**A cautionary case of casual causality â€“ Diagnosing (a)causality in the EFT of gravity**

In recent years, causality has emerged as a powerful criterion to distinguish between effective field theories (EFTs) arising from physical and unphysical high-energy theories.

A direct way to ensure a given EFT is causal is to demand a lower bound on scattering time delays, which essentially imposes a speed limit averaged over the trajectory. In flat space, this is unambiguously dictated by the Minkowski light cones, but the situation is more subtle with dynamical gravity. I will make the case that the relevant notion is so-called infrared (IR) causality.

As an example, I will apply this to study Gauss-Bonnet gravity in $D \geq 5$. Previously we found that, on spherically symmetric black hole backgrounds, potential signatures of IR causality violation are not large enough within the EFT of gravity to discard the Gauss-Bonnet operator. The parameter space allowed by IR causality is in fact consistent with gravitational positivity bounds. Attempts to bypass this conclusion use configurations of shockwaves designed to artificially enhance causality violation. In this talk I will emphasise that this is not possible.

This talk is based on arXiv:2112.05031 and upcoming work in collaboration with C. de Rham, A. Margalit, and A. J. Tolley.