This Week’s Hype

Bogus media stories about how “physicists finally find a way to test string theory” have now been with us for decades, with a large number of them documented here. Recently this phenomenon seemed to finally be dying down, with such stories the province of more obscure media outlets and the press offices of not very well-known institutions. Yesterday though saw a new example of the genre, coming to us from IAS faculty and the Fermilab/SLAC publication Symmetry, which announces that Theorists from the Institute for Advanced Study have proposed a way forward in the quest to test string theory.

The source of all this is the Arkani-Hamed/Maldacena paper Cosmological Collider Physics from earlier this year. As usual with a lot of these bogus stories about “testing string theory”, the work in question actually has nothing to do with string theory. It’s about possible ways to look for particle physics effects in subtle effects in non-Gaussianities in the CMB. This is a theoretically interesting topic, but suffers from the obvious problem that, experimentally, there are no non-Gaussianities in the CMB. The limits on non-Gaussianity from Planck and other CMB experiments are quite strong.

The connection to string theory is given in the article as:

But scientists are working out ways that experiments could at least begin to test parts of string theory. One prediction that string theory makes is the existence of particles with a unique property: a spin of greater than two.

This is of course complete nonsense, since there are plenty of known particles of spin greater than two. String theory arose as an attempt to explain some of these, but it turned out that it didn’t work, the actual explanation was QCD, a quantum field theory. The author seems to have gotten this argument from the following statement in the Arkani-Hamed/Maldacena paper:

Of course, if we were to detect the contribution of a spin 4 state in the non-gaussianity, it would be a strong indication of string theory during inflation, since we suspect that a structure like string theory follows when we have weakly interacting particles with spin s greater than 2.

Knowing that there is a spin 4 state up at the inflation or Planck scale would of course be quite interesting, but I don’t see any reason to believe that effective field theory would apply to it or that this would “of course” “be a strong indication of string theory”. This argument would actually make better sense at lower energy. I suppose you could claim that lots of work being done at the LHC is “a way forward in the quest to test string theory”, since any day it could lead to evidence for a new weakly interacting spin 4 state. That would of course be pretty silly, but less silly than this article.

Combined with the bogus “test”, the article includes a large helping of the usual promotional material, ending with a section on “The value of strings”. We’re told that

Witten and others believe that such successes in other fields indicate that string theory actually underlies all other theories at some deeper level.

“All other theories”???

Update: I should make clear that my comment about the strength of the limits on non-Gaussianities was about the quality of the experimental results, and my impression that there are not near-term prospects for doing much better. Depending on what models one is talking about, such results are often not strong constraints. A correspondent suggest this source for more information about all of this.

This entry was posted in This Week's Hype. Bookmark the permalink.

10 Responses to This Week’s Hype

  1. Nathalie says:

    We are indeed living in the golden age of bogus speculations presented as facts.
    Here is another example from a recent talk by Hawking, taken from

    “The existence of alternative histories with black holes suggests this might be possible,” Hawking said. “The hole would need to be large and if it was rotating it might have a passage to another universe. But you couldn’t come back to our universe.”

  2. Shantanu says:

    Peter nice article. I sincerely hope the author of this article reads your critique
    and wish journalists for such magazines are more responsible.

  3. KJ says:

    Note that the author’s comment that string theory “underlies all other theories” does not qualify that he is talking about correct, or even approximately correct, theories. Thus, string theory underlies even all false theories. If there are really 10^500 versions of string theory, this may very well be true!

  4. Avattoir says:

    KJ, then that’s even more exciting, given there’s no limits at all, not even theoretically, on the number of false theories.

  5. George Ellis says:

    “There are plenty of known particles of spin greater than two.” – Theoretically proposed, or experimentally verified? What are they?

  6. TS says:

    The particle data group lists 17 states of spin 4 or above. Now the PDG is partly a political entity and one should be a bit careful about these listings (something LHCb’s now-famous pentaquark analysis wasn’t), but e.g. the a_4(2040) is a spin-4 resonance established beyond all doubt. Actually, the next edition of the PDG listings will include my measurement of its properties — and I can assure you that the data are unambiguous about its spin being 4.

  7. vmarko says:


    I think that Arkani-Hamed and Maldacena are talking about spin-4 fundamental (i.e. elementary) particles, as opposed to composite particles which can be of any spin (some atomic nuclei can have spin 9/2 or such). The fundamental particles with spin higher than 2 are sort-of forbidden in QFT due to a no-go theorem (Coleman-Mandula, Weinberg, Weinberg-Witten — see for example 1007.0435 for a review). But apparently they are not forbidden in string theory, so if (say) a spin-4 elementary particle has some signature in the CMB, string theory would be able to describe it while ordinary QFT wouldn’t. Or something along those lines.

    Of course, observation of spin-4 fields is not a *prediction* of string theory (AFAIK). IOW, the existence of such particles is *allowed* by ST, but not *required* by it.

    HTH, 🙂

  8. Peter Woit says:

    The problem is that how do you tell if a spin 4 state up at that scale is “elementary”? Composite particles look elementary unless you can probe them at some higher scale. This is why they specify “weakly interacting” instead. It’s extremely unlikely that you can see evidence at all that such a state exists, much more so that you can study its interactions. And all of this is going to be taking place up near the Planck scale, where all bets based on effective field theory arguments are off.

  9. vmarko says:


    I fully agree with you. It’s just that I saw some confusion in the comments regarding the question whether spin-4 particles have already been detected in nature or not. So I wanted to clarify that the existence of composite spin-4 particles is not controversial, while the existence of elementary spin-4 particles is. Of course, establishing (experimentally) that a given particle is indeed elementary is a very hard task even at the LHC scales, let alone near the Planck scale.


Comments are closed.