Pith. sign in

REVIEW 6 minor 31 references

Null distance on cosmological spacetimes and monotone convergence

T0 review · 0 major / 6 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Monotone sequences of cosmological spacetimes have null distances that converge uniformly to a compact limit, and the limit's causal regularization recovers the causal information of the possibly non-smooth limit metric.

desk verdict A solid, genuinely useful contribution to the null-distance convergence program: explicit completion for cosmological spacetimes, a monotone-convergence theorem, and an answer to the causally-null question for limit spaces; the main lines check out, with only presentation-level blemishes. read the letter →

arxiv 2607.20063 v1 pith:KLESPGFJ submitted 2026-07-22 math.DG gr-qcmath.MG

Miguel Prados-Abad , Omar Zoghlami This is my paper · ORCID
classification math.DGgr-qcmath.MG MSC 53B3053C2353C50
keywords nulldistancecosmologicalspacetimestimedmetricspacesGromov-Hausdorffconvergencefuturedevelopedcausally-nullcompactifiablemonotonelimittensor
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proves that a monotone sequence of cosmological spacetimes with compact spatial slices—Lorentzian manifolds of the form (0, τmax) × M with metric −dt² + h_t—has null distances that converge uniformly to a compact limit distance, provided the spatial metrics increase and stay bounded by one fixed Riemannian metric. From this it derives convergence of the associated timed metric spaces in the future-developed and timed-Hausdorff senses. Its central result identifies the causal information of the limit: the causally-null distance induced by the limit distance equals the null distance of the possibly non-smooth pointwise limit metric tensor. This matters because the plain limit distance can fail to be causally-null; the causal regularization recovers what the limit forgets. The result gives a clean sufficient condition for causal-structure preservation in spacetime convergence.

What carries the argument

The central mechanism is the sandwich comparison of null distances: monotonicity of the spatial metrics implies monotonicity of the null distances, so d̂_1 ≤ d̂_j ≤ d̂_{g̃} for every j, where g̃ is built from the uniform bound h̃. For compact M, both d̂_1 and d̂_{g̃} are bi-Lipschitz equivalent to the taxi product distance |t−s| + d_{t0}(x,y) on [0, τmax] × M, forcing d∞ to be compact and the convergence uniform. The proof then uses a limit-curve argument to identify the causal relation of the limit as the intersection of the approximating causal relations, and shows that this relation is exactly the one induced by the pointwise limit tensor g∞.

What would settle it

Compute the null-distance limit for a monotone sequence without the uniform bound, e.g., h_j(t) = (1 + t/j)h on a compact manifold; the pointwise limit distance is non-compact, so if Theorem 1.2 were to hold without assumption (B), it would fail.

Watch

Extended reading notes

Core claim

The paper establishes Theorem 1.4: for a sequence of cosmological spacetimes (N, g_j) with compact slices and metrics g_j = −dt² + h_{j,t} satisfying assumption (B)—the spatial metrics are increasing in j and uniformly bounded above by a fixed Riemannian metric—the pointwise limit distance d∞ is compact, the sequence converges in the future-developed and timed-Hausdorff senses, and the causally-null distance d̂_{d∞,τ} of the limit space coincides with the extension of the null distance d̂∞ of the possibly non-smooth pointwise limit metric tensor g∞. In other words, even when the plain limit distance is not causally-null (as Example 5.8 shows), the causally-null regularization still retains t

Load-bearing premise

The argument hinges on assumption (B): the existence of a single Riemannian metric h̃ on the compact manifold M such that h_{j,t} ≤ h̃ uniformly, together with the compactness of M.

Editorial extensions

If this is right

  • Cosmological spacetimes with compact slices and continuous metric extension to the boundary are causally-null compactifiable, with metric completion bi-Lipschitz to [0, τmax] × M under a taxi distance.
  • For monotone sequences satisfying a uniform upper bound, the null distances converge uniformly and the full sequence of timed metric spaces converges in the future-developed and timed-Hausdorff senses.
  • The causal relation of the limit space is exactly the intersection of the causal relations of the approximating spacetimes.
  • The causally-null distance of the limit space equals the null distance of the possibly non-smooth limit metric tensor, even when the plain limit distance is not causally-null.
  • Without the uniform upper bound, limit spaces can be non-compact and non-product (Example 5.5), so the compactness conclusion genuinely depends on assumption (B).

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same sandwich argument may extend to sequences whose spatial metrics only converge in measure or L¹, as long as monotonicity and the uniform upper bound hold, giving a route to causal-structure preservation under weaker regularity than uniform tensor convergence.
  • The causally-null accessibility condition that makes completions causally-null could serve as a practical criterion for detecting whether a timed metric space has 'dead' boundary points that disconnect the causal relation.
  • These results give a template for checking, in numerical relativity or approximate Lorentzian models, that a monotone family of discrete approximating spacetimes preserves causal order even if the metric converges only pointwise to a non-smooth tensor.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

0 major / 6 minor

Summary. The paper studies null distances on cosmological spacetimes with compact slices, i.e. manifolds (0,τmax)×M with Lorentzian metrics -dt² + h_t. Under a continuous-extension assumption (A), Theorem 1.1 identifies the metric completion of the null-distance space with [0,τmax]×M equipped with a taxi distance, via explicit bi-Lipschitz bounds. For monotone sequences of metrics satisfying a uniform upper tensor bound (assumption (B)), Theorem 1.2 establishes pointwise and uniform convergence of the null distances, future-developed Gromov–Hausdorff convergence, and timed-Hausdorff convergence. Theorems 1.3 and 1.4 then show that the causally-null distance of the limit distance space coincides with the null distance of the (possibly non-smooth) pointwise limit metric tensor. Several examples test the necessity of the hypotheses, including a non-compact limit without (B), a non-causally-null completion without an accessibility condition, and a discontinuous limit tensor requiring absolutely continuous causal curves.

Significance. If the results hold, this is a substantial step in the Sakovich–Sormani null-distance convergence program. It gives the first explicit topological description of null-distance completions for cosmological spacetimes with compact slices, and a monotone-convergence theorem analogous to Perales–Sormani but in the Lorentzian/null-distance setting. The causally-null regularization result (Theorem 1.4) is conceptually interesting: it shows that the limit distance can fail to be causally-null while its causally-null regularization still recovers the null distance of the limit tensor. The paper has clear strengths: a self-contained alternative proof of Nigri's theorem (Theorem 3.5), explicit global bi-Lipschitz estimates, clean compactness arguments, and examples that sharply separate the hypotheses. I found no circularity or parameter-fitting; the assumptions are stated precisely and the proofs are largely checkable.

minor comments (6)
  1. [Example 5.6] The triangle is defined by vertices (0,-1), (1,0), (0,1+ε), but the boundary point r is written as (1+ε,0). This point has t=1+ε, x=0 and is not on the boundary of the described triangle. Presumably r should be (0,1+ε) (or the vertices should be changed). This makes the subsequent angle discussion and the figure caption confusing.
  2. [Abstract / Assumption (B)] The abstract states that the paper treats monotone sequences 'with a uniform upper bound on the spatial diameter', but assumption (B) requires a uniform upper bound on the spatial metric tensors: h_{j,t} ≤ \tilde{h} on TM. This is a stronger condition than a diameter bound. The wording should be aligned with the actual hypothesis, e.g. 'uniform upper bound on the spatial metrics'.
  3. [Proposition 4.10 / Theorem 1.4] The proof that the completed space (\bar{N}, \hat{d}_∞) is causally-null is dispatched with the single sentence 'By the same reasoning as in Theorem 1.3 one can see...'. Since this is the key step linking \hat{d}_{d∞,τ} to \hat{d}_∞, please spell out the accessibility check for boundary points with respect to \hat{d}_∞ (e.g. that the vertical curves are timelike for g∞ and hence give the required ≤_{ \hat{d}_∞,τ } relations).
  4. [Proposition 4.6 proof] In the proof of Proposition 4.6, the sentence 'we only need to prove the opposite inequality, i.e., that \hat{d} ≥ \hat{d}_{\hat{d},τ}' states the wrong direction. Since [CP26, Proposition 1.9] gives \hat{d}_{\hat{d},τ} ≥ \hat{d}, the missing inequality is \hat{d}_{\hat{d},τ} ≤ \hat{d}, which is what the subsequent argument actually proves. The wording should be corrected.
  5. [Theorem 3.5 proof] In the estimate after constructing the ODE solutions, the display ends with '= M(K+˜δ)|c_k−c_{k+1}| = M(K+˜δ)/N'. Since |c_k−c_{k+1}| = (d-c)/N, the factor (d-c) has been omitted. The choice of N should read N > M(K+˜δ)(d-c)/min(ε, δ_t/2). This is a typo that does not affect the argument, but the formula as written is incorrect.
  6. [Example 5.9] The example should state explicitly that the smooth approximants f_j can be chosen to satisfy the continuous extension condition (A) at t=0 and t=1 (for instance by taking f_j(0)=f_j(1)=1), so that the sequence falls under the hypotheses of Proposition 4.10. Also, the remaining-measure expression should be typeset unambiguously, e.g. (1−3a)/(1−2a), rather than the current ambiguous inline formula.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the new theorems are proved from stated assumptions and external published results, with no self-authored citations carrying the argument.

full rationale

The derivation chain is self-contained in the relevant sense. Theorem 1.1 is proved directly via Lemma 3.3, Corollaries 3.8/3.10 and a bi-Lipschitz comparison with a taxi distance. Theorem 1.2 follows from Theorem 4.2 (monotone uniform convergence), Theorem 4.4 (future-developed property) and Proposition 4.3 (metric-pair convergence). Theorem 1.4 is obtained from Propositions 4.8 and 4.10, which establish equality of causal relations by elementary limit arguments plus the standard limit-curve theorem [Min08]. No parameter is fitted to data and then renamed a prediction; the quantity d_infinity is genuinely a pointwise limit of the null distances, and the equality in Theorem 1.4 is proved rather than assumed. The only result credited to another source, Theorem 3.5, is attributed to Nigri [Nig25] and the paper supplies an independent proof. The black-box use of published theorems such as [PS25, Theorem 3.2], [Per26], [CP26], [Min08], [KS18], and [BG24] is standard mathematical practice, not circularity, and none of these is a self-citation of the present authors. No uniqueness theorem from the authors' own prior work is invoked, and no ansatz is smuggled in through a citation. The examples explicitly demonstrate the necessity of the hypotheses, confirming that the results have independent content. I therefore find no circular step.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The central claims rest on the structural class definition (cosmological spacetimes), two regularity/bound hypotheses (A) and (B), compactness of M, and a dozen cited black-box theorems. No free parameters: nothing is fitted or empirically calibrated. No invented entities: the technical vocabulary (null distance, causally-null, timed metric spaces, future developed) is inherited from prior literature; the 'taxi distance' is a proof tool, not a postulate.

assumptions (6)
  • domain assumption N = (t1,t2) x M with g = -dt^2 + h_t and h_t(∂t,·) = 0 ('cosmological spacetime', Definition 2.1).
    The entire paper is restricted to this class; it is what makes the coordinate t equal to the cosmological time and the null distance explicitly computable.
  • domain assumption M is compact (compact slices).
    Used in Lemma 3.1 (regularity of cosmological time), Lemma 3.3 (compact unit tangent bundle), Lemma 3.7 (diameter bounds), and Proposition 4.10's limit-curve argument.
  • domain assumption Assumption (A): h_t extends continuously to t = t1, t2 as a Riemannian metric.
    Introduced in Section 1; needed so Lemma 3.3's comparison functions live on the closed interval and Theorem 1.1's bi-Lipschitz constants are finite.
  • domain assumption Assumption (B): h_{j,t} <= h_{j+1,t} for all j, and a fixed Riemannian metric h-tilde with h_{j,t} <= h-tilde for all j,t.
    Assumption (B), Section 1; monotonicity of metrics implies monotonicity of null distances (AB22 Lemma 4.15), and the uniform bound yields the compact sandwich limit in Theorem 4.2.
  • domain assumption tau = t is a regular cosmological time and the null distance encodes causality on each (N, g_j).
    Lemma 3.1 plus cited results (AGH98 Prop 2.7, SV16/SS23 Def 2.5, BG24 Prop 2.9); needed for definiteness of d-hat_j and for the chain arguments in Propositions 4.6 and 4.10.
  • standard math Black-box theorems: PS25 Thm 3.2, Per26 Thm 1.5, CP26 Prop 1.9 and 5.6, Min08 Thm 3.1, KS18 Lemma 2.21, Gra+20 AC-causality, SV16, AB22 Lemma 4.15, Nig25 Thm 1.2.
    Invoked at the locations cited in Sections 2-4; the paper relies on them rather than re-proving them.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Null distance on cosmological spacetimes and monotone convergence." pith.science (2026). https://pith.science/paper/KLESPGFJ

@misc{pith2026260720063,
  author       = {Pith},
  title        = {Pith review of: Null distance on cosmological spacetimes and monotone convergence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KLESPGFJ}},
  note         = {Machine review of arXiv:2607.20063}
}
abstract

The metric theory of spacetimes studies Lorentzian manifolds using tools of metric geometry. This is achieved via the null distance, which is a definite distance constructed from a time function on a spacetime. This enables the study of Gromov-Hausdorff-type convergence of spacetimes, a program recently initiated by Sakovich and Sormani. In this paper we study such notions of convergence for cosmological spacetimes with compact slices, i.e., $(a,b)\times M$ endowed with a Lorentzian metric $-dt^2+h_t$, where $h_t$ is a family of Riemannian metrics on the compact manifold $M$. Assuming mild extension properties of $h_t$, we first establish that these spacetimes are causally-null compactifiable and future developed. We then study monotone sequences with a uniform upper bound on the spatial diameter, obtaining uniform convergence of the null distances, as well as convergence of the associated timed metric spaces in the future developed Gromov-Hausdorff sense. Finally, we prove that causally-null compactifiable spacetimes satisfying a mild causal accessibility condition are causally-null, and relate the causally-null distance induced by the limit distance with the null distance induced by the (possibly non-smooth) limit metric tensor. Examples are provided to motivate the necessity of our hypotheses.

Figures

Figures reproduced from arXiv: 2607.20063 by the authors.

Figure 1
Figure 1. Situation described in the proof of Lemma 5.2. The point p ∗ has greater t-component than those of p1 and p2, from which one deduces that the piece of the curve from p − to p + cannot be minimal. Therefore, the whole curve cannot be minimal. then there exists a point x ∗ ∈ M such that r := (t ∗ , x∗ ) is in the intersection of the causal pasts of p and q. Indeed, consider a minimal curve β in M joining xp and xq. If… view at source ↗
Figure 2
Figure 2. Situation described in Example 5.6. The point r, in the metric boundary of the triangle, is not related to any point in the completion with the causal relation ≤d,τ ˆ . Therefore, r is at infinite ˆdd,τ ˆ -distance from any other point in △¯ . Remove p q [PITH_FULL_IMAGE:figures/full_fig_p031_2.png] view at source ↗
Figure 3
Figure 3. Situation described in Example 5.7. Points p and q are timelike related with respect to any Lorentzian metric ga = −dt2 + a 2dx2 with a < 1, but they are not even causally related with respect to g1. Example 5.7. In Proposition 4.10 one proves the equality T j ≤j=≤∞ for the case of a sequence of cosmological spacetimes. In this example we show that the inclusion ⊆ is not true for general spacetimes. Specifically, co… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Future cone of the origin in the non-compact case [PITH_FULL_IMAGE:figures/full_fig_p032_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 7 canonical work pages

  1. [1]

    Monotone Sequences of Metric Spaces with Compact Limits

    Monotone. doi:10.48550/arXiv.2406.18795 , author =

    work page Pith review arXiv doi:10.48550/arxiv.2406.18795

  2. [2]

    doi:10.48550/arXiv.2410.16800 , author =

    Introducing. doi:10.48550/arXiv.2410.16800 , author =

    doi:10.48550/arxiv.2410.16800

  3. [3]

    Sakovich, Anna and Sormani, Christina , TITLE =. J. Math. Phys. , FJOURNAL =. 2023 , NUMBER =

    2023

  4. [4]

    Burtscher, Annegret and Garc\'ia-Heveling, Leonardo , TITLE =. Comm. Math. Phys. , FJOURNAL =. 2024 , NUMBER =

    2024

  5. [5]

    Perales, Raquel , TITLE =. Bol. Soc. Mat. Mex. (3) , FJOURNAL =. 2026 , NUMBER =

    2026

  6. [6]

    Grant, James D. E. and Kunzinger, Michael and S\"amann, Clemens and Steinbauer, Roland , TITLE =. Lett. Math. Phys. , FJOURNAL =. 2020 , NUMBER =

    2020

  7. [7]

    Minguzzi, Ettore , TITLE =. J. Math. Phys. , FJOURNAL =. 2008 , NUMBER =

    2008

  8. [8]

    Kunzinger, Michael and S\"amann, Clemens , TITLE =. Ann. Global Anal. Geom. , FJOURNAL =. 2018 , NUMBER =

    2018

Show all 31 references
  1. [9]

    , TITLE =

    Galloway, Gregory J. , TITLE =. Classical Quantum Gravity , FJOURNAL =. 2024 , NUMBER =

    2024

  2. [10]

    and Sánchez, M

    Minguzzi, E. and Sánchez, M. , year =. The causal hierarchy of spacetimes , copyright =. doi:10.48550/ARXIV.GR-QC/0609119 , abstract =

    doi

  3. [11]

    Allen, Brian and Burtscher, Annegret , TITLE =. Int. Math. Res. Not. IMRN , FJOURNAL =. 2022 , NUMBER =

    2022

  4. [12]

    Classical Quantum Gravity , FJOURNAL =

    Sormani, Christina and Vega, Carlos , TITLE =. Classical Quantum Gravity , FJOURNAL =. 2016 , NUMBER =

    2016

  5. [13]

    and Howard, Ralph , TITLE =

    Andersson, Lars and Galloway, Gregory J. and Howard, Ralph , TITLE =. Classical Quantum Gravity , FJOURNAL =. 1998 , NUMBER =

    1998

  6. [14]

    and S\'anchez, Miguel , TITLE =

    Bernal, Antonio N. and S\'anchez, Miguel , TITLE =. Comm. Math. Phys. , FJOURNAL =. 2003 , NUMBER =

    2003

  7. [15]

    and S\'anchez, Miguel , TITLE =

    Bernal, Antonio N. and S\'anchez, Miguel , TITLE =. Comm. Math. Phys. , FJOURNAL =. 2005 , NUMBER =

    2005

  8. [16]

    S\'anchez, Miguel , TITLE =. Gen. Relativity Gravitation , FJOURNAL =. 2022 , NUMBER =

    2022

  9. [17]

    and Grant, James D

    Chru\'sciel, Piotr T. and Grant, James D. E. and Minguzzi, Ettore , TITLE =. Ann. Henri Poincar\'e , FJOURNAL =. 2016 , NUMBER =

    2016

  10. [18]

    Che, Mauricio and Galaz-Garc\'ia, Fernando and Guijarro, Luis and Membrillo Solis, Ingrid Amaranta , TITLE =. J. Appl. Comput. Topol. , FJOURNAL =. 2024 , NUMBER =

    2024

  11. [19]

    2025 , DOI=

    Null Distance and Temporal Functions , author=. 2025 , DOI=

    2025

  12. [20]

    Kunzinger, Michael and Steinbauer, Roland , TITLE =. Ann. Henri Poincar\'e , FJOURNAL =. 2022 , NUMBER =. doi:10.1007/s00023-022-01198-6 , URL =

    2022 doi

  13. [21]

    Allen, Brian , TITLE =. Gen. Relativity Gravitation , FJOURNAL =. 2023 , NUMBER =. doi:10.1007/s10714-023-03167-8 , URL =

    2023 doi

  14. [22]

    Living Reviews in Relativity , author =

    Lorentzian causality theory , volume =. Living Reviews in Relativity , author =. 2019 , pages =. doi:10.1007/s41114-019-0019-x , number =

    2019 doi

  15. [23]

    doi:10.48550/arXiv.2604.22362 , author =

    Convergence of. doi:10.48550/arXiv.2604.22362 , author =

    doi arXiv

  16. [24]

    Choquet-Bruhat, Yvonne and Geroch, Robert , TITLE =. Comm. Math. Phys. , FJOURNAL =. 1969 , PAGES =

    1969

  17. [25]

    Differential Geom

    Che, Mauricio and Perales, Raquel and Sormani, Christina , TITLE =. Differential Geom. Appl. , FJOURNAL =. 2026 , PAGES =. doi:10.1016/j.difgeo.2026.102364 , URL =

    2026

  18. [26]

    , TITLE =

    Burtscher, Annegret Y. , TITLE =. New York J. Math. , FJOURNAL =. 2015 , PAGES =

    2015

  19. [27]

    and Grant, James D

    Chru\'sciel, Piotr T. and Grant, James D. E. , TITLE =. Classical Quantum Gravity , FJOURNAL =. 2012 , NUMBER =

    2012

  20. [28]

    2025 , keywords =

    Lorentzian. 2025 , keywords =. doi:10.48550/arXiv.2504.10380 , author =

    2025 doi arXiv

  21. [29]

    Minguzzi, Ettore and Suhr, Stefan , TITLE =. Lett. Math. Phys. , FJOURNAL =. 2024 , NUMBER =. doi:10.1007/s11005-024-01813-z , URL =

    2024 doi

  22. [30]

    Bykov, Aleksei and Minguzzi, Ettore and Suhr, Stefan , TITLE =. Lett. Math. Phys. , FJOURNAL =. 2025 , NUMBER =. doi:10.1007/s11005-025-01941-0 , URL =

    2025 doi

  23. [31]

    Convergence of

    Ketterer, Christian , year =. Convergence of. doi:10.48550/arXiv.2605.11271 , abstract =

    doi arXiv

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.