Local(ish) gravity theories in conformal superspace

TL;DR

This paper explores a class of gravity theories with conformal symmetries in superspace, leading to a static or conformally diffeomorphic universe, with implications for quantum gravity and causal structure.

Contribution

It introduces a new class of geometrodynamical gravity models with conformal symmetries, analyzing their dynamics, matter couplings, and quantization properties.

Findings

Two gravitational degrees of freedom are maintained.

The model admits a geometric interpretation as a geodesic in conformal superspace.

The propagator exhibits higher-order spatial derivatives, suggesting regularization properties.

Abstract

Motivated by well-known obstacles to quantum gravity, I look for the most general geometrodynamical symmetries compatible with a reduced physical configuration space for metric gravity. I argue that they lead either to a completely static Universe, or one embodying spatial conformal diffeomorphisms. Demanding locality for an action compatible with these principles determines the sort of terms one can add, both for the purely gravitational part as well as matter couplings. The symmetries guarantee that there are two gravitational propagating physical degrees of freedom (and no explicit refoliation-invariance). The simplest such system has a geometric interpretation as a geodesic model in infinite dimensional conformal superspace. An approximate solution to the equations of motion corresponds to a static Bianchi IX spatial ansatz. The unique coupling to electromagnetism forces its propagation equations to be hyperbolic, enabling us to \"build" a standard space-time causal structure. There are, however, deviations from the standard Maxwell equations when space-time anisotropies become too large. Regarding quantization, with the geometric interpretation and the lack of refoliation invariance, the path integral treatment of the symmetries becomes much less involved than the similar approaches to GR. The symmetries form an (infinite-dimensional) Lie algebra, and no BFV treatment is necessary. We find that the propagator around the flat solution has up to 6-th order spatial derivatives, giving it plausible regularization properties as in Horava-Lifschitz.