Title: Determining the normalization of the quantum field theory vacuum, with implications for quantum gravity

Abstract: In a standard quantum field theory the norm $\langle \Omega\vert \Omega\rangle$ of the vacuum state is taken to be finite. In this paper we provide a procedure, based on constructing an equivalent wave mechanics, for determining whether or not $\langle \Omega\vert \Omega\rangle$ actually is finite. We provide an example based on a second-order plus fourth-order scalar field theory, a prototype for quantum gravity, in which it is not. In this example the Minkowski path integral with a real measure diverges though the Euclidean path integral does not. Thus in this example contributions from the Wick rotation contour cannot be ignored. Since $\langle \Omega\vert \Omega\rangle$ is not finite, use of the standard Feynman rules is not valid. And while these rules not only lead to states with negative norm, they in fact lead to states with infinite negative norm. However, if the fields in that theory are continued into the complex plane, we show that then there is a domain in the complex plane known as a Stokes wedge in which one can define an appropriate time-independent, positive and finite inner product, viz. the $\langle L\vert R\rangle$ overlap of left-eigenstates and right-eigenstates of the Hamiltonian; with the vacuum state then being normalizable, and with there being no states with negative or infinite $\langle L\vert R\rangle$ norm. In this Stokes wedge it is the Euclidean path integral that diverges while the Minkowski path integral does not. The concerns that we raise in this paper only apply to bosons since the matrices associated with their creation and annihilation operators are infinite dimensional. Since the ones associated with fermions are finite dimensional, the fermion theory vacuum is automatically normalizable. Our results are relevant to the construction of a consistent, unitary and renormalizable quantum theory of gravity.