Some evildoer at ScienceBlogs asks "Is String Theory an Unphysical Pile of Garbage?" and references

this paper (in particular p.54 to 57) and other heretic stuff.

"... in string theory, we don't know both the variables and the equations. In fact, unless another theory (...) comes along that encompasses and expands upon string theory, string theory isn't a fundamental theory at all, due to instabilities."

I think this is yet another job for SuperLumo, defender of the one true string theory!

While we wait for SuperLumo, let me add a few remarks about this (but keep in mind that I am not a string theorist).

The blog post and the paper it references basically complain about the lack of a fundamental, non-perturbative formulation of string theory. But it seems to me that the Maldacena conjecture provides for such a non-perturbative description, currently at least for AdS.

A while ago string

*field*theory has been proposed as a more fundamental theory and the referenced paper suggests that it suffers from instabilities. I cannot judge the argument in detail, but it seems to me that string

*field*theory is indeed no longer such a hot topic.

Most of the effort seems today focused on generalizing and understanding the AdS/CFT correspondence.

However, I also think it is a mistake to underestimate 'string theory as perturbation theory'.

After all it provides for the only known way to deal with quantum gravity but at the same time keep local Lorentz invariance (i.e. a smooth spacetime) and quantum theory as we know it. And it gets surprisingly far, including consistent calculations of black hole entropy, making use of the amazing string dualities.

Many questions remain unanswered and perhaps early hopes that string theory will answer all questions of particle physics turned into consternation about the multiverse, but I think the answer to the "incendiary title" of the blog post obviously has to be "No!".