The Great Cosmic Roller-Coaster Ride

More than three years ago Scientific American ran a feature article by Bousso and Polchinski promoting the then new idea of The String Theory Landscape. Now that this pseudo-science has become well-entrenched in the physics community, this month’s issue of the magazine has a feature article on The Great Cosmic Roller-Coaster Ride, describing how

one of the emerging themes of 21st-century cosmology is that the known universe, the sum of all we can see, may just be a tiny region in the full extent of space

and claiming that this is “stimulating a thorough rethinking of the early universe in terms of string theory.” There is quite a bit of defensiveness about string theory in the article, where it is described as the “leading candidate for the foundational laws of nature”. The authors note that “String theory has received some unfavorable press of late”, and characterize criticism of the theory as due to the fact that it “has yet to be tested experimentally”, ignoring the fact that much of the criticism is about string theory’s inherent untestability. Not only has it not been tested yet, but no one has any idea how to test it ever. They admit as much when it comes to predictions about particle physics:

string theory has disappointed because it has not yet been possible to test it experimentally, despite more than 20 years of continued investigation. It has proved hard to find a smoking gun – a prediction that, when tested, would decisively tell us whether or not the world is made of strings. Even the Large Hadron Collider (LHC) – which is now nearing completion near CERN , the European laboratory for particle physics near Geneva – may not be powerful enough.

At the same time, they imply that the answer to string theory’s problems is that it will produce testable predictions about cosmology. They describe work of their own and other people attempting to use as an inflaton field positions of branes or moduli parameters describing positions in the Landscape. What predictions do they see coming out of this?

  • CMB experiments will continue to not see effects due to gravitational waves. In other words, the prediction is just that CMB experiments won’t see anything relevant. Not exactly a distinctive prediction of string theory, and Lenny Susskind lists seeing such effects as the main hope for observational evidence of the Landscape. If such effects are seen, I doubt that string cosmologists will give up on string theory, but just come up with models that do “predict” such effects.
  • Cosmic superstrings, observed by their gravitational lensing effects. One can in principle construct scenarios where cosmic superstrings are produced in the early universe in just such a way as to have eluded all observation until now, but such that one will turn up as we look more closely at more galaxies. I don’t know of any reason for this other than wishful thinking. If cosmic superstrings don’t show up over the next few years, I don’t think anyone will take this as evidence against string theory.
  • I don’t think it does much for the public understanding of science or increases respect for scientists when they decide to go to the public in this way, promoting extremely speculative and complex ideas that lack not only a glimmer of experimental evidence, but also any plausible idea about how they can be tested.

    Update: There’s a new review of string cosmology up on the arXiv tonight. The authors contradict the SciAm article’s claim about whether these models can accomodate observed effects of gravitational waves:

    As an example, in many – but not necessarily all – string inflation models, the primordial tensor signal is very small.

    giving examples of models with detectable gravitational waves (see here).

    While the article as a whole is pretty much unadulterated hype for string cosmology, it ends on a downbeat:

    Despite these promising signs, it remains to be seen whether this endeavor will lead to genuine contact between experiment and Planck-scale physics. In many scenarios, inflation is described by a well-controlled, albeit fine-tuned, effective field theory Lagrangian, and inflation lasts long enough to obscure all evidence of a pre-inflationary stage. If we live in such a universe, cosmological observations can, at best, teach us about the nature of the inflaton, but will provide few clues about more fundamental physics, except perhaps through the enduring mystery of dark energy.

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

    9 Responses to The Great Cosmic Roller-Coaster Ride

    1. roland says:

      “Not only has it not been tested yet, but no one has any idea how to test it ever.”

      This might very well be true, but a good half of your postings
      feature a similar statement. Perhaps you should choose it as a
      caption for the whole blog.

      ..

      However, maybe it can’t be said often enough and in fact
      i like your blog anyway.

    2. Coin says:

      This might very well be true, but a good half of your postings
      feature a similar statement. Perhaps you should choose it as a
      caption for the whole blog.

      Technically, isn’t it expressed in the title of the blog?

    3. roland says:

      i noticed that right after posting.

      but it might as well have been some sort of weird sarcasm.

    4. Kris Krogh says:

      Hi Roland,

      Trouble is, string theorists keep coming up with new “tests,” and others cite those without checking them.

      See this exchange with Joe Polchinski, items 35, 43, 44. He never responded to the last one.

    5. fulo says:

      See this exchange with Joe Polchinski, items 35, 43, 44. He never responded to the last one.

      Kris,

      I’m very far from sharing Prof. Polchinski’s optimistic view of the landscape, but I think his answer #43 is perfectly sensible. What he says is that having a theoretically tractable model of a strongly coupled non-abelian gauge theory out of equilibrium is important and interesting. It is.

      I think there are at least three reasons why he didn’t answer your comment about quoting a non-refereed talk as a basis for his remarks.

      First, because if you really want references to refereed papers, you just have to look them up in the talk he mentions.

      Second, because he was posting his answer to a weblog, not to a scientific journal, so there’s no need to be so picky about references.

      And third, because actually whether a paper is refereed or not is of no importance at all to any practicing physicist. People download papers right from the arXiv, read them, and decide by themselves whether they believe them or not. You only download papers from journals when they are too old to be found in arXiv or, else, when you are very interested in the figures, which are usually better printed in the published version.

    6. Quixotik says:

      This blog is a bit one-note, true, but many good blogs are…if someone ever actually comes up with a universally accepted way to test string theory, this blog will lose it’s raison d’etre.

    7. nerd says:

      Quixotik, they can test string theory, the whole problem is that there are 10^500 versions to falsify before any alternative ideas can get a look in…….

    8. roland says:

      maybe browsing through that math book, they just chose the wrong page. i mean the not even wrong page.

    9. Thomas Love says:

      nerd said: “…. there are 10^500 versions to falsify before any alternative ideas can get a look in…….”

      What a horrifying possibility…that would tie up progress in physics for 10^490 years! Or longer! I won’t be able to sleep tonight thinking about that.

    Comments are closed.