Holography and Correlation Functions of Huge Operators: Spacetime Bananas

TL;DR

This paper explores the holographic duals of heavy operators in CFTs, revealing that their correlators correspond to backreacted geometries called 'spacetime bananas', and demonstrates this with two-point functions involving black hole geometries.

Contribution

It introduces the concept of 'spacetime bananas' as new geometries for heavy operator correlators and develops a holographic framework for their analysis.

Findings

Two-point functions of heavy operators are described by black hole geometries.

The on-shell action reproduces CFT results for heavy operators.

Boundary terms on the stretched horizon are crucial for correct computations.

Abstract

We initiate the study of holographic correlators for operators whose dimension scales with the central charge of the CFT. Differently from light correlators or probes, the insertion of any such maximally heavy operator changes the AdS metric, so that the correlator itself is dual to a backreacted geometry with marked points at the Poincar\'e boundary. We illustrate this new physics for two-point functions. Whereas the bulk description of light or probe operators involves Witten diagrams or extremal surfaces in an AdS background, the maximally heavy two-point functions are described by nontrivial new geometries which we refer to as "spacetime bananas". As a universal example, we discuss the two-point function of maximally heavy scalar operators described by the Schwarzschild black hole in the bulk and we show that its onshell action reproduces the expected CFT result. This computation is nonstandard, and adding boundary terms to the action on the stretched horizon is crucial. Then, we verify the conformal Ward Identity from the holographic stress tensor and discuss important aspects of the Fefferman-Graham patch. Finally we study a Heavy-Heavy-Light-Light correlator by using geodesics propagating in the banana background. Our main motivation here is to set up the formalism to explore possible universal results for three- and higher-point functions of maximally heavy operators.