On local scale invariance and the questionable theoretical basis of the conformal transformations' issue (1811.02458v2)

Abstract: Here we follow the mainstream of thinking about physical equivalence of different representations of a theory, regarded as the consequence of invariance of the laws of physics -- represented by an action principle and the derived motion equations -- under given transformations; be it coordinate, gauge or conformal transformations. Accordingly the conformal transformations' issue is discussed by invoking the assumed invariance of the laws of physics -- in particular the laws of gravity -- under conformal transformations of the metric. It is shown that Brans-Dicke and scalar-tensor theories are not well-suited to address physical equivalence of the conformal frames since the corresponding laws of gravity are not invariant under the conformal transformations or Weyl rescalings. The search for conformal symmetry leads us to explore the physical consequences of Weyl-invariant theories of gravity, that represent a natural arena where to discuss on physical equivalence of the conformally related representations. We show that conformal invariance of the action of a (supposedly conformal invariant) theory and of the derived motion equations is not enough to ensure actual Weyl invariance. It is required, also, that the underlying geometrical structure of the background spacetime be, at least, Weyl-integrable. Otherwise, if assume (as usual) spacetimes of (pseudo)Riemannian geometrical structure, the resulting -- apparently conformal invariant -- theory is anomalous in that, only massless matter fields can be consistently coupled. Gauge freedom, a distinctive feature of actually Weyl-invariant theories of gravity, leads to very unusual consequences.