← all systems · see it in The System
Cosmic censorship hypothesis
'The weak and the strong cosmic censorship hypotheses are two mathematical conjectures about the structure of gravitational singularities in the context of general relativity.' The weak hypothesis, conceived by Roger Penrose in 1969, 'posits that no naked singularity is visible from infinity'; the strong hypothesis 'asserts that generically, general relativity must be a deterministic theory and, thus, our universe must be globally hyperbolic'.
Properties published scienceour proposal
Sourced fields, and our own readings.
| boundary | Inside: the claim about generic, nonsingular, asymptotically flat initial data with suitable matter: the maximal Cauchy evolution has complete future null infinity (weak) and cannot be extended as a regular Lorentzian manifold (strong). Outside: non-generic data, unsuitable matter (fluids with shell crossing), spacetimes not asymptotically flat (anti-de Sitter, de Sitter), more than four dimensions, and the quantum-gravity regime. |
| nested level | A conjecture about the global structure of solutions of Einstein's equations: built on event horizons, Cauchy developments and null infinity; it forbids naked singularities and, in the strong form, demands global hyperbolicity. |
| representation | True |
| scale | cosmological |
Measures (7) published science
| first formulated | 1969 (Penrose, Nuovo Cimento) year |
| Kerr condition for a horizon | a < 1 (a = J/M) G = c = 1 |
| extremal Kerr-Newman bound | M^2 = Q^2 + (J/M)^2 G = c = 1 |
| Penrose inequality for null-dust shells (proven) | A(S) <= 16 pi M^2 G = c = 1 |
| first counterexample in four dimensions (AdS4) | 2017 year |
| very strong (C0) form disproven | 2018 (Dafermos and Luk, Kerr Cauchy horizon) year |
| characteristic clock (human test of the idea, sys:science): the Thorne-Preskill-Hawking bet | made 1991, conceded 1997 (6 years); the revised bet still open |
Open contentions (1) open contention
Where sources or readings disagree. Both stay on the map.
- C-052 Is the inner horizon of a rotating or charged black hole stable?
booked: The map shows where that inner horizon sits and leaves open whether it survives; on the censorship page it follows the view that the horizon is unstable and a revised censorship holds.
the other reading: B: stable Cauchy horizons (AWEC violation; de Sitter); strong censorship 'false'
Partners and parts published science
Built from / builds (4)
- Four-dimensional Lorentzian manifold (events and world lines)
- Event horizon
- Globally hyperbolic spacetime
- Cauchy horizon
Sits inside / holds (2)
Trades with (valence) (13)
- Cauchy horizon via sys:spacetime · Cosmic censorship hypothesis even
- Superextremal Reissner-Nordstrom spacetime via sys:spacetime · Cosmic censorship hypothesis even
- Penrose-Hawking singularity theorems via sys:spacetime · Cosmic censorship hypothesis gain
- Strong cosmic censorship hypothesis via sys:spacetime · Cosmic censorship hypothesis gain
- Naked singularity via sys:spacetime · Cosmic censorship hypothesis even
- Black hole via sys:gravitational-field · Cosmic censorship hypothesis even
- Anti-de Sitter spacetime via sys:spacetime · Cosmic censorship hypothesis loss
- Event horizon via sys:spacetime · Cosmic censorship hypothesis gain
- Globally hyperbolic spacetime via sys:spacetime · Cosmic censorship hypothesis even
- Five-dimensional spacetime (Kaluza-Klein) via sys:spacetime · Cosmic censorship hypothesis loss
- Extremal Reissner-Nordstrom black hole via sys:electromagnetic-field · Cosmic censorship hypothesis even
- Chronology protection conjecture via sys:spacetime · Cosmic censorship hypothesis even
- Maximal globally hyperbolic development via sys:gravitational-field · Cosmic censorship hypothesis gain
expressed by (2)
floor filled by (28)
- Reissner-Nordstrom spacetime
- Cauchy horizon
- Averaged dominant energy condition (ADEC)
- Killing horizon
- Ring singularity
- Overextreme Kerr spacetime
- Kerr-Vaidya null dust
- Superextremal Reissner-Nordstrom spacetime
- Gravitational singularity
- Penrose singularity theorem
- Penrose-Hawking singularity theorems
- Strong cosmic censorship hypothesis
- Naked singularity
- Electromagnetic field
- Black hole
- de Sitter spacetime
- Anti-de Sitter spacetime
- Event horizon
- Globally hyperbolic spacetime
- Scalar field
- Trapped surface
- Kerr-Newman spacetime
- Weak cosmic censorship hypothesis
- Near-extremal Reissner-Nordstrom black hole
- Naked singularity of scalar-field collapse (Christodoulou)
- Black-string pinch-off singularity
- Roberts scalar-field spacetime
- Maximal globally hyperbolic development
represents (1)
Kinds (2)
Notes published science
- Wald's framing: could a mad scientist with finite resources destroy the universe? Weak cosmic censorship says whatever singularity he makes stays behind a horizon, so distant observers 'can live out their lives in their entirety'.
- Hawking lost the 1991 naked-singularity bet in 1997 and re-bet; the prize was 'clothing to cover the winner's nakedness'.
- Throwing charge into a nearly extremal hole looks like it should break the horizon (Hubeny); twenty years later the second-order self-force bookkeeping showed nature always refuses (Sorce-Wald).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
cosmic censorship (own, global structure of generic solutions)
- L1: Initial data: nonsingular, asymptotically flat (h_ab, K_ab, psi) on a 3-manifold that is the connected sum of R^3 and a compact manifold, for 'suitable matter' (well-posed, energy conditions; the Maxwell and Klein-Gordon fields; fluids excluded for shell crossing)
- L2: Evolution to singularities: the maximal Cauchy evolution; singularities are 'inevitable in physically reasonable situations' (Penrose-Hawking); trapped surfaces form when enough mass is in a small region
- L3: Weak form: generically the evolved spacetime is asymptotically flat at future null infinity with complete I+; singularities sit in the black hole region B = M - I^-(I+) behind H; distant observers 'can live out their lives in their entirety'
- L4: Strong form: the maximal Cauchy development of generic compact or asymptotically flat data is not locally extensible as a regular Lorentzian manifold: GR is deterministic, the universe globally hyperbolic. Very strong (C0) form disproven in 2018 (Dafermos-Luk, Kerr Cauchy horizon); weak and strong are mathematically independent
- L5: 'Generically' and the evidence for it: linear stability of Schwarzschild (Vishveshwara 1970, Price) and Kerr (Whiting: no unstable modes); failed Penrose-Gibbons counterexamples (A(S) <= 16 pi M^2 proven); Christodoulou's theorem (spherical Einstein-Klein-Gordon: naked singularities non-generic); no over-charging or over-spinning of Kerr-Newman holes under the null energy condition (Sorce-Wald)
- L6: Limits: the Roberts scalar-field counterexample (1985); five-dimensional black strings (generic violation); Einstein-Maxwell in AdS4 (first four-dimensional counterexample); overcharging of asymptotically de Sitter 'black holes'; Hawking conceded the 1991 bet in 1997 over special-case 'technicalities'
tests by throwing matter in (electromagnetic field)
- L1: Charged or spinning test bodies aimed at an extremal hole (M^2 = Q^2 + (J/M)^2: 'no stationary black hole solutions exist when the right side ... exceeds the left side')
- L2: Refusal at the door: 'not possible to get a black hole to swallow too much charge or angular momentum'; for Kerr a particle with l = 2Me meets a maximum potential of exactly (e^2 - 1)/2: 'too much angular momentum to fall in'
- L3: Near-extremal at test order: Hubeny finds 'one can overcharge a near extremal Reissner-Nordstrom black hole by throwing in a charged particle, as long as the backreaction effects may be considered negligible' (a charged imploding shell upholds censorship)
- L4: Second order: the full correction delta^2 M including self-force and finite-size effects shows 'no gedanken experiments of the generalized Hubeny type can ever succeed', given the null energy condition
- L5: Asymptotically de Sitter extremal 'black holes': 'over-charging can occur in this case'; not against the asymptotically flat weak form, but evidence against some strong formulations
Who fills each layer
- L1: Strong cosmic censorship hypothesis
- L4: Strong cosmic censorship hypothesis, Maximal globally hyperbolic development
Sources (11) published science
Wald, R. M. (1997) Gravitational collapse and cosmic censorship, arXiv:gr-qc/9710068 (behind its links)
Crisford, T., Santos, J. E. (2017) Violating weak cosmic censorship in AdS4, arXiv:1702.05490 (behind its links)
Sorce, J., Wald, R. M. (2017) Gedanken experiments to destroy a black hole II: Kerr-Newman black holes cannot be over-charged or over-spun, arXiv:1707.05862 (behind its links)
Cauchy horizon (Wikipedia, rev 1369451305, 2026-08-15) (behind its links)
Joshi, P. S., Malafarina, D. (2011) Recent developments in gravitational collapse and spacetime singularities, arXiv:1201.3660 (secs. 1-3, 7) (behind its links)
Lehner, L., Pretorius, F. (2010) Black strings, low viscosity fluids, and violation of cosmic censorship, arXiv:1006.5960 (behind its links)