Hmm seems like they just made it up to make their equations make sense.
It's just a hunch but it seems remarkably elegant to me.
You may be more right than you know.
I've Read Brian Greene's "The Fabric of the Cosmos" and it is clear that he is both a very good writer and a very good physicist. 'Cosmos' was a well written and a very entertaining and informative book. In particular, his explanation of Bell's Theorem (weird action at a distance) is far superior to any other description that I have seen - it's actually understandable.
But when Greene get's to String Theory his arguments get very weak. Here is a theory that has been under development for over three decades, promises to explain anything and everything but has yet to be subjected to even a single simple experiment that would validate the basic assumptions, predictions and principles of the theory. So I say: What good is such a theory? In my very, very humble opinion, it's not worth much. I expect that in the end String Theory will to fail to produce because that has been it's track record for a good share of my life. Enough I say. Let some of our bright young physicists work on something else.