Perturbative Unitarity and the Wavefunction of the Universe

TL;DR

This paper explores how unitarity constraints shape the wavefunction of the universe in cosmology, using geometric methods to connect cosmological and flat-space optical theorems and clarify the role of the iε-prescription.

Contribution

It introduces the optical polytope as a geometric structure encoding unitarity in cosmological wavefunctions, linking cosmological and flat-space optical theorems through polytope subdivisions.

Findings

The iε-prescription is determined by positivity and orientation of the cosmological polytope.

Unitarity is encoded in a non-convex part of the polytope called the optical polytope.

The cosmological optical theorem arises from polytope triangulations and subdivisions.

Abstract

Unitarity of time evolution is one of the basic principles constraining physical processes. Its consequences in the perturbative Bunch-Davies wavefunction in cosmology have been formulated in terms of the cosmological optical theorem. In this paper, we re-analyse perturbative unitarity for the Bunch-Davies wavefunction, focusing on: 1) the role of the $i\epsilon$-prescription and its compatibility with the requirement of unitarity; 2) the origin of the different "cutting rules"; 3) the emergence of the flat-space optical theorem from the cosmological one. We take the combinatorial point of view of the cosmological polytopes, which provide a first-principle description for a large class of scalar graphs contributing to the wavefunctional. The requirement of the positivity of the geometry together with the preservation of its orientation determine the $i\epsilon$-prescription. In kinematic space it translates into giving a small negative imaginary part to all the energies, making the wavefunction coefficients well-defined for any value of their real part along the real axis. Unitarity is instead encoded into a non-convex part of the cosmological polytope, which we name optical polytope. The cosmological optical theorem emerges as the equivalence between a specific polytope subdivision of the optical polytope and its triangulations, each of which provides different cutting rules. The flat-space optical theorem instead emerges from the non-convexity of the optical polytope. On the more mathematical side, we provide two definitions of this non-convex geometry, none of them based on the idea of the non-convex geometry as a union of convex ones.