Relaxation and the Final-State Conjecture

The final-state conjecture says, roughly, that generic isolated gravitational collapse or merger with suitable asymptotically flat data has an exterior that eventually approaches a member of the Kerr family, with excess energy radiated away. Numerical ringdown supports this picture in broad binary regimes, but a waveform that is well fitted by damped modes is not the global nonlinear theorem. The unresolved task is to connect large-data horizon formation, exterior completeness, decay, and parameter selection in one argument.

Helpful background. Stationary uniqueness and horizon regularity build on Black-Hole Geometry and Mechanics; linear decay belongs to Black-Hole Perturbation Theory; hyperbolic formulations come from Formulations of General Relativity; nonlinear evidence uses Computational Relativity; and radiation escape is described in Gravitational Radiation and Waveforms.

What convergence to Kerr would mean

Fix an asymptotically flat vacuum initial-data class and its maximal development. A strong exterior statement must identify a complete future null infinity, an event horizon with controlled geometry, final parameters $(M_f,J_f)$, a gauge in which the exterior metric approaches Kerr, and decay rates strong enough to control nonlinear error terms. It must also distinguish convergence on compact exterior regions from behavior near timelike infinity, null infinity, and the horizon.

Linear mode stability excludes exponentially growing separated modes, while boundedness and decay estimates control general perturbations. For Schwarzschild, the Regge–Wheeler and Zerilli sectors can be treated by robust energy estimates; Dafermos, Holzegel, and Rodnianski prove boundedness and decay for the full linearized system modulo linearized Kerr and gauge solutions Dafermos, Holzegel, and Rodnianski 2019, Theorems 1–2 and §§ 9–13. The nonlinear problem must additionally track changing Kerr parameters and prevent slowly decaying interactions from accumulating.

Final-state results through 2026-08-12

• Numerical evidence: vacuum binary evolutions across wide mass-ratio and spin domains form a common horizon and settle toward Kerr-consistent charges, multipoles, and ringdown spectra.
• Perturbative analytic evidence: linear stability and decay are established in important Schwarzschild and Kerr regimes, with hypotheses on regularity, gauge, and asymptotics.
• Asymptotically flat nonlinear stability: Klainerman and Szeftel prove that sufficiently small general asymptotically flat perturbations of slowly rotating Kerr have complete future null infinity and converge in its causal past to a nearby Kerr spacetime Klainerman and Szeftel 2023, § 1.2.1, Theorem 1.2.1, pp. 22–24. Hintz’s versioned 2026 preprint proves nonlinear stability and relaxation to nearby Kerr for small smooth vacuum perturbations throughout the strictly subextremal parameter range; its full statement permits structured, partially polyhomogeneous decay at spatial infinity Hintz 2026, arXiv:2606.28253v2, Theorem 1.1, pp. 2–3, and Theorem 13.1.
• Nonlinear theorems with different asymptotics: Kerr–de Sitter stability exploits cosmological redshift and exponential decay; Hintz and Vasy state the small-perturbation theorem and asymptotic parameter selection Hintz and Vasy 2018, Theorem 1.1 and § 1.1. Positive cosmological constant is essential to that formulation.
• Unresolved claim: a single theorem does not yet begin with generic large asymptotically flat merger data and prove the complete Kerr exterior final state throughout the full subextremal parameter range.

The numerical and analytic positions are complementary rather than competing measurements of one quantity. Simulation evidence covers strongly nonlinear comparable-mass formation but only finite time and resolution. The asymptotically flat stability theorems control infinite time but begin with data close to Kerr, and Hintz’s full-spin result is, at this cutoff, a version-specific preprint with stated spatial-decay structure. The formulation axis is therefore initial-data size, asymptotics, spin range, regularity and decay class, matter model, and the norm used for convergence.

Residual fields and counterexamples to overstatement

Power-law tails mean “settles” need not mean exponential disappearance in asymptotically flat spacetime. Massive or charged fields can introduce long-lived bound states, and matter remaining outside the horizon changes the candidate final family. Extremal horizons possess distinct decay and instability phenomena. A stable-looking finite waveform is thus a counterexample to the inference that every spacetime norm is already small, even when the ringdown fit is excellent.

This comparison covers results through 2026-08-12. Independent verification or publication of the full-subextremal preprint would clarify its mathematical standing. Theorems weakening its structured spatial-decay assumptions, treating extremality or relevant matter, or connecting generic large-data trapped-surface formation to a complete final exterior would extend its scope. Conversely, reproducible numerical evidence of a persistent non-Kerr exterior under controlled convergence could challenge the broader final-state conjecture. A short-duration waveform fit alone cannot settle these questions.

Check your understanding

A simulation shows a stable two-mode Kerr fit for a duration $100GM/c^3$ after merger, with $M$ the remnant mass scale. Which additional facts would be needed to turn this into a final-state theorem?

Solution

One would need a precisely stated initial-data class, global existence of the relevant exterior, completeness and controlled geometry of the horizon and null infinity, estimates that bound nonlinear errors for arbitrarily late time, convergence in declared spacetime norms, and proof that the limiting parameters equal the conserved-charge balance. A finite interval and one observable cannot supply those global analytic statements.

References

• Dafermos, Mihalis, Gustav Holzegel, and Igor Rodnianski. “The Linear Stability of the Schwarzschild Solution to Gravitational Perturbations.” Acta Mathematica 222 (2019): 1–214. doi:10.4310/ACTA.2019.v222.n1.a1.
• Hintz, Peter, and András Vasy. “The Global Non-Linear Stability of the Kerr–de Sitter Family of Black Holes.” Acta Mathematica 220 (2018): 1–206. doi:10.4310/ACTA.2018.v220.n1.a1.
• Hintz, Peter. “Nonlinear Stability of Subextremal Kerr Black Holes.” arXiv:2606.28253v2 [gr-qc] (2026), 339 pp. Open PDF.
• Klainerman, Sergiu, and Jérémie Szeftel. “Kerr Stability for Small Angular Momentum.” Pure and Applied Mathematics Quarterly 19, no. 3 (2023): 791–1678. doi:10.4310/PAMQ.2023.v19.n3.a1.