Last Friday I went to hear a talk by David Gross at the CUNY Graduate Center on “The Coming Revolutions in Fundamental Physics”. This was more or less Gross’s standard advertisement for string theory that he has been giving for nearly 20 years now. He explicitly started off with the claim:

“Fundamental Physics = String Theory”

His first PowerPoint slide was a quote from “Hannibal” by Thomas Harris, a sequel to the novel “Silence of the Lambs”. Evidently late in this novel the main character is depicted manipulating the “symbols of string theory”, with “equations that begin brilliantly and end in wishful thinking”. It sounds like Harris really has got the right idea about string theory, but Gross said his talk was designed to argue that string theory was not wishful thinking.

He then went on to claim that in 3-4 years there will be a headline in the New York Times about the discovery of supersymmetry at the LHC. From what I can tell, the LHC should have first beams in 2007, but even if everything goes according to plan, it won’t be until 2008 that the experiments there will have accumulated a non-trivial amount of data and analyzed it. Experience with colliders generally has been that getting them running at a useful luminosity can take quite awhile after they are first turned on. So 3 years from now is definitely out, 4 years is optimistic. This is now getting close enough that Gross and others seem intent on ignoring the failures of string theory, desperately hoping that superpartners will pop out of the LHC, thereby providing at least some vindication of the train of reasoning that lead to string theory. What will be interesting to see will be what Gross et. al. do when this doesn’t happen. Will they drop string theory? Quite possibly the LHC will revolutionize physics by showing us what is really causing the spontaneous breaking of the electroweak gauge symmetry. If this happens, everyone will abandon string theory and start working on this, 1984-2008 then becoming a period in the history of physics that particle theorists try and not think about.

Gross’s talk contained the usual tendentious pro-string theory points, here’s a few of them with commentary:

1. ” String theory is in a period like that of 1913-1925, it’s like the Bohr model, we’re waiting for the analog of Heisenberg’s or Schrodinger’s breakthroughs”

The problem with this is that the Bohr model was actually predictive, for instance it predicted a lot about atomic spectra that could be experimentally checked. There clearly was something right about the Bohr model, there is no good evidence there is something right about string theory.

2. “String theory is better than QFT, because QFT Feynman diagrams have these interaction vertices you can assign any interaction strength you want to”

This is not true of gauge theories, the different vertices are related by gauge symmetry. True you have to pick over-all gauge groups and representations, and the Higgs sector is problematic, but the claim that there is just one string theory is just wishful thinking.

3. “String theory is better than QFT because interactions are not at points, so short distance behavior is better”

Gross should be well aware that asymptotically free gauge theories are extremely well-behaved at short distances despite having point-like interactions, since he discovered this. It is also true that string theory perturbation theory is only known to be well-behaved up to two loops. My colleague Phong claims that higher loops remain very much not understood.

4. “String theory is a consistent, finite quantum theory of gravity”

Simply not true. Peturbative string theory is a divergent expansion, non-perturbative definitions don’t work for four large flat dimensions, rest small.

5. “String theory inspired brane-world scenarios, although I don’t really believe these”

Why would you think that an argument in a theory’s favor was that it inspired some clearly wrong models that you don’t believe and that don’t predict anything?

While Gross mentioned the “discretium”, he didn’t really explain exactly how disastrous this is for string theory, since it makes it essentiallly vacuous. He made a big deal of string theory implying that our notions of space and time need to be changed, but made it clear that no one really has a viable idea about how to change them. He puts his hopes in the fact that we still don’t understand what string theory is. This seems to me to be exactly the sort of wishful thinking that he claimed at the beginning was not what string theorists were doing.

His talk went on for more than an hour and a half. Several questions from the audience were taken, including one from Michio Kaku who claimed that dark energy was evidence of supersymmetry and asked about theories with two times. Gross didn’t seem interested in saying much about such theories. I noticed that two string theory postdocs I know were in the audience. They’ve both told me that they think the subject is at a point of crisis and they are thinking of quitting. I don’t think anything Gross said was likely to encourage them to continue.

Epicycles, aether, string theory?

I don’t believe at all that it would be a “stunning triumph” to find a consistent version of string theory that is compatible with any QFT. Quite the opposite: it would be a stunning failure. It would mean that, as a theory of particle physics, string theory was completely and utterly vacuous. There’s no reason other than wishful thinking to believe that such a theory would predict anything (if your theory fundamentally has nothing to say about matter, there’s no reason to believe it’s going to be able to precisely predict anything about any physical process).

Gross doesn’t believe this nonsense. He’s known to start quoting Churchill, shouting “Never, never, never, never give up!” when people like Susskind (or Srednicki) start talking about how great it would be to have a theory that is compatible with anything. At his CUNY talk and elsewhere he emphasized that it is the role of particle theorists to find a theory that explains the things about particle physics that the standard model doesn’t, not to congratulate themselves for having found a theory that is consistent with anything.

Gross now stakes his hopes on the idea that the presently known version of string theory is inconsistent and that when a consistent theory is found it will have better uniqueness properties. I (and many others, including Susskind) think this is wishful thinking. I also think he should ditch his powerpoint slide that says that string theory is a finite and consistent theory of quantum gravity, when later in his talk he is pinning his hopes on its inconsistency.

I think Mark’s point is this: it would be a “stunning triumph” to find a candidate for a UV completion of the Standard Model including gravity. Surely this would make new predictions. Granted, there is the issue that perturbative string theory is not provably finite. But, this does not mean the string theory would not let us calculate things we could not otherwise do.

Of course, this won’t lead us immediately to the “theory of everything,” whatever that might mean. But it could be an important step. We could try to test, for instance, whether the real world looks like some intersecting brane scenario.

We really do strongly disagree here.

First of all, even if you believe in finiteness of higher loops in string perturbation theory, perturbative string theory is not a finite theory. All evidence is that the expansion in the string coupling constant is a divergent series. What you’ve got is presumably an asymptotic expansion. This may give a useful approximation if the string coupling constant is very small, a useless one if it isn’t. Saying that perturbative string theory is a “finite, consistent theory of quantum gravity” is simply not true. You can’t sum the series to get something finite, and if you cut it off at a finite order your theory is inconsistent (not unitary).

So if you can get any QFT + gravity out of perturbative string theory + a choice of background, what have you actually got? You’ve got an exceedingly complex theory that predicts absolutely not a single thing about anything and is inconsistent to boot. Would this be a “stunning triumph”?

If you want to engage in extreme wishful thinking and believe that string theorists will come up with a finite, non-perturbative string theory that could have any QFT as its low energy limit, calling such a thing a “stunning triumph” would still be complete hype. It would predict nothing about anything. Theorists should save terms like “stunning triumph” for the day they actually manage to make a single solid prediction that in anyway goes beyond the standard model.

“While Gross mentioned the “discretuum”, he didn’t really explain exactly how disastrous this is for string theory, since it makes it essentiallly vacuous.”

I strongly disagree with this. The extreme case of the discretuum is that *every* QFT is realizable as the low-energy limit of string theory. This probably isn’t right, but suppose it is. Then we could take *any* QFT (with some exceptions: no Landau poles below the Planck scale, for example), include gravity, and have a finite quantum theory (assuming string finiteness holds up, which it may not). This should be considered a stunning triumph!

Furthermore, probably not *every* QFT is realizable in this way. Some are, some aren’t. Which class is the Standard Model in? It seems to me that this is a very important (though also very hard) question.

The great dream was that there would be only *one* QFT realizable as the low energy limit of string theory, and that it would turn out to be the Standard Model. It would’ve been great, but right now it doesn’t seem likely. That certainly doesn’t mean that string theory is vacuous.

Hi Fred!

The interesting thing about Gross’s talk was that he was kind of going out on a limb on the issue of supersymmetry at the LHC. I’m more and more convinced that he and a lot of others are getting discouraged about string theory and no supersymmetry at the LHC will be the final straw.

The worst possible thing for particle theory would be if the standard model Higgs shows up, behaving like a standard elementary scalar field with a certain mass. Then we would still be in the situation of having no idea where the Higgs potential or couplings come from, and no prospects for doing experiments in our lifetime to find out.

I’m also hoping the LHC doesn’t find a standard Higgs field, but evidence for some more interesting way of breaking electroweak gauge symmetry, one that we haven’t thought of yet.

Hi Peter..

If they don’t find any SUSY at LHC, I doubt it will be the death knell of String Theory. Alas, there is enough degrees of freedom there to just argue that it breaks at a higher scale, no big deal!

The nasty thing of course is that it just adds more fine tuning to minimal SUSY, and much of the original point would be lost.

I’m more interested though, in the (seemingly absurd) case that the LHC doesn’t discover the Higgs! AFAICS, all reasonable models put it firmly in reach of the LHC.. If we don’t find it, well, something drastic has got to give. I don’t see many people talking about that, so entrenched is the SM gospel in our minds. What a strange higgs sector that would imply.

Hi Matt,

Have you decided where to go to grad school yet?

Today I’ve put up somewhat more positive posts, and in general hope to mix the positive and negative. Unfortunately even the positive posts will probably be so speculative as to be “Not Even Wrong”. When I figure out a unification idea that really works I’ll dump the blog and start instead enjoying the fruits of fame and fortune that accrue to super-star physicists.

Your idea of pursuing applications of string theory to better understand strongly interacting gauge theories is quite reasonable. I don’t have any problem with that part of the string theory program. My problem very specifically is with the continual hyping of the failed idea of unifying the standard model and gravity in a 10/11d supersymmetric string/M-theory.

I’m not so convinced that new information about strongly interacting gauge theories coming from string theory duals will solve the problem of spontaneous gauge symmetry breaking. The technicolor idea is quite beautiful, but all implementations of it seem to involve adding a huge amount of new structure to one’s theory, while not being able to actually calculate anything one wants to calculate. On the other hand, maybe one can use string duals to

find a way of calculating things, and if this is the way the world works, there should be real evidence for it a few years from now.

Hi Sean, thanks for the advice about graphics, I’ll give that some thought. I’m still trying to see exactly how this thing works, so not sure how to comment on your comment in some way that news of this gets back to you. If you’re lacking topics to write about in your own blog, I’ve put up an entry here today with some questions for cosmologists.

Interesting to see that you have a blog. I’ve followed some of your s.p.r posts in the past with interest.

I’m an undergrad now, but I’ll be going into particle theory as a grad student next year. I have some skepticism about the virtue of a lot of research in string theory, but I can’t share your extreme skepticism.

I notice you divide your links into “non-string theory” and “string theory,” and title your blog “Not Even Wrong.” Does this indicate you intend to focus on anti-string theory polemics? Personally I would find some more positive posts interesting.

I was much more skeptical of string theory until the past year or so. Curiosity about various models of “large extra dimensions” led me to read papers of Hall and Nomura on orbifold GUTs, which then led to Nomura’s papers on warped geometry and “Higgsless” models. These led me to all sorts of other phenomenological articles, and at some point it became clear that the AdS/CFT correspondence is quite important in the phenomenology of Randall-Sundrum models and their variations.

Although it’s certainly not proven (but then, little in field theory is), I think the indications of dualities of field and string theories are too compelling to ignore. You bring up that asymptotically free gauge theories are well behaved. One interesting possible for electroweak symmetry breaking, as I’m sure you know, is the possibility that the Standard Model is embedded in a strongly interacting gauge theory at higher energies. The simplest technicolor scenarios aren’t feasible, and harder ones are tough to construct. Thus the discovery that certain such models might be dual to stringy models in higher dimensions seems to me a compelling reason to learn about string theory. These dualities might provide calculational tools that are lacking in field theory. Effective field theory, lattice gauge theory, and string theory, I think, should all be viewed as powerful tools for elucidating the behavior of physical theories we otherwise can’t understand.

I’m rather more skeptical of claims that string theory, on its own, will give us a theory of fundamental physics. Certainly we need further experimental input to progress much farther. The apparent existence of a “string theory landscape” that doesn’t uniquely predict a vacuum is disappointing. But, as they say, let’s not throw the baby out with the bathwater.

Peter, welcome to the blogosphere. I found your blog in the usual way — your link to mine. I suspect I will disagree with many things you say (I’m a fan of string theory, although not a string theorist myself), but that’s all part of the fun.

By the way, if you want to get a lot of hits, I’ve found it helps to stick up a pretty picture from the Hubble Space Telescope. Posts with words in them don’t seem to draw as well, for some reason.

Pingback: Not Even Wrong » Blog Archive » Various Links

Pingback: Not Even Wrong » Blog Archive » Pressure Mounts to Tie String Theory to the Real World