This week’s press release trumpeting a bogus “test of string theory” comes from the University of Illinois at Urbana-Champaign, which is headlined Scientists propose test of string theory based on neutral hydrogen absorption, and informs us that
Ancient light absorbed by neutral hydrogen atoms could be used to test certain predictions of string theory, say cosmologists at the University of Illinois. Making the measurements, however, would require a gigantic array of radio telescopes to be built on Earth, in space or on the moon.
String theory – a theory whose fundamental building blocks are tiny one-dimensional filaments called strings – is the leading contender for a “theory of everything.” Such a theory would unify all four fundamental forces of nature (the strong and weak nuclear forces, electromagnetism, and gravity). But finding ways to test string theory has been difficult.
Now, cosmologists at the U. of I. say absorption features in the 21-centimeter spectrum of neutral hydrogen atoms could be used for such a test.
One peculiar aspect of this press release is that it seems that the relevant paper is not yet publicly available. Supposedly it has been submitted to PRL and accepted, but it has not yet appeared in PRL, and I don’t see a preprint on the arXiv (the authors do have a PRL paper with an arXiv preprint from last year on a different topic, one that also came with a press release, but this one didn’t mention string theory).
As far as I can tell from the press release the idea behind this “test of string theory” is the same as lots of other similar ones that invoke cosmic strings. Among the huge variety of string theory backgrounds and the many possible ways to try and use such backgrounds to model the big bang, some will (just like some non-string theory GUT models) produce macroscopic “cosmic strings”. Astronomers have looked hard for evidence of such things and found none, but one can always imagine that, miraculously, such things exist, with characteristics exactly such that they wouldn’t have shown up so far, but would in some new, improved astronomical observations. In this case, I guess to come up with some new possible observation not already ruled out, the authors of the paper invoke a possible radio telescope with an area of a thousand km2.
There seem to be at least a couple reasons for the recent flood of bogus “we’ve found a test of string theory” press releases. One is that PRL evidently encourages authors to issue press releases whenever they have a paper appearing in PRL. Another reason is that string theorists are on the defensive, and some of them have decided that finding some way to claim that string theory really is testable, no matter how dubious, is the way to fight back. Earlier this month, one such claim hyped in New Scientist carried the headline “slammed for their failure to explain how our particular universe came to exist, string theorists are fighting back.” In an interview with string theorist Thibault Damour in today’s edition of the Swiss paper le Temps, he promotes three possible tests of string theory. One is the possibility (which he describes as “very speculative”) that one might observe extra dimensions at the LHC, another is cosmic strings, and finally there are his claims that string theory leads to violations of the equivalence principle. Lubos Motl strongly disagrees. Lubos also has a posting about this latest hype, where he comments:
Such possibilities highlight that creative people may often solve questions that look too difficult at the beginning. They also emphasize how incredibly idiotic are the aggressive crackpots’ proclamations that modern theoretical physics in general and string theory in particular is untestable.
Not clear who it is who believes that “modern theoretical physics” is untestable. While at Lubos’s blog, you might want to see what you can make of his posting on his new book The Bogdanov Equation: the secret of the universe?
Update: This story is appearing lots of places, including the UPI newswire, and at Wired, where the writer seems to realize that the bogosity level here may be problematic, including the unusual disclaimer:
Disclosure: I have no idea whether this makes sense.
Update: A correspondent points me to another recent “test of string theory”, one where for some reason the authors don’t seem to have issued a press release. The article is Toward a test of string theory using Rydberg atoms, and it begins by referencing my book and then claiming that
… measurable effects are predicted by String Theory on normal quantum scales, which the current criticisms have apparently overlooked.
What is discussed in the paper is actually not string theory, but just the idea of adding spatially non-commuting terms to the Heisenberg commutation relations. Certainly such terms should have experimentally observable effects. I suppose you can claim that such terms, of any size you want, come from a “string theory background”, but, as with all these “tests of string theory”, what is going on here just reflects the fact that you can pretty much get anything you want out of string theory, which is why it’s not testable…