Penrose's singularity theorem and the Kerr space-time
Jonathan Brook, Chris Stevens
TL;DR
The paper investigates how Penrose's singularity theorem applies to Kerr spacetime, clarifying that the theorem requires global hyperbolicity and a trapped surface. In the maximally extended Kerr spacetime these conditions fail, so the theorem does not force a singularity there; however, in the non-extended region up to the inner horizon $r_-$ the hypotheses can hold and yield incomplete future-directed null geodesics even though scalar curvature invariants remain finite, exemplified by a detailed Kerr-Schild analysis and geodesic ray tracing. A concrete calculation with $M=1$, $2a=M$, and $r_-\approx0.13398$, $r_+\approx1.86603$ shows a trapped surface with positive null-convergences and a geodesic crossing $r_-$ at finite affine parameter, while the Kretschmann scalar remains $I=48/r^6$ on the equatorial plane for $r>0$. The results illustrate that geodesic incompleteness does not necessarily indicate a physical curvature singularity and discuss the Cauchy-horizon nature and potential instabilities of the Kerr interior under perturbations.
Abstract
In this short paper, Penrose's famous singularity theorem is applied to the Kerr space-time. In the case of the maximally extended space-time, the assumptions of Penrose's singularity theorem are not satisfied as the space-time is not globally hyperbolic. In the case of the unextended space-time -- defined up to some radius between the inner and outer event horizons -- the assumptions of the theorem hold, but scalar curvature invariants remain finite everywhere. Calculations are done in detail showcasing the validity of the theorem, and misconceptions regarding the characterization of physical singularities by incomplete null geodesics are discussed.
