot more to them than writing down an action functional.

    Whatever time you’ve spent with Polchinski, it seems to have left you with:

    1. Not much of an idea about what either a string or superstring actually is.

    2. The conviction that, even though some constructions appear ad hoc to you, they really aren’t for some unknown reason.

    3. An incredibly ignorant and arrogant attitude, using the small amount of misinformation you’ve picked up to justify personally attacking people people who actually understand things you don’t as “unwilling to learn”.

    Reading Polchinski doesn’t seem to have been an intellectually healthy experience for you, quite the opposite.

    I’ve had some bizarre exchanges with string theory partisans over the years, but this one really takes the cake.

  • Plato

    From being an O, it is important that the physics is brought into perspective with R, yet how would such an adventourous mind of those in O appeal to the strict formulations in R?

    The people in O had to have some understanding as Lee states, of the concepts formulated by the degrees lead too, by seeing such Physics correlations of R in O?

    Now such historical correlations and derivatives of string/M theory needed the valuation of R to develope views in O. Although R is strictly mathematical(?) the generalization moved to concept developement arising from R, moved O to consider such physics correlations. They had to go hand in hand, and reval the precipice of change such a model might institue in how we percieve in O.

    I hope this makes sense.:)

  • Greg Kuperberg

    Yes, I know that string theory is a quantum theory. Although I would only be a beginning student of string theory, I do know quantum mechanics. In quantum mechanics, you don’t really quantize objects, you quantize their dynamics. Also, since quantum mechanics is true, all dynamics in the universe are quantized. So a string in string theory is not logically any more or less quantum than a shoelace. Of course its dynamics are much more quantum, but only because it’s much smaller.

    Well, you could call the fermionic part of a superstring inherently quantum, in the sense that fermionic fields are inherently quantum. But you could also call fermionic fields “anti-classical” rather than quantum, given that bosonic fields (absent quantum dynamics) are classical.

    Moreover, there really isn’t any more to quantum mechanics than “writing down” an action, or more precisely, solving it. The laws of perturbative superstring theory really are just a superization of the area functional, as in the Plateau problem. (Granted, the usual Plateau problem is in Euclidean space, but I can excuse physicists for preferring Minkowski space.) All of the complicated stuff comes from trying to solve the dynamics of this simple action. It seems that some people are confusing laws with solutions in this business.

    When I said that some constructions seemed ad hoc, I was careful to explain that I wasn’t referring to type II superstrings, which don’t seem at all ad hoc to me. As for the other three superstring theories, they don’t seem very ad hoc at first glance, just more than zero. One side of them that certainly isn’t ad hoc is that they are chosen for their logical consistency.

    I admit that I don’t really know string theory, but on the other hand no string theorist has disputed my comments.

    Finally, Peter, I got the idea that you don’t want to learn string theory from your own words, not from any string theorist. If you don’t like that inference, then you have misrepresented yourself.

  • Peter Woit


    You’re just parading a truly massive amount of ignorance and making it crystal clear that you have no idea of what you are talking about here. From statements like

    “there really isn’t any more to quantum mechanics than “writing down” an action, or more precisely, solving it”

    you make clear that you not only don’t understand string theory, you also don’t understand quantum mechanics.

    The idea that “I don’t want to learn string theory” is something you made up for yourself out of arrogance and ignorance. On Dave Bacon’s blog I wrote:

    “I have a Ph.D. in particle theory from Princeton, and have devoted much of the last twenty years to learning as much as I can about the subject, including a great deal of string theory. Arguably I’ve spent far too much of these twenty years trying to learn about string theory, instead of spendng my time on something more fruitful.”

    Let me repeat this in simpler terms: I have had about the best high-level training in particle theory you can get in this world, and then have spent twenty years professionally working in this area, including spending a sizable fraction of that time learning about string theory. You admit you don’t know much about the subject, and make this clear with everything you write. And yet, you are willing to attack my competence to discuss this subject. What is wrong with you? What is wrong with this subject that its partisans behave like this?

  • Clifford

    Peter, Greg. Chill Dudes! Or take it outside. Not in my house, ok?



  • Peter Woit


    Hey, it’s not me who’s coming into your place and launching personal attacks on anyone. I’m getting the impression that the way much of the string theory community intends to deal with my arguments is to not answer them, but to engage in personal, ad hominem attacks on my competence. If that’s not something you want to be part of, you might want to consider telling people who do this they’re out of line. Otherwise, they’ll hear from me….

  • Clifford


    You have every right to defend yourself, and you’ve done so (and quite well). I don’t think anyone is under any illusions….. so now I think you can safely move on.
    Honor is restored on both sides….. let’s call a truce. Then we can all come back and fight the good fight another day. ๐Ÿ˜‰



  • Pingback: Not Even Wrong » Blog Archive » Quantum Gravity Commentary()

  • Pingback: Not Even Wrong » Blog Archive » Comedy()

  • David Foster R.G. ie{regular guy}as i have no degree lol

    thanks Pete its so nice to hear that somebody is still trying to find new ways to find out what we may never know …..but in every different idea the seed of truth may be sowed. i hope n pray that men such as yourself are reading your interview in Discover this month ……and starting to THINK…..thanks fer yer time rock on dude!

  • David Foster R.G. ie{regular guy}as i have no degree lol

    P.S> tell Clifford hes an ass. the guy argues fer a whole page and then tries to make up?…..camon cliff you may be surprised when CERN comes on line in 2007…..we may all thank Peter for being the voice of reason….we just dont know yet fer sure …do you cliff?

  • Pingback: Crackpots, contrarians, and the free market of ideas | Cosmic Variance()

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


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