The case for background independence

TL;DR

Smolin argues that a fundamental theory of gravity must be background independent, contrasting it with the background-dependent string landscape. He surveys leading relational programs—causal sets, loop quantum gravity, and causal dynamical triangulations—and discusses their successes (e.g., discrete quantum geometry, emergence of 3+1 spacetime) and open problems (classical limit, inverse problem, time). He critiques the string theory landscape as inherently tied to fixed backgrounds and advocates a relational program, including cosmological natural selection, as a more falsifiable path toward unification and cosmology. The work highlights that background independence can yield testable predictions and suggests relational extensions to quantum theory and cosmology as promising directions for future research.

Abstract

The aim of this paper is to explain carefully the arguments behind the assertion that the correct quantum theory of gravity must be background independent. We begin by recounting how the debate over whether quantum gravity must be background independent is a continuation of a long-standing argument in the history of physics and philosophy over whether space and time are relational or absolute. This leads to a careful statement of what physicists mean when we speak of background independence. Given this we can characterize the precise sense in which general relativity is a background independent theory. The leading background independent approaches to quantum gravity are then discussed, including causal set models, loop quantum gravity and dynamical triangulations and their main achievements are summarized along with the problems that remain open. Some first attempts to cast string/M theory into a background independent formulation are also mentioned. The relational/absolute debate has implications also for other issues such as unification and how the parameters of the standard models of physics and cosmology are to be explained. The recent issues concerning the string theory landscape are reviewed and it is argued that they can only be resolved within the context of a background independent formulation. Finally, we review some recent proposals to make quantum theory more relational.