The Templeton Foundation has recently been sponsoring a series of Bloggingheads diavlogs, under the name Percontations. This week’s episode is Fiddling With the Knobs of the Universe, and it has cosmologist Anthony Aguirre and string theorist Clifford Johnson doing their best to hype string theory and the landscape. Critics are dismissed as people who believe obviously wrong things like “if it’s statistical it’s not science”.

Johnson argues that string theory landscape research is just like any other kind of science, capable of making testable statistical predictions, predictions based on generic properties of the theory (e.g. T-duality), and predictions of some parameters based upon fixing others by observation. He neglects to mention that decades of work by people trying to do such things have shown that there are very solid reasons why they don’t work. Not only have no predictions come out of this, but the reasons why have become clear.

While hyping the landscape, he acknowledges that string theory has had a problem with hype in the past. “We all bought into it to some extent” that string theory was going to give the Standard Model, and it was bad that this was promoted in the press as a polished, definitive story of how the world works. He claims to be happy that this has been backed away from in the last several years (although he never seems to have been happy about the existence of string theory critics who have raised the issue of the problems publicly).

In a recent posting, Johnson partially resolves a mystery I’d always wondered about, that of why he left Cosmic Variance. He explains that one reason was that Cosmic Variance was taking “the obnoxious route of calling someone an idiot or stupid for their religious beliefs at the outset.” Bloggingheads has recently featured Sean Carroll and Mark Trodden of Cosmic Variance also discussing cosmology and the multiverse (see here and here), but these episodes were not sponsored by Templeton.

This entry was posted in Multiverse Mania. Bookmark the permalink.

35 Responses to Percontations

  1. NonScientist says:

    “He neglects to mention that decades of work by people trying to do such things have shown that there are very solid reasons why they don’t work. Not only have no predictions come out of this, but the reasons why have become clear.”

    This may be an omission, but I don’t think it’s a fault. How many theories or other scientific discoveries have come to light after such a small quantum of time as decades were not fruitful?

    Don’t confuse this as a slight to, or promotion of string theory. I’m more speaking about Johnson’s desire to promote his research.

  2. Marcus says:

    Was “Bloggingheads diavlogs”
    a diavolical Sleudian Frip?

  3. Peter Woit says:

    No Marcus, “diavlog” is the word the bloggingheads people came up with to describe these things.


    I think promoting one’s research by going to the public and making an argument for it that neglects to mention there are well-known and well-understood reasons why this argument doesn’t work is not just an omission, but a fault.

  4. Phil says:


    Diavlog is a second-order blend, by the way: it blends dialog and vlog, with the latter element representing a blend of video and blog. Or make that third-order, since blog blends Web and log.

  5. Marcus says:


    that means thanks very much.

  6. M says:

    I would say that the main problem was not the hype, but that hyping the Unique Theory with the Unique Vacuum, string theorists collectively misunderstood the physics of strings. Being very smart in formal stuff and doing good physics are two different things.

  7. alex says:


    that means the last paragraph is missing a closing quote (for the one that opens “the obnoxious… )

  8. Peter Woit says:

    Thanks alex, fixed.

    I’m learning all sorts of new words….

  9. Tim vB says:

    Just listened to the “Fiddling With the Knobs of the Universe” episode.
    Another interesting remark of Clifford Johnson is that one should not expect anymore that string theory will explain all input parameters of the standard model
    (that would mean that one’s concept of what a scientific theory should be like or could do for you is too narrow), or that string theory will be a theory of everything (because it is no longer clear what that would even mean).
    Seems to be quite a strategic withdrawal.

  10. per says:

    Well, I think Clifford deserves credit if he left Cosmicvariance for the reasons states in his post. Starting out a discussion with arrogance usually does not lead down a fruitful path (which I am sure you Peter are aware of after all the pro / con string theory discussions)


  11. Peter Woit says:


    At this point, I think just about all string theorists think it unlikely that “string theory will explain all input parameters of the standard model”.

    But this is really a misleading way of putting things. All evidence is that if the landscape picture is correct, string theory not only can’t explain “all” parameters of the standard model, but it can’t explain a single one. It’s completely vacuous, exactly what you would expect of a wrong idea.

    In dealing with the public, string theory proponents often use this misleading tactic, referring to string theory’s problems with predicting everything, when the problem is that it predicts nothing.

  12. Tim vB says:

    Hello Peter,
    agreed, Clifford uses a very interesting multiple tactics approach that clearly shows that he has learned some politics during the string wars:
    – repetition of well known hype arguments (“strings are beautiful, string theory is the only game in town”),
    – references to tremendous successes of the theory and of (verified?) predictions (maybe I missed the point but I think he does not explain which ones),
    – a nearly complete retreat from the original research program (“you cannot expect that string theory has to explain anything, e.g. the electron mass, and fits on a T-shirt”,
    “string theory is just a tool that can be of use in many different contexts”),
    – and the promise that the string theory hive is still open to new ideas, as any scientists should be (“as soon as something more promising shows up, we will stop the string theory business and think about that”).
    I’ve always wondered if the “only game in town” is some sort of insider joke, because according to the original anecdote the player that plays that game loses all the time…

  13. DB says:

    Propaganda was a major weapon in the arsenal of String Theorists during their period of advance, as they acquired funding and tenured positions. It’s just as important during their retreat, to preserve funding and protect the careers of postdocs and non-tenured staff as far as possible.
    They are masters of cynical media manipulation and spin.
    The long-term damage to the discipline of theoretical physics is hard to quantify but is substantial.

  14. Nigel Cook says:

    I think that the string theorists are able to ride over critics easily because they have got the big advantage of a critical mass. They have each other to discuss ideas with. Whenever you have a group of people all enthusiastic about something, they can’t resist talking the easy road to ignore critics.

    E.g., the ‘only game in town’ claim is infuriating because it’s untrue. But when you point out that it’s untrue, they always (without any shame) shift to claiming that it’s the ‘best claim in town’ by arguing that it’s had more people, money, and time spent working on it than all other speculative ideas.

  15. When I was researching my upcoming book, String Theory for Dummies, I found the landscape to be one of the hardest things to deal with logically. I finally used an analogy of the “texas sharpshooter” – he shoots at a wall and then draws the target around where the bullets landed. You basically look around where you’re at and then redefine that position as the goal of what you were working toward. This doesn’t necessarily mean he’s a bad shot, but it certainly makes him look like a far better shot than he is in reality.

  16. Chris W. says:

    Despite or rather because of its accuracy, I don’t think string theorists will be too fond of that analogy, Andrew. Good luck…

  17. nbutsomebody says:

    Clifford is completely vague in what he is saying. The information content is zero but the comedic content is high. is it a stand up (phone up maybe) comedy !

    “String theory or method of string theory” (humm subtle, a person of finer articulation)
    “It is telling us to grow up” (have I heard it correct, hope not ! )
    “Interesting to regime of physics … blah blah” (zzz, sleepy),
    String theory USEFUL ( 🙂 )

    well, what is going on?

    Anyway such a bad performance by a string theorist is really concerning.

  18. chris says:

    Tim vB,

    string theory indeed was tremendously successful recently. i actually think it is close to a breakthrough right now. give it just a few more years and you will see, that it is a fabulous phenomenological model for strongly coupled electron systems. and maybe – in some more distant future – it will come close to lattice gauge theory in describing the IR part of QCD.

  19. Tim vB says:

    allow me to be a bit naive and demagogic here: If you could replace the standard model lagrangian
    with, say, the nambu-goto lagrangian of naive string theory with one free parameter
    (e.g. the string tension), that would be beautiful. I would be very grateful if I would
    not have to memorize the standard model lagrangian any more.
    So let’s assume for the moment that the impression “string theory is beautiful” stems from this observation.
    If you give up the hope that string theory will ever explain resp. replace the standard model, why is it still beautiful?
    This is one of the main points I do not get, I mean: Clifford cites many of the arguments pro string theory
    that originated from the hope that the standard model could be explained by it, while (nearly) giving up all the hope that this could ever happen.

  20. it works! says:

    I don’t know whether you guys have come across this claim, but one string theorist told me he did string theory “because it works.” I don’t know what he meant by that, but I kept quiet just for the sake of politeness….

  21. Peter Woit says:


    That’s not really news, just a good explanation of the complexity of the LHC startup schedule.

    Latest estimate I’ve seen has first beam injection around Nov. 18, with sector 67 the last one to be ready (its cooldown recently had to be interrupted to fix a short-circuit).

  22. Peter Orland says:


    I have hear such promises before…

    Right now string theory can describe strongly-coupled gauge theory (by the way, so can non-numerical lattice gauge theories). OK, fine, one can play with the idea and try some phenomenology. There are also stringy ideas for summing planar diagrams, and a lot of other interesting ideas.

    BUT… no one can show that string theory methods gives the RIGHT strongly-coupled gauge theory, i.e. that obtained by integrating out all the short-distance degrees of freedom from QCD, starting near the UV-free fixed point. My impression is similar for approaches to condensed-matter physics.

    Promising a breakthrough doesn’t count as much as delivering one. Any good physicist who truly has a breakthrough in the works, would work on finishing it, and not making promises (that only spurs the competition).

  23. Marcus says:

    Peter Orland,
    you took Chris’s post seriously, as perhaps was proper, but to me it seemed humorous, somewhere on the scale between subtle and sidesplitting. He said:

    “…string theory indeed was tremendously successful recently. i actually think it is close to a breakthrough right now. give it just a few more years and you will see, that it is a fabulous phenomenological model for strongly coupled electron systems. and maybe – in some more distant future – it will come close to lattice gauge theory in describing the IR part of QCD.”

  24. Peter Orland says:

    Hi Marcus,

    It could be that I’m the guy in the room who doesn’t get the joke (especially if it’s on him), but in this case, I suspect Chris was serious. I think he is expressing the feeling of a lot of younger people in the field. If I’m right about this, they need to hear that promises mean nothing (except in the public-relations world), and only results count.

    Peter O.

  25. onymous says:

    Peter Orland said:

    BUT… no one can show that string theory methods gives the RIGHT strongly-coupled gauge theory, i.e. that obtained by integrating out all the short-distance degrees of freedom from QCD, starting near the UV-free fixed point.

    Indeed, one knows that they do not, and the reason, amusingly, is their failure to be sufficiently stringy. Calculability in these models relies on a hierarchy between the AdS curvature scale and the string scale. On the field theory side this means a large gap in operator dimensions that does not exist in QCD. In principle one can imagine a model that gets closer to QCD — or even is QCD — but in such a model this hierarchy would not exist, and one would have to do string theory rather than effective field theory in the AdS background. So far this is an insurmountable technical obstacle. The AdS/QCD models remain compelling and amusing toy models, but everyone working on them should (and, for the most part, does) understand that they are, in many ways, toys.

    As far as I can tell all the same objections apply to AdS/CMT. They might be valuable toy models that shed light on interesting phenomena, but they are unlikely to ever be models of real physical systems until the technical obstacle of doing string theory on highly curved spacetimes is overcome.

  26. TomInCalif says:

    Peter wrote: “… string theory proponents often use this misleading tactic, referring to string theory’s problems with predicting everything, when the problem is that it predicts nothing… ”

    As a layman in this field, I have a question: Can the proponents even construct all the correct parameters of our actual universe, in a consistent way, just by twiddling whatever “free parameter knobs” exist in string “theory” ?

    Or is even that challenge akin to recreating all the content of the Sunday NY Times, from a pile of burned ash?

  27. Greek says:

    as a European practitioner in the field that you may call String Theory, it is amusing to see that the String Wars is – perhaps – more than an urban legend! I see that (some) serious people take it seriously and (sometimes) use serious arguments. So here is mine:

    perhaps there is reason to be optimistic – whichever side of the trench one sees oneself. On the one hand, it is by now obvious to the honest physicists that the Scherk-Schwartz leap to connect the spin-2 string state to the graviton is probably misleading. It has lead to an enormous amount of work, much of which is of high quality both in physics and maths, however it has produced a “doomsday” religion regarding string theory: i.e. that string theory is a T.O.E. This nicely fit the old “particle” ideas regarding “effective field theory”, which are based on the philosophical conceptions such as the “atomic” theory of the ancient Greeks! Unfortunately, both the ancient Greeks and the modern non-Greeks are proven – by experiment – wrong.

    On the other hand, the honest physicist now observes that String Theory rather helps in the deconstruction of the old particle physics “effective field theory” philosophy! Indeed, “fundamental d.o.f” such as strings and (M)branes describe low energy phenomena in condensed matter. So what is “fundamental”, the string or the superconducting vortex observed in the lab? But this is supporting the idea (see P. Anderson etc) of “emergent physics”.

    So, if the recent AdS/CMT ideas work – even a little bit – String Theory will close a circle and get back to where it started; to a model describing the “phenomenology” of matter!

    Then, for one that is not a string-religion-fanatic, there is a clear winner in the String vs LQW war. It is String Theory of course – but viewed from this newer perspective. Indeed, many of the young generation view things that way – so in 10 years time when the old guard will retire the field will be completely different.

    So, cheer up, indeed.

    PS: Sorry for the long comment, I am completely new to blogging.

  28. Peter Woit says:


    Thanks a lot for the interesting comment. I agree with what I take to be your perspective that a lot of the string theory community, especially the younger part, has given up on the idea of using string/M-theory in 10/11 dimensions to unify physics, and moved on to trying to do other things, typically related in one way or another to AdS/CFT duality.

    That’s fine, but to me there still remains the problem that, non-perturbatively, QFT still remains mysterious in many ways, often in ways that AdS/CFT has no relevance to. It would be nice if the giving up on one failed idea could lead to people being willing to investigate a variety of new ideas, not just one particular one that historically grew out of the TOE failure.


    The problem is that the string theory machine used to construct “string theory backgrounds” that might look like our world turns out to have to be very complicated, without much calculational control. So, the state of the art is that as far as anyone can tell, in principle you can get our physics with the right parameters (and just about any parameters you want) out of such a “string theory background”. But actually doing this in a precise and well-defined way is way beyond current calculational capabilities, and to me it has never been clear that these calculational problems are not problems of principle, not just practicality.

  29. Greek says:

    Hi Peter,

    sure one should try new ideas, away from AdS/CFT. In fact, a particularly interesting clue has recently emerged in d=3. It seems that one needs to go beyond the usual group-based gauge-theory structures to understand three-dimensional theories. More complicated structures (based on 3-algebras) seem to be needed. Other ideas are welcome too!
    e.g. keep the term “systolic geometry” for the time when it will be applied to critical systems and the black-hole entropy..

    Again on the philosophical side (its very late here in the “old world” so I am allowed): this ear looks very much like the beginning of the 21s century when people would construct everything from machines -then came quantum mechanics. It also looks like the 70’s when people were seeing UFOs everywhere and nuclear science would explain everything – then came quantum gravity to spoil the party. It seems that in the coming decade something else will emerge that will change our current views.. hopefully.

  30. Pawl says:


    I am puzzled by your comment that String Theory “viewed from this newer perspective” is the winner in the “String vs LQW [sic] war.” Do you mean LQG (Loop Quantum Gravity)?

    If you mean that String Theory provides a more successful approach to quantum gravity, I don’t really see, based on what you’ve written, why you would hold that view — much less how it could be presented as something more than an opinion. But perhaps I’m just not understanding your point.

  31. Greek says:

    my point was that “dreams of a quantum gravity” (to paraphrase Weinberg) is not such a big thing after all.. There is much more to physics than QG.i.e. turbulence to just name a cliche’ example. Even if LQG provides the path to QG (which I believe is at least as doubtful as the opposite statement), it will not be relevant! On the other hand, ideas and maths that are spin-offs of String Theory touch upon much more physical structures (turbulence included!). That is why I think that the winner in String Theory. I hope I am clearer now.

  32. Pawl says:


    Thanks; I understand you now.

    I do not really agree that reconciling quantum theory with gravity is “not such a big thing” — although I agree strongly that there are many other important problems in physics. (I’ll point out too that I was not arguing that LQG was likely to be correct, but I was comparing its progress versus string theory’s on quantum gravity.)

  33. chris says:

    Peter Orland,

    unfortunately i am not as young as you imply – i wish i were. and just to clarify:

    if i was working on a candidate TOE and after 20 years discover that it might be a passable effective description for some solid state physics or quark gluon plasma, i’d not really call it a breakthrough.

  34. Peter Orland says:

    Well, Chris, I have to say, if it were true that string theory could solve real problems, I would consider that a success, not a disappointment. Having said that, I don’t see any sign of success.

    Any new solution in theoretical physics is a breakthrough. I personally would be much more happy and excited to solve Yang-Mills or High-T_{c} superconductivity than find a FUT (Fully Unified Theory – a more strident-sounding moniker than the melifluous TOE), since the former are related to experiments. I don’t understand the feeling of some physicists that all life begins at the Planck scale.

    What I said above (and see onymous’s comment also) is that string-theory ideas have hit a wall in field-theoretic applications. In fact, they hit this wall very soon after the ideas were proposed.

    The promise of a breakthrough means little. I wish people would stop making promises. People on the verge of breakthroughs don’t advertise – they break through!

Comments are closed.