Yesterday at the KITP Wati Taylor gave a talk entitled Freedom and Constraints in the Landscape of Intersecting/Magnetized Branes. During the talk he explained in detail the problem of lack of predictivity caused by the landscape. As far as anyone knows, to the extent you can calculate anything, you can get whatever you want: “Anything Goes”, and string theory is useless for ever predicting anything. He was looking at some particular classes of vacua, chosen for their computational tractability, and hoping to find some constraints among the quantities one can compute. There’s no known reason to expect this, but one can compute anyway and hope. The end result was the expected one: you can get whatever you want. Here are some quotes from the talk:
So, We’re really in a very challenging situation where we don’t really know how to define the theory, we don’t know what the set of solutions are, and even if we did we would have a very hard time making a sensible statement about what that means for predictions…
Every piece of data we have so far I would say is consistent with the notion that everything is pretty much uniformly and randomly distributed in the landscape.
There was extensive discussion of the predictivity problems and overwhelming evidence string theory can’t ever predict anything below the Planck scale (this wasn’t discussed, but I don’t see how it predicts much above the Planck scale either). For some reason there was no drawing of the obvious conclusion that one should just give up on the idea and try something else.