Complementarity, not Firewalls

Daniel Harlow

Published 2012-07-26, updated 2012-08-09Version 3

In this note I argue that a version of complementarity is possible which evades the need for the "firewalls" recently proposed by Almheiri, Marolf, Polchinski, and Sully to burn up observers falling into black hole horizons. In particular I claim that it is consistent for an infalling observer to fall through an "old" black hole horizon without burning up, without this observer or any other seeing information loss or a violation of low energy effective field theory in an unexpected place. The reason that AMPS find the opposite conclusion is because they attempt to use low energy physics to translate between the quantum mechanics of different observers rather than to describe the experiments of only a single observer; I argue that this translation is polluted by short-distance physics related to stretched horizons.