I heard recently in a talk by Alain Connes that a certain Woodword (or Woodward or Woodard, not sure with reccording quality) showed an important result that a quantum theory of gravity cannot be built directly from Planck-scale elementary degrees of freedom (like causal sets do, I presume), and required instead a more global approach (like non-commutative geometry or loop quantum gravity), even though those global approaches may eventually lead to planck-scale quantised behaviours (as LQG does).
I was a bit surprised, own to the abundant literature on "bottom-up" quantum gravity, and suspected that there should be specific assumptions behind such a result - but I could not find the reference for this work, to check that point.
So does anyone know this result? And if so, do you have the refence for it?
Thanks beforehand