Is spacetime curved? Assessing the underdetermination of general relativity and teleparallel gravity

TL;DR

This paper examines whether the empirical equivalence of general relativity and teleparallel gravity challenges the notion that spacetime curvature is an essential feature of reality, analyzing arguments and conceptual issues involved.

Contribution

It critically assesses claims that TEGR's flat spacetime undermines the realism about spacetime curvature in GR, and discusses conceptual problems related to operational and visual interpretability.

Findings

TEGR is empirically equivalent to GR despite different geometric assumptions

Arguments claiming no underdetermination are challenged and reevaluated

Conceptual issues of operationalisability and visualisability are analyzed

Abstract

Realism about general relativity (GR) seems to imply realism about spacetime curvature. The existence of the teleparallel equivalent of general relativity (TEGR) calls this into question, for (a) TEGR is set in a torsionful but flat spacetime, and (b) TEGR is empirically equivalent to GR. Knox (2011) claims that there is no genuine underdetermination between GR and TEGR; we call this verdict into question by isolating and addressing her individual arguments. In addition, we anticipate and evaluate two further worries for realism about the torsionful spacetimes of TEGR, which we call the "problem of operationalisability" and the "problem of visualisability".