Weak Cosmic Censorship

Weak cosmic censorship asks whether singular behavior produced from regular, asymptotically flat initial data is generically hidden from distant observers. As of 2026-08-20, the evidence does not support an unqualified yes or no. Symmetry-reduced matter models contain both instability results and a topology-dependent stability claim for naked singularities, while structured characteristic constructions now include vacuum and nonspherical examples. Unrestricted four-dimensional vacuum evolution remains open. The working assessment is therefore open and contested, with high confidence that the formulation itself controls the verdict.

Helpful background. Cosmic Censorship and Inextendibility supplies maximal developments, null infinity, extendibility, and the distinction between weak and strong censorship. This record uses those definitions and concentrates on the current genericity evidence.

Which weak-censorship proposition is being tested?

Fix an Einstein–matter system, complete asymptotically flat initial data, and a topology on the constraint set. A finite version asks whether future null infinity is complete for an open dense, residual, or full-measure class of those data. That statement is stronger than the existence of an event horizon in examples, and different from asking whether one locally naked singular endpoint can be constructed.

Six axes must remain visible: vacuum versus matter; spherical symmetry versus unrestricted perturbations; Cauchy versus characteristic or exterior data; smooth, Hölder, Sobolev, or bounded-variation regularity; the chosen meaning of genericity; and the causal conclusion at null infinity. In particular, stability in a localized $C^{1,\alpha}$ norm and instability in a BV norm need not conflict, because the neighborhoods contain different perturbations.

Assessment at 2026-08-20	Meaning
Question status	Open
Evidence status	Contested
Confidence	High for the formulation-dependent verdict; not high for either universal answer
Domain	Four-dimensional asymptotically flat collapse, with theorem cells separated by matter, symmetry, regularity, and data topology
What could change the conclusion	A proof, generic counterexample, or decisive formulation change in a named model

A starting project: compare perturbation topologies

Fit and preparation. Choose this direction if you enjoy precise theorem statements and can work with characteristic initial data, total variation, and Hölder seminorms. A useful readiness test is to explain why convergence in one norm need not preserve smallness in another. This is a pencil-and-paper project with optional symbolic checks; an adviser familiar with Einstein–scalar evolution is more important than computing power.

Reading and baseline. First extract the exceptional-set conclusion and the allowed perturbations from Christodoulou 1999, Theorem 4.1, pp. 215–216. Then extract the background regularity, localization, and smallness conditions from Zheng 2026, Theorems 1.3 and 3.1. Reconstruct one smooth, localized profile in the notation of each paper and calculate the norms as its amplitude and support scale vary. A norm comparison is preparatory analysis; it is not a new family of constraint-satisfying spacetimes.

Deliverable. Produce a short comparison with the two exact data spaces, their quantifiers, and a worked scaled-profile calculation. Mark every step that would be needed to turn the profile into admissible characteristic data. A successful reproduction explains why the two conclusions do not presently contradict one another; it need not settle an inclusion between their full data spaces.

Possible extension and stopping point. After checking subsequent papers and obtaining expert feedback, investigate one precisely stated inclusion or failure of inclusion for a restricted admissible family. That could become research only if the comparison is absent from the literature. Stop if the proposed contradiction changes matter, asymptotics, or topology, or if the profile violates the characteristic constraints. Narrow the project to the norm calculation before attempting nonspherical stability or complete Cauchy-data formation.

What is established in the stated domain

Naked singularities are not merely heuristic. Christodoulou constructed them in spherical Einstein–massless-scalar collapse in the 1994 construction, pp. 607–653. He later proved the two instability alternatives and the positive-codimension structure of the exceptional set within a rough BV framework in Theorems 2.1 and 3.1, pp. 192–214, and Theorem 4.1, pp. 215–216. This is strong evidence for censorship in that specified topology, not a theorem about every smoother topology or nonspherical vacuum data.

An and Tan claim an analogous censorship result for the spherical Einstein–Maxwell–charged-complex-scalar system. Their exact v1 theorem is formulated for characteristic profiles in $BV([0,\infty),dv)\cap L^1((0,\infty),dv/v)$: data outside an exceptional set have complete future null infinity, and through each exceptional datum passes a complex two-dimensional transverse family containing no other exceptional datum in Theorem 1.1 and Remark 1, printed pp. 2–3, and precise Theorem 9.1, printed p. 136. This remains an unrefereed, symmetry-reduced theorem claim rather than a result for unrestricted matter or vacuum data.

The vacuum construction also extends beyond an exterior region. Rodnianski and Shlapentokh-Rothman construct the exterior in their main theorems, §§1 and 3; Shlapentokh-Rothman’s companion v1 interior paper glues the two regions into the maximal development of a complete asymptotically flat characteristic cone with incomplete future null infinity in Theorem 1.3, printed pp. 5–6. The initial data have arbitrarily high angular regularity but only $C^{1,s}$ regularity in the null direction in Remark 1.2, printed pp. 5–6. This is a full characteristic construction, but not formation from smooth complete Cauchy data and not a stability theorem.

Nonspherical Einstein–scalar results point in both directions. An shows that small anisotropic characteristic perturbations can cover the Christodoulou singularity with an apparent horizon in the main theorem and its stated perturbation norms. An and Wu’s July 2026 v1 preprint instead constructs nonspherical double-null data retaining incomplete future null infinity: it prescribes structured $\kappa$-self-similar incoming data and small outgoing perturbations, and also finds a $C^{1,(\kappa+\delta)/(1-\kappa)}$ obstruction for the scalar field at the singular inner Cauchy horizon in Theorem 1.3, printed pp. 3–4. This removes spherical symmetry for a constructed family, but does not prove open-set stability or genericity in an unrestricted data space.

A separate topology-dependent shift is Zheng’s 2026 preprint: a one-parameter family of continuously self-similar spherical naked singularities is claimed to be nonlinearly stable in a localized Hölder topology in Theorem 1.3, printed pp. 2–3, and precise Theorem 3.1. Its linear input is a separate companion preprint Singh and Zheng 2026, main theorem. The result is important, but it is neither vacuum nor nonspherical, and it does not contradict the BV theorem.

Results and open questions at 2026-08-20

Claim or evidence	What it supports	What it does not establish
Spherical scalar naked-singularity construction	Naked endpoints occur in a self-gravitating model	Stability or genericity
BV instability theorem	Censoring perturbations are available in a rough topology	Instability in every smoother topology
Charged-scalar BV theorem claim	Complete null infinity outside a transverse exceptional set in one spherical characteristic class	Peer-reviewed confirmation, nonspherical data, or unrestricted matter models
Anisotropic apparent-horizon formation	A nonspherical mechanism can cover the scalar singularity	Open density for unrestricted vacuum Cauchy data
Vacuum exterior and companion interior	A vacuum naked solution arises from a complete asymptotically flat characteristic cone	Smooth complete Cauchy formation or stability
Hölder-stability preprint and companion	An open neighborhood is claimed within one localized spherical norm	Independent confirmation, BV stability, or nonspherical stability
Nonspherical scalar double-null construction	Incomplete future null infinity survives in one structured nonspherical family	An open neighborhood, genericity, or vacuum formation

Theorems are weighted by their exact hypotheses, not counted as votes. The apparent contradiction is resolved once the perturbation space is included in the proposition. The remaining dispute is whether one topology captures the geometrically and physically relevant notion of generic collapse.

Competing interpretations of genericity

The censorship-favoring interpretation emphasizes the BV instability theorems in the real- and charged-scalar models and anisotropic horizon formation: known naked solutions can be removed by specified rough or nonspherical perturbations. Its strongest limitation is that none of these results proves an open dense or residual theorem for complete unrestricted vacuum data.

The opposing interpretation emphasizes the 2026 localized Hölder neighborhood: within that topology, naked behavior is not isolated. Zheng’s stability theorem remains spherical, belongs to a restricted self-similar family, and depends on companion analysis from the same programme. An–Wu removes spherical symmetry only for a structured double-null construction, not for an open stability theorem. Calling either result “a counterexample to weak cosmic censorship” without naming the data space would erase the central mathematical content.

Counterexamples, null results, and failed routes

The vacuum exterior-plus-interior construction defeats the idea that vacuum equations alone forbid a full naked solution from complete asymptotically flat characteristic data. Its nonsmooth null regularity and absence of a stability result prevent transfer to smooth complete Cauchy data. The Hölder-stability claim defeats a topology-free assertion that every known naked scalar singularity is unstable, but it does not defeat Christodoulou’s BV statement. An–Wu removes symmetry for a constructed Einstein–scalar family, while An’s apparent horizon shows that this existence result is still insufficient for a four-dimensional genericity claim.

No verified result at the cutoff supplies a stable vacuum naked singularity arising from complete regular Cauchy data. No verified result proves generic weak censorship for unrestricted four-dimensional vacuum data either. These absences are gaps in theorem coverage, not numerical null searches with quantified sensitivity.

What would change the assessment

A decisive advance would put censoring and naked outcomes in the same constraint-satisfying data space. An–Wu supplies nonspherical existence for structured double-null data, but the stability route still requires an open neighborhood in a geometrically natural nonspherical norm. The opposite route is a proof that horizon-forming perturbations are dense in that same norm. For vacuum gravity, either an open stable formation theorem from complete Cauchy data or an open-dense completeness theorem for future null infinity would materially change the record.

Research checklist

• State the Einstein–matter model, asymptotics, data surface, regularity, and topology before using “generic.”
• Distinguish existence of a naked endpoint, formation from regular data, nonlinear stability, and visibility from null infinity.
• Reconstruct the BV and localized Hölder perturbation classes on a common family; identify which inclusions fail.
• Check that nonspherical perturbations satisfy the constraints and the smallness norm actually used by the theorem.
• Trace Zheng’s nonlinear theorem through the Singh–Zheng linear input instead of treating the companion as independent replication.
• Keep the An–Wu constructed double-null family separate from an open stability statement, and keep the Shlapentokh-Rothman complete characteristic cone separate from smooth Cauchy data.
• Treat An–Tan’s charged-scalar result as a v1 theorem claim in its exact $BV\cap L^1(dv/v)$ characteristic class.
• Report whether completeness, apparent-horizon formation, or local causal visibility is the measured conclusion.

Source and data map

The original positive and adverse results are the Christodoulou construction and instability theorems, An’s anisotropic theorem, the Rodnianski–Shlapentokh-Rothman vacuum exterior, and Shlapentokh-Rothman’s companion interior. An–Tan supplies the charged-scalar BV theorem claim; Zheng and Singh–Zheng supply the localized Hölder claim and its explicit dependency; An–Wu supplies the structured nonspherical double-null construction. The search covered asymptotically flat formation and genericity evidence. Inner Cauchy horizons were included only where the same constructed spacetime also has incomplete future null infinity; test-body destruction of an existing hole remained excluded because it tests a different proposition.

No independent reproduction was performed, and no machine-checkable proof archive was found. Reproduction here means an expert comparison of theorem hypotheses, especially the topology definitions in the 2026 preprints, the anisotropic perturbation norms, and the regularity of characteristic data. An–Tan v1, Shlapentokh-Rothman v1, An–Wu v1, and the two May 2026 companion papers had no journal reference at the cutoff.

Return to the seminar

Record this page’s worksheet, return to Case 1 in the Research Seminar, then continue to Case 2 — Nonlinear Kerr stability. Read its seminar instructions before treating it as a reproduction candidate.

References

• An, Xinliang. “Naked Singularity Censoring with Anisotropic Apparent Horizon.” Annals of Mathematics 201 (2025), 775–908. Journal record.
• An, Xinliang, and Hong Kiat Tan. “A Proof of Weak Cosmic Censorship Conjecture for the Spherically Symmetric Einstein–Maxwell–Charged Scalar Field System.” arXiv:2402.16250v1 (2024). arXiv v1.
• An, Xinliang, and Shengrong Wu. “Naked Singularities beyond Spherical Symmetry: Singular Inner Cauchy Horizons for the Einstein–Scalar Field System.” arXiv:2607.07134v1 (2026). arXiv v1.
• Christodoulou, Demetrios. “Examples of Naked Singularity Formation in the Gravitational Collapse of a Scalar Field.” Annals of Mathematics 140 (1994), 607–653. DOI.
• Christodoulou, Demetrios. “The Instability of Naked Singularities in the Gravitational Collapse of a Scalar Field.” Annals of Mathematics 149 (1999), 183–217. DOI.
• Rodnianski, Igor, and Yakov Shlapentokh-Rothman. “Naked Singularities for the Einstein Vacuum Equations: The Exterior Solution.” Annals of Mathematics 198 (2023), 231–391. DOI.
• Shlapentokh-Rothman, Yakov. “Naked Singularities for the Einstein Vacuum Equations: The Interior Solution.” arXiv:2204.09891v1 (2022). arXiv v1.
• Singh, Jaydeep, and Weihao Zheng. “Nonlinear Stability of Continuously Self-Similar Naked Singularities for the Einstein-Scalar Field Equations II: Linearized Stability.” arXiv:2605.16095v1 (2026). arXiv.
• Zheng, Weihao. “Nonlinear Stability of Continuously Self-Similar Naked Singularities for the Einstein-Scalar Field Equations I: Main Results.” arXiv:2605.16235v1 (2026). arXiv.