I am nowhere near qualified to critique String/M Theory on its mathematical foundations, but I do have to wonder where it all is supposed to end. Every once in a while I read about a new “discovery” in the field, but it turns out what has been “discovered” is essentially a new bit of math. That’s great, but does it have any relevance at all to the physical world? I know String/M Theory will prove to be useful, even if it’s totally wrong, because mathematical physicists are pushing the envelope of mathematics itself. Doubtless people even today are finding interesting and powerful new insights that grew out of M maths that give them better tools to pack spaces or tie knots or what have you.
But when does physics turn into pure mathematics? I understand that M-Theory is tightly constrained in some respects by nature. You can’t have arbitrary numbers of dimensions and have it resemble our universe. The dimensions can’t have too many holes; the strings/branes can’t be too tightly or loosely wound; and so on. But it seems that even with these constraints, the possibilities that are left over are virtually endless. There could be an infinite number of ways to formulate consistent M Theories, and if one doesn’t turn out to fit what we observe, maybe another one will. I have to wonder if M Theory could ever be disproven, since it appears to be so adjustable. Dimensions can be curled up; or not. We could live in a “brane-world”. Or not. The curled-up dimensions might be big enough to measure with the LHC. Or not. Things can get bigger or smaller, wound up or unwound, string-like, brane-like, perforated, etc., etc., and it’s all allowed, it would seem, so long as it isn’t strictly forbidden by our current level of understanding (which everyone claims is poor).
Meanwhile, we’ve got massive neutrinos, dark matter, dark energy/cosmological constant, an estimation of the cosmological constant that is 120 orders of magnitude too big, the Higgs mechanism, and so on, and as far as I know, M-theory hasn’t given us a single testable insight into any of it that sets it apart from other candidates. We’ve got tons of new observations that may take us way beyond the Supersymmetric Standard Model, and yet did M Theory anticipate any of it? I would guess, since it’s got so many free parameters, one could probably argue yes, it did, but obscure postdictions tend not to be very convincing to skeptics. When are the M Theorists going to nail something down? Can they be effectively refuted by any observation, or will a new version of M-Theory inevitably be formulated to fit the new constraints?
That’s my biggest question about extra dimensions.