From Sean Carroll’s weblog I see that he’s in Austin now for a session at a meeting of the Philosophy of Science Association. Philosophers of science seem to actually write up their talks in advance, and many of the talks for this meeting are already available online. Poking around on their website with papers from earlier conferences, I ran into one on Scientific Realism and String Theory by Richard Dawid.
Dawid appears to have swallowed the hype about string theory hook, line and sinker. He believes that string theory exhibits a new paradigm of how to do physics, one where the idea of being able to calculate anything about the real world and compare it to observations is passe. All that matters now is “theoretical uniqueness”, that one’s theory is the only possible one. He doesn’t seem to notice that there’s something kind of funny about people claiming that they have a wondrous unique theory, but don’t know quite what it is and can’t calculate anything about the real world with it. The S-matrix theorists of the 60s also promoted the idea that they had a wondrous unique theory, but didn’t know quite what it was. Probably one can dig up philosophy of science articles from that period about how a whole new paradigm of how to do science was required.
Dawid also seems to believe that the dualities of M-theory imply the “dissolution of ontology”, that “the ontological object has simply vanished”. In reality, what has vanished is not the ontological object, but the theory.
Over at Robert Helling’s web-site you can read an example of the latest philosophical excuses about why string theory now can’t predict anything, together with implausible wishful thinking about how this might change since “It’s just at this stage we are not yet powerful enough to make these kinds of predictions”.
For the life of me I can’t figure out why smart physicists and philosophers can’t see the obvious fact that is staring them in the face. You don’t need a new paradigm of how to do science, the old one works just fine. If you have a conjectural theoretical scientific idea, there are two ways in which it can turn out to be wrong. Either it predicts something that disagrees with experiment, or it is so vacuous that it predicts nothing. The evidence is now overwhelming that, if string theory is consistent at all, it is wrong for the second reason.
Update: Dawid actually has a whole fancy web-site about Realism and String Theory. He also has a newer paper on Undetermination and Theory Succession from a String Theoretical Perspective.
The PhilSci archive does have another paper about string theory, one by Reiner Hedrich entitled Superstring Theory and Empirical Testability. Hedrich is much less credulous than Dawid, noting about superstring theory “above all, it has fundamental problems with empirical testability – problems that make questionalbe its status as a physical theory at all.”