Infinitesimal Minkowskianity for manifolds with continuous Lorentzian metrics

Vanessa Ryborz · 2026

Read Infinitesimal Minkowskianity for manifolds with continuous Lorentzian metrics by Vanessa Ryborz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-22420.txt.

Introduction

Ryborz proves that any metric measure spacetime from a smooth manifold with a continuous Lorentzian metric is infinitesimally Minkowskian, if the manifold is causally simple. That metric splits spacetime into time and space directions, with light cones at each point. A spacetime is simple in causal terms when it has no flaws such as closed timelike curves. Infinitesimally Minkowskian means that, at small scales, it looks like flat model spacetime. The result extends small-scale flatness to spacetimes with a merely continuous metric.

Catalog record

CREATORVanessa Ryborz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE88909
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.22420

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Infinitesimal Minkowskianity for manifolds with continuous Lorentzian metrics Vanessa Ryborz Abstract We prove that any metric measure spacetime arising from a smooth manifold M M endowed with a continuous Lorentzian metric g g is infinitesimally Minkowskian, under the assumption that ( M , g ) (M,g) is causally simple. † † footnotetext: MSC2020: 53C50, 53C23, 53B30 Keywords . Lorentzian manifolds of low regularity, metric measure spacetimes, Lorentzian length spaces, causal functions Acknowledgements. The author gratefully acknowledges Andrea Mondino for his supervision and guidance, and thanks Clemens Sämann for valuable discussions. The author is supported by a postgraduate scholarship from the Mathematical Institute at the University of Oxford. 1 Introduction The geometry of metric (measure) spaces is an active field of study across analysis, metric and differential geometry, and probability. Metric measure spaces can be seen as a generalisation of smooth Riemannian manifolds, which turn out to carry enough information to allow generalisations of several core notions from differential geometry. These include upper and lower sectional curvature bounds, which can be defined on length spaces (see for instance [ 2 ] ), the celebrated curvature dimension condition \mathsf{CD}(K,N) based on optimal transport introduced by Lott-Villani [ 34 , 33 ] and Sturm [ 46 , 47 ] (a synthetic definition of lower Ricci curvature and upper dimension bounds), and a generalised first order calculus (see [ 4 ] ). Analogous approaches were recently started in Lorentzian signature: After the pioneering work by Kronheimer and Penrose [ 28 ] and Busemann [ 15 ] , Kunzinger and Sämann [ 31 ] proposed the notion of Lorentzian (pre-)length spaces, a synthetic approach to Lorentzian geometry. They also studied timelike sectional curvature bounds through timelike triangle comparison. A burst of activity in the field of non-smooth Lorentzian geometry followed: Cavalletti and Mondino [ 19 ] developed optimal transport tools and a theory of synthetic timelike lower Ricci curvature bounds for Lorentzian length spaces (building on [ 36 ] and [ 42 ] that treated the case of smooth Lorentzian manifolds), together with a synthetic Hawking singularity theorem. Mondino and Suhr proposed a synthetic definition of upper timelike Ricci curvature bounds leading to a synthetic formulation of the Einstein equations in [ 42 ] . In the flavour of the celebrated singularity theorems on smooth spacetimes, main developments in the non-smooth case include the study of cones and related singularity theorems [ 1 ] , the relationship between spacetime inextendability and synthetic curvature blow-up established in [ 26 ] , singularity theorems under variable timelike lower Ricci curvature bounds [ 9 ] , and an extension of the Penrose singularity theorem to continuous spacetimes satisfying a synthetic null energy condition [ 17 ] . Further developments include a Lorentzian splitting theorem for Lorentzian length spaces with non-negative timelike sectional curvature [ 8 ] , a definition of Lorentzian Hausdorff measures and dimension [ 35 ] , the study of time functions on Lorentzian length spaces [ 13 ] , a definition of timelike lower Ricci curvature bounds on Finsler spacetimes [ 10 ] , and a new Lorentzian isoperimetric-type inequality under timelike lower Ricci curvature bounds [ 18 ] . In [ 7 ] , the authors introduce the notion of metric measure spacetimes, a framework largely consistent with the works by Kunzinger-Sämann, Cavalletti-Mondino, and Mondino-Suhr. In this setting, they prove a d’Alembertian comparison (see also [ 11 ] ) and develop a Lorentzian differential first order calculus. Based on that, they define a synthetic criterion, called infinitesimal Minkowskianity , to distinguish those metric measure spacetimes whose time separation arises from a Lorentzian inner product from Lorentz-Finsler structures. This criterion can be seen as a Lorentzian counterpart of infinitesimal Hilbertianity, which was introduced in [ 5 , 24 ] for metric measure spaces and which provided a natural framework for remarkable structural results [ 40 , 12 ] when combined with Lott-Sturm-Villani’s \mathsf{CD}(K,N) condition. A vast class of metric measure spacetimes arises from manifolds M endowed with non-smooth Lorentzian metrics g . They have been studied in [ 21 , 45 , 25 , 16 ] . Such manifolds serve to describe various phenomena from general relativity, examples being the inside or outside of a star, spacetimes with conical singularities, and cosmic strings ( [ 48 , 49 ] ). In this paper, we study the case that g is continuous. A physically relevant example is provided by Penrose’s impulsive gravitational waves (see for instance [ 43 ] and [ 27 , Chapter 20] ). Both metric measure spacetimes and manifolds endowed with continuous Lorentzian metrics can be seen as generalisations of smooth Lorentzian manifolds, hence a naturally emerging question is whether these two generalisations are compatible. In Riemannian signature, such compatibility questions were studied in [ 14 , 29 , 41 , 30 , 22 ] . The goal of this paper is to compare the classical first order calculus on manifolds with the first order calculus on metric measure spacetimes developed in [ 7 ] . This work can be seen as a Lorentzian counterpart to [ 14 ] and [ 41 , Section 4] . The main result of this paper is: Theorem 1 . Let M be a (d+1) -dimensional manifold and g be a continuous Lorentzian metric on M such that (M,g) is a causally simple spacetime. Then the metric measure spacetime (M,\ell_{g},\mathrm{vol}_{g}) is infinitesimally Minkowskian. Here, the causal simplicity is an assumption of technical nature as we merely use strong causality in our proofs, see Remark 4.6 . The proof of Theorem 1 essentially consists of two steps: • We verify that for any sufficiently regular causal curve \gamma\colon[0,1]\to M , and almost every t\in[0,1] , the causal speed |\dot{\gamma}_{t}| which is almost everywhere given by |\dot{\gamma}_{t}|=\lim_{s\searrow t}\frac{\ell(\gamma_{t},\gamma_{s})}{|s-t|} essentially coincides with the Lorentzian length of the derivative |\dot{\gamma}_{t}|_{g}=\sqrt{-g(\dot{\gamma}_{t},\dot{\gamma}_{t})} . This step is contained in Corollary 4.16 . • We prove that, for a causal function f , the maximal weak subslope is given by |\nabla_{g}f|_{g} , wherever f is finite and differentiable, and +\infty wherever f is infinite or not differentiable. The idea of the proof is similar to the proof in the Riemannian signature [ 41 , Proposition 4.24] but in this setting an extra difficulty arises as we may work with functions that are not C^{1} . We use that causal functions are of bounded variation (see [ 7 ] ), so the distributional derivative of f is given by a measure \mathcal{D}f , which potentially has a singular part \mathcal{D}f^{s} . Using the Besicovitch covering theorem allows us to suitably localise away from the singular part \mathcal{D}f^{s} (Lemma 3.2 ) and then perform a contradiction argument in Lemma 4.22 . The paper is organised as follows: In Section 2 , we recall the synthetic Lorentzian theory from [ 7 ] . Thereafter, in Section 3 , we establish relevant measure theoretic results that will be needed in the proof of Theorem 1 . Finally, in Section 4 , we recall how a manifold with a continuous Lorentzian metric enters the class of metric measure spacetimes and then compare the synthetic first order calculus to the classical manifold calculus to prove Theorem 1 . Notation Given m\in\mathbb{N}

Read the complete text