A localized construction of Kasner-like singularities

Grigorios Fournodavlos; Nikolaos Athanasiou

REVIEW 3 major objections 4 minor 49 references

A localized construction of Kasner-like singularities

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper constructs local singular solutions to the Einstein vacuum equations whose past boundary is Kasner-like, and proves uniqueness and smoothness of such solutions.

desk verdict A substantial, carefully argued localization of the Fournodavlos–Luk construction; the main theorem is believable, with two localized soft spots (an omitted computation in Lemma 4.11 and a terse gluing argument in Corollary 1.1) that should be fixed but don't sink the paper. read the letter →

arxiv 2412.16630 v1 pith:77GY54D5 submitted 2024-12-21 math.AP gr-qcmath-phmath.DGmath.MP

Nikolaos Athanasiou , Grigorios Fournodavlos This is my paper · ORCID

classification math.APgr-qcmath-phmath.DGmath.MP MSC 35Q7635L4583C0583C75

keywords EinsteinvacuumequationsKasnersingularitiessingularityformationsymmetrichyperbolicsystemsasymptoticdatalocalizedconstructionbigbangLorentzianmetrics

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

The paper proves that, on a spatial cube, any smooth asymptotic data set of Kasner type—functions $p_i(x)$, $c_{ij}(x)$ satisfying the Kasner relations and a differential constraint—can be extended to an actual local solution of the Einstein vacuum equations with a spacelike singularity at $t=0$. The constructed metric has the leading form $-dt^2 + \sum_i t^{2p_i(x)}\,\omega^i\omega^i$, with corrections that vanish to arbitrarily high polynomial order as $t\to 0$. The novelty is that the construction is localized in space: it uses a first-order symmetric hyperbolic formulation of the field equations in the connection coefficients of a parallelly propagated orthonormal frame, so no elliptic estimates are needed and the energy argument works in a local domain with spacelike boundary. The same data also determine the solution uniquely, and smooth data produce smooth solutions.

What carries the argument

The argument is carried by a first-order symmetric hyperbolic system for the connection coefficients $k_{IJ}$ and $\gamma_{IJB}$ of an orthonormal frame that is parallelly propagated along $\partial_t$; here $k_{IJ}$ is the second fundamental form of the constant-time slices and $\gamma_{IJB}$ the spatial connection coefficients. This formulation replaces the third-order metric formulation used in the global construction and eliminates the derivative loss that forced elliptic estimates. Around it, the proof layers an iteration scheme that builds an approximate solution $g^{(n)}$ with Ricci tensor decaying as $t^{-2+n\varepsilon}$, weighted $H^s$ energy estimates with large $t$-weights to control the remainder, and a domain whose spacelike boundary is chosen so that all boundary terms in the energy identity have the favorable sign.

What would settle it

Take spatially homogeneous asymptotic data with constant $p_i$ and $c_{ij}$ satisfying the constraints, for which the explicit Kasner metric is an exact vacuum solution; if the iterative approximate solutions $g^{(n)}$ constructed in Section 4 do not converge to that metric, or the remainder bound $\|g^{(d)}\|_{H^s(U_t)}^2 \le t^{2N_0}$ fails, the existence theorem is wrong.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that Kasner-like asymptotic data are not just formal: for every smooth triplet $(p_1,p_2,p_3,c_{ij})$ satisfying Definition 1.1, there is a $C^2$ Lorentzian metric $g$ on a local domain $\{U_t\}$ with spacelike future boundary that solves $R_{\mu\nu}=0$ and has the prescribed leading behavior (1.1) as $t\to 0$. The metric is produced as $g^{(n)}+g^{(d)}$, where $g^{(n)}$ is an explicit iterative approximate solution whose Ricci tensor decays like $t^{-2+n\varepsilon}$ and $g^{(d)}$ is a remainder whose $H^s$ norm on the time slice is bounded by $t^{2N_0}$, with $\varepsilon=\min\{1-p_3,\,p_3-p_2\}>0$. Theorems 1.2 and 1.3 add that the solution is smooth when the data are smooth and that any two solutions with the same asymptotic data coincide, so the data genuinely parametrize the local singularity.

Load-bearing premise

The construction needs the three Kasner exponents to stay strictly ordered with $p_2(x) < p_3(x) < 1$ everywhere, so that $\varepsilon = \min\{1-p_3,\, p_3-p_2\}$ is a positive number; every decay estimate in the proof is a power of $t^\varepsilon$, and if $p_2$ and $p_3$ touch or $p_3$ reaches 1 the argument yields no decay.

Editorial extensions

If this is right

Any admissible local asymptotic data set of Kasner type is realized by an actual vacuum spacetime, not merely by a formal expansion.
Two solutions with the same asymptotic data are the same, so the asymptotic data genuinely parametrize the local singularity.
Smooth asymptotic data lead to smooth solutions, so the construction produces classical solutions, not just weak ones.
Because the construction is local and carries uniqueness, Kasner-like patches can be glued along overlaps to form a global singular spacetime on a closed spatial manifold from covariant data, as stated in Corollary 1.1.
Increasing the iteration order $n$ makes the approximate solution vanish the Ricci tensor to any desired polynomial rate, which is what allows the remainder to be controlled uniformly as $t\to 0$.

Reading between the lines

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

The strict separation $\varepsilon>0$ is load-bearing; the paper sets aside contact points where $p_2=p_3$ or $p_3=1$. A natural extension would be to track whether the construction can be modified with logarithms or different weights at such points.
Because the proof avoids elliptic estimates, the same first-order hyperbolic frame should adapt to lower-regularity data or to Einstein equations coupled to matter, with only the constants adjusting.
One could use the local uniqueness theorem to match a Kasner-like patch against a patch with oscillatory behavior near the singularity, building a spacetime whose singularity changes character from point to point; the paper mentions this as motivation but does not carry it out.

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.

Desk Editor's Note

read the letter

Dear colleague,

Quick take: this is a genuine advance, not just a repackaging. It localizes the global T^3 Kasner-like construction, drops elliptic estimates by working with a first-order symmetric hyperbolic system in a parallelly propagated frame, adds a cleaner existence statement for asymptotic data (Prop 2.1), and proves a sharper uniqueness result. The iterative construction is written in detail; the weighted energy estimates for the remainder are intricate and, as far as I can tell, internally consistent. I believe Theorem 1.1 is very likely correct.

The soft spots are two, both localized. First, Lemma 4.11's r=1 case ("We omit the details") is exactly what feeds Proposition 4.3 and Lemma 5.2, so it is load-bearing, not cosmetic. The computation is plausibly a tedious bookkeeping exercise along the lines of the r=0 case, but the paper should either include it or at least detail the cancellation structure; a referee should check this carefully. Also, the indices in Lemma 4.11 carry |pI-pJ|; the proof of Proposition 4.3 should spell out why these extra powers are harmless for the n chosen later (for large n, t^{-2+nε} is weaker than the |pI-pJ|-dependent bound, but this needs to be said). Second, Corollary 1.1's gluing over T^3 is sketched in four sentences. The compactness argument is fine, but the claim that overlapping patches coincide by Theorem 1.3 needs a common coordinate/gauge discussion; as written, the transition from local charts to a global solution is too quick. Both issues look fixable, not fatal.

The paper is honest about its limitations: contact points between Kasner exponents are excluded, and the data are local. The citations to [18] and [20] are appropriate tool citations, not self-promotion.

Verdict: this deserves a serious referee. I'd suggest a conditional accept with a request to fill in Lemma 4.11 and expand the gluing proof. It is a paper I would bring to a reading group and would cite once the details are confirmed.

Referee Report

3 major / 4 minor

Summary. The paper constructs local, in space, singular solutions to the Einstein vacuum equations in 1+3 dimensions with prescribed Kasner-like asymptotic data, using a first-order symmetric hyperbolic ADM-type formulation relative to a parallelly propagated orthonormal frame. The main theorem (Theorem 1.1) builds an iterative approximate solution g[n] whose spacetime Ricci tensor decays like t^{-2+nε} for arbitrary n, then solves for a small remainder in weighted H^s spaces on a localized domain with favorable boundary terms, obtaining a C^2 metric satisfying R_{\mu\nu}=0 with remainder of order t^{N_0}. The paper also proves smoothness (Theorem 1.2), a refined uniqueness statement under pointwise asymptotic bounds (Theorem 1.3), an existence result for asymptotic data satisfying the constraint conditions (Proposition 2.1), and a local-to-global corollary for covariant data (Corollary 1.1). The main novelty relative to prior work is the use of a first-order symmetric hyperbolic system, which removes the need for elliptic estimates and permits a localized energy argument.

Significance. If the construction is fully correct, this is a substantial contribution: it localizes the earlier global Kasner-like singularity construction of Fournodavlos--Luk, removes elliptic estimates from the proof, provides a flexible generation of asymptotic data, and gives a uniqueness statement that works under relatively weak pointwise assumptions. The paper is carefully structured, with detailed iterative estimates, weighted energy inequalities, and a clear separation between the approximate solution and remainder problem. The ε positivity in (4.10) is automatic from the Kasner algebraic relations and strict ordering on a compact domain, so the reader's concern about an 'ε-gap' is not an actual weakness. However, the proof as written has one explicitly omitted r=1 computation in Lemma 4.11 that is load-bearing for the approximate-solution accuracy, and the gluing argument in Corollary 1.1 is only sketched.

major comments (3)

[§4.4, Lemma 4.11] The r=1 estimate in (4.60) is asserted with the sentence 'We omit the details.' This estimate controls ∂t(k[n]_IJ - ~k[n]_IJ), and it is directly used in Proposition 4.3 to convert the evolution-error bound of Lemma 4.10 into the spacetime Ricci decay |∂_x^α R[n]_IJ| ≤ C_{α,n} t^{-2+nε}, which is point 3 of Theorem 4.1 and the first display of Theorem 1.1. Lemma 5.2 also relies on the r=1 bound through the term ∂t(k[n]_IJ - ~k[n]_IJ) in (I[n]_k)_IJ, and the large-M decay of that term is what makes the remainder estimate (5.26) hold for arbitrarily large N0. Since the r=1 computation is load-bearing for the central existence theorem, the omitted details should be supplied or replaced by a complete alternative argument.
[§7, Corollary 1.1] The proof of Corollary 1.1 is only a short paragraph and does not spell out the gluing mechanism. Theorem 1.3 is stated and proved for two solutions written in a common gauge, with a common approximate metric g[n] and a common domain {U_t} defined relative to that g[n]. In the finite cover of Corollary 1.1, different patches are constructed in different coordinate charts and with possibly different n and N0; the proof does not explain how the hypotheses (1.16)-(1.17) of Theorem 1.3 are verified in the overlaps, nor why uniqueness on overlaps yields a single globally defined metric rather than merely compatible local metrics. Since Corollary 1.1 is advertised as a main application, this gluing argument should be written out.
[§7.1, proof of Theorem 1.3(ii)] The implication (1.18) ⇒ (1.17) is completed with the sentence 'We may continue iteratively improving the bounds ... for each tε improvement we sacrifice two spatial derivatives, which is possible provided M1 ∼ M/ε.' As written, the proof does not give a formal induction statement for the derivative loss, and the stated dependence of M1 on ε is not fully justified: the argument makes M1 depend on M0 through the choice of n with M(n) ≥ M0, while M0 in point (i) is tied to the constant C* from the energy estimates. The authors should either make the induction explicit and specify the exact dependence of M1 on ε (and on M0, if needed), or weaken the statement in Theorem 1.3(ii).

minor comments (4)

[§5.3, Lemma 5.3] In the proof of Lemma 5.3, the phrase 'initial data ... are trivial on Uδ' appears to be a typo: the initial slice in the local-existence argument is Uη, and the integral curves of ∂t emanating from Uη rule the domain {U_t}_{t∈[η,T]}; please correct the notation.
[References] References [10] and [11] are both listed with arXiv:2308.07475; if these are distinct papers, one of the identifiers is incorrect and should be fixed.
[§4.1, Lemma 4.1] The estimate for (g[0]_error)_{11} and (g[0]_error)_{12} absorbs logarithms into t^ε by shrinking tn; this is legitimate, but it would be clearer to state explicitly that the final constants depend on the number of derivatives through the log-power exponent.
[§5.1, Eq. (5.1)] The definition of X_a^± uses σ∇[n]x^a / |∇[n]x^a|; since σ is later chosen large to control boundary fluxes, it may help the reader to note at (5.1) that the normalization is taken with respect to g[n], as is done in the proof of Lemma 5.1.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the construction takes Kasner data as free input and proves existence via an explicit iterative scheme; self-citations are not load-bearing.

full rationale

The paper's derivation chain is a genuine construction: given prescribed asymptotic data p_i, c_ij satisfying Definition 1.1, it builds an approximate solution g[n] by an explicit iteration scheme (Section 4) and then solves for a small remainder in a weighted energy argument (Section 5), recovering the vacuum equations in Section 6. The Kasner exponents and metric coefficients are free inputs, not fitted values, and the leading-order behavior t^{2 p_max} is the target of the construction rather than an output derived from hidden assumptions. No step defines a predicted quantity in terms of itself: the approximate solution's leading term is set equal to the ansatz, and the remainder estimates are proved by induction from the iteration equations. The citations to [18] and [20] provide the prior framework and the symmetric hyperbolic formulation, but the paper reproduces the relevant equations and proves the main estimates itself; these citations do not import the target theorem. The only notable gap is in Lemma 4.11, where the r=1 time-derivative bound is asserted with 'We omit the details'; this is a completeness and correctness concern, not a circularity, because it concerns an omitted computation rather than an input that defines the conclusion. Overall, there is no circular reduction of the central claim to its assumptions.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard PDE tools, a gauge choice, and the physical integrability condition for Kasner behavior. No new entities or fitted parameters are introduced; the Kasner exponents are prescribed data subject to algebraic constraints.

assumptions (5)

standard math Standard local well-posedness for first order symmetric hyperbolic systems with smooth coefficients.
Used in Lemma 5.4 to start the remainder solution from a regular slice U_eta; this is a standard textbook result.
domain assumption Gaussian time foliation and a parallelly propagated orthonormal frame exist for metrics of the form g = -dt^2 + g_ij dx^i dx^j.
The entire formulation in Section 1.3 uses such a frame; this is always possible locally for a Lorentzian metric, with the singularity synchronized at t = 0.
domain assumption The asymptotic data satisfy the Kasner algebraic relations and the differential constraint (1.6), equivalently (4.9).
This is the non-generic integrability condition needed for vacuum Kasner behavior; it is imposed in Definition 1.1 and used in Lemma 4.13 to propagate the constraints.
domain assumption The strict ordering p1 < p2 < p3 with epsilon > 0 holds on the data.
All decay rates are powers of t^epsilon; contact points are excluded in Section 1.2 and Definition 1.1 point 2.
domain assumption Ringström's local equivalence between gauge-dependent data (Definition 1.1) and covariant data (Definition 1.2) is valid.
Used in Corollary 1.1 to patch local solutions on T^3; quoted from [44] without proof.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A localized construction of Kasner-like singularities." pith.science (2026). https://pith.science/paper/77GY54D5

@misc{pith2026241216630,
  author       = {Pith},
  title        = {Pith review of: A localized construction of Kasner-like singularities},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/77GY54D5}},
  note         = {Machine review of arXiv:2412.16630}
}

read the original abstract

We construct local, in spacetime, singular solutions to the Einstein vacuum equations that exhibit Kasner-like behavior in their past boundary. Our result can be viewed as a localization (in space) of the construction in \cite{FL}. We also prove a refined uniqueness statement and give a simple argument that generates general asymptotic data for Kasner-like singularities, enjoying all expected degrees of freedom, albeit only locally in space. The key difference of the present work with \cite{FL} is our use of a first order symmetric hyperbolic formulation of the Einstein vacuum equations, relative to the connection coefficients of a parallelly propagated orthonormal frame which is adapted to the Gaussian time foliation. This makes it easier to localize the construction, since elliptic estimates are no longer required to complete the energy argument.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 45 canonical work pages

[18]

Fournodavlos and J

G. Fournodavlos and J. Luk, Asymptotically Kasner-lik e singularities. Amer. J. Math. , 145(4):1183–1272, 2023

work page 2023
[20]

Fournodavlos and J

G. Fournodavlos and J. Smulevici. The initial boundary value problem for the Einstein equations with totally geodesic boundary. Comm. Math. Phys. , 385(3):1615–1653, 2021

work page 2021
[1]

Stable space-like singularity formation for axi-symmetric and polarized near-Schwarzschild black hole interiors

S. Alexakis and G. Fournodavlos. Stable space-like sing ularity formation for axi- symmetric and polarized near-Schwarzschild black hole int eriors. arXiv:2004.00692. 56

work page Pith review arXiv 2004
[2]

E. Ames, F. Beyer, J. Isenberg, and P. G. LeFloch. Quasili near hyperbolic Fuchsian sys- tems and A VTD behavior in T 2-symmetric vacuum spacetimes. Ann. Henri Poincar´ e, 14(6):1445–1523, 2013

work page 2013
[3]

E. Ames, F. Beyer, J. Isenberg, and P. G. LeFloch. A class o f solutions to the Einstein equations with A VTD behavior in generalized wave gauges. J. Geom. Phys. , 121:42–71, 2017

work page 2017
[4]

Andersson and A

L. Andersson and A. D. Rendall. Quiescent cosmological s ingularities. Comm. Math. Phys., 218(3):479–511, 2001

work page 2001
[5]

J. D. Barrow. Quiescent cosmology. Nature, 272:211–215, 1978

work page 1978
[6]

V. A. Belinski and M. Henneaux. The cosmological singula rity. Cambridge University Press, 2018

work page 2018

Show all 49 references

[7]

V. A. Belinski and I. M. Khalatnikov. Eﬀect of scalar and ve ctor ﬁelds on the nature of the cosmological singularity. ˇZ. `Eksper. Teoret. Fiz., 63:1121–1134, 1972

1972
[8]

V. A. Belinski, I. M. Khalatnikov, and E. M. Lifshitz. Osc illatory approach to the singular point in relativistic cosmology. Soviet Physics Uspekhi , 13(6):745–765, 1971

1971
[9]

B´ eguin and T

F. B´ eguin and T. Dutilleul, Chaotic Dynamics of Spatial ly Homogeneous Spacetimes. Comm. Math. Phys. , 399:737–927, 2023

2023
[10]

Beyer, T

F. Beyer, T. A. Oliynyk. Localized big bang stability fo r the Einstein-scalar ﬁeld equations arXiv:2308.07475

arXiv
[11]

Beyer, T

F. Beyer, T. A. Oliynyk. Past stability of FLR W solution s to the Einstein-Euler-scalar ﬁeld equations and their big bang singularities. arXiv:230 8.07475
[12]

Choquet-Bruhat, J

Y. Choquet-Bruhat, J. Isenberg, and V. Moncrief. Topol ogically general U(1) sym- metric vacuum space-times with A VTD behavior. Nuovo Cimento Soc. Ital. Fis. B , 119(7-9):625–638, 2004

2004
[13]

P. T. Chru´ sciel, J. Isenberg, and V. Moncrief. Strong c osmic censorship in polarised Gowdy spacetimes. Classical Quantum Gravity , 7(10):1671–1680, 1990

1990
[14]

Damour, M

T. Damour, M. Henneaux, A. D. Rendall, and M. Weaver. Kas ner-like behaviour for subcritical Einstein-matter systems. Ann. Henri Poincar´ e, 3(6):1049–1111, 2002

2002
[15]

Demaret, M

J. Demaret, M. Henneaux, and P. Spindel. Nonoscillator y behaviour in vacuum Kaluza-Klein cosmologies. Phys. Lett. B , 164(1-3):27–30, 1985. 57

1985
[16]

Fajman and L

D. Fajman and L. Urban. Cosmic Censorship near FLR W spac etimes with negative spatial curvature. arXiv:2211.08052

arXiv
[17]

Fajman and L

D. Fajman and L. Urban. On the past maximal development o f near-FLR W data for the Einstein scalar-ﬁeld Vlasov system. arXiv:2402.08544

arXiv
[19]

Fournodavlos, I

G. Fournodavlos, I. Rodnianski, and J. Speck. Stable Bi g Bang formation for Einstein’s equations: The complete sub-critical regime. J. Amer. Math. Soc. , 36:827–916, 2023

2023
[21]

Franco-Grisales

A. Franco-Grisales. Developments of initial data on bi g bang singularities for the Einstein-nonlinear scalar ﬁeld equations. arXiv:2409.17 065
[22]

H. O. Groeniger, O. Petersen, and H. Ringstr¨ om Formati on of quiescent big bang singularities. arXiv:2309.11370
[23]

J. M. Heinzle, C. Uggla, and W. C. Lim. Spike oscillation s. Phys. Rev. D , 86:104049, 2012

2012
[24]

J. M. Heinzle, C. Uggla, and N. R¨ ohr. The cosmological b illiard attractor. Adv. Theor. Math. Phys. , 13:293–407, 2009

2009
[25]

Isenberg and V

J. Isenberg and V. Moncrief. Asymptotic behaviour in po larized and half-polarized U(1) symmetric vacuum spacetimes. Classical Quantum Gravity , 19(21):5361–5386, 2002

2002
[26]

Kichenassamy and A

S. Kichenassamy and A. D. Rendall. Analytic descriptio n of singularities in Gowdy spacetimes. Classical Quantum Gravity , 15(5):1339–1355, 1998

1998
[27]

P. Klinger. A new class of asymptotically non-chaotic v acuum singularities. Ann. Physics, 363:1–35, 2015

2015
[28]

W. Li. BKL bounces outside homogeneity: Gowdy symmetri c spacetimes. arXiv:2408.12427

arXiv
[29]

W. Li. BKL bounces outside homogeneity: Einstein-Maxw ell-scalar ﬁeld in surface symmetry. arXiv:2408.12434

arXiv
[30]

E. M. Lifshitz and I. M. Khalatnikov. Investigations in relativistic cosmology. Adv. in Physics, 12:185–249, 1963. 58

1963
[31]

J. Lott. On the initial geometry of a vacuum cosmologica l spacetime. arXiv:1908.02185

1908 arXiv
[32]

J. Lott. Kasner-like regions near crushing singularit ies. arXiv:2008.02674

2008 arXiv
[33]

Van de Moortel

M. Van de Moortel. The breakdown of weak null singularit ies inside black holes. Duke Math. J. , 172(15):2957–3012, 2023

2023
[34]

M. Z. A. Moughal and W. C. Lim. Cylindrical spikes. Classical Quantum Gravity , 38(7):075029, 2021

2021
[35]

A. D. Rendall. Fuchsian analysis of singularities in Go wdy spacetimes beyond analyt- icity. Classical Quantum Gravity , 17(16):3305–3316, 2000

2000
[36]

A. D. Rendall and M. Weaver. Manufacture of Gowdy spacet imes with spikes. Class. Quantum Grav., 18(15):2959–2975, 2001

2001
[37]

Reiterer and E

M. Reiterer and E. Trubowitz. The BKL Conjectures for Sp atially Homogeneous Spacetimes. arXiv:1005.4908

arXiv
[38]

Ringstr¨ om

H. Ringstr¨ om. The Bianchi IX attractor. Annales Henri Poincar´ e, 2:405–500, 2001

2001
[39]

Ringstr¨ om

H. Ringstr¨ om. Existence of an asymptotic velocity and implications for the asymptotic behavior in the direction of the singularity in T 3-Gowdy. Comm. Pure Appl. Math. , 59(7):977–1041, 2006

2006
[40]

Ringstr¨ om

H. Ringstr¨ om. Strong cosmic censorship in the case of T 3-Gowdy vacuum spacetimes. Classical Quantum Gravity , 25(11):114010, 14, 2008

2008
[41]

Ringstr¨ om

H. Ringstr¨ om. Strong cosmic censorship in T 3-Gowdy spacetimes. Ann. of Math. (2) , 170(3):1181–1240, 2009

2009
[42]

Ringstr¨ om

H. Ringstr¨ om. On the geometry of silent and anisotropi c big bang singularities. arXiv:2101.04955
[43]

Ringstr¨ om

H. Ringstr¨ om. Wave equations on silent big bang backgr ounds. arXiv:2101.04939

arXiv
[44]

Ringstr¨ om

H. Ringstr¨ om. Initial data on Big Bang singularities. arXiv:2202.04919

arXiv
[45]

Rodnianski and J

I. Rodnianski and J. Speck. A regime of linear stability for the Einstein-scalar ﬁeld system with applications to nonlinear big bang formation. Ann. of Math. (2) , 187(1):65– 156, 2018

2018
[46]

Rodnianski and J

I. Rodnianski and J. Speck. Stable big bang formation in near-FLR W solutions to the Einstein-scalar ﬁeld and Einstein-stiﬀ ﬂuid systems. Selecta Math. (N.S.) , 24(5):4293– 4459, 2018. 59

2018
[47]

Rodnianski and J

I. Rodnianski and J. Speck. On the nature of Hawking’s in completeness for the Einstein-vacuum equations: The regime of moderately spati ally anisotropic initial data. preprint, 2018

2018
[48]

J. Speck. The maximal development of near-FLR W data for the Einstein-scalar ﬁeld system with spatial topology S3. Comm. Math. Phys. , 364(3):879–979, 2018

2018
[49]

St ˚ ahl

F. St ˚ ahl. Fuchsian analysis of S2 × S1 and S3 Gowdy spacetimes. Classical Quantum Gravity, 19(17):4483–4504, 2002. 60

2002

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [18]

Fournodavlos and J

G. Fournodavlos and J. Luk, Asymptotically Kasner-lik e singularities. Amer. J. Math. , 145(4):1183–1272, 2023

work page 2023

[2] [20]

Fournodavlos and J

G. Fournodavlos and J. Smulevici. The initial boundary value problem for the Einstein equations with totally geodesic boundary. Comm. Math. Phys. , 385(3):1615–1653, 2021

work page 2021

[3] [1]

Stable space-like singularity formation for axi-symmetric and polarized near-Schwarzschild black hole interiors

S. Alexakis and G. Fournodavlos. Stable space-like sing ularity formation for axi- symmetric and polarized near-Schwarzschild black hole int eriors. arXiv:2004.00692. 56

work page Pith review arXiv 2004

[4] [2]

E. Ames, F. Beyer, J. Isenberg, and P. G. LeFloch. Quasili near hyperbolic Fuchsian sys- tems and A VTD behavior in T 2-symmetric vacuum spacetimes. Ann. Henri Poincar´ e, 14(6):1445–1523, 2013

work page 2013

[5] [3]

E. Ames, F. Beyer, J. Isenberg, and P. G. LeFloch. A class o f solutions to the Einstein equations with A VTD behavior in generalized wave gauges. J. Geom. Phys. , 121:42–71, 2017

work page 2017

[6] [4]

Andersson and A

L. Andersson and A. D. Rendall. Quiescent cosmological s ingularities. Comm. Math. Phys., 218(3):479–511, 2001

work page 2001

[7] [5]

J. D. Barrow. Quiescent cosmology. Nature, 272:211–215, 1978

work page 1978

[8] [6]

V. A. Belinski and M. Henneaux. The cosmological singula rity. Cambridge University Press, 2018

work page 2018