Asymptotic Completeness and Open Problems

Asymptotic completeness is the assertion that a theory’s admissible states at infinity and its global solutions determine one another. For Einstein’s equation this goal is more demanding than global existence, decay, a regular null boundary, or even a forward radiation field. It requires a precisely defined space of asymptotic data, constraint and gauge compatibility, a forward map with no missing energy, and an inverse construction with uniqueness modulo diffeomorphism.

This frontier assessment uses literature versions checked through 2026-08-12. A later revision, independent verification, counterexample, or new theorem can change individual entries.

Required background. Analysis for Mathematical Relativity supplies the weighted spaces and nonlinear estimates, and exterior algebra and differential forms supplies constraint and flux formulations.

Helpful background. The Einstein equation supplies the gauge-reduced system; Causality and Global Structure supplies the asymptotic ends; PDEs and Green Functions for Gravitation supplies scattering theory; and Gravitational Radiation and Waveforms supplies the radiative interpretation.

Four nonequivalent conclusions

Let $\mathcal D$ be a constrained Cauchy-data space modulo spatial diffeomorphisms, and let $\mathcal R^\pm$ be compatible past and future asymptotic-data spaces. A complete nonlinear theory would construct

$$\mathscr F^\pm:\mathcal D\longrightarrow\mathcal R^\pm, \qquad \mathscr S=\mathscr F^+\circ(\mathscr F^-)^{-1},$$

and establish the following separate properties:

1. global existence and completeness: solutions in the data class reach the declared asymptotic boundaries;
2. forward asymptotics: every solution has a radiation state in the named norm;
3. scattering from infinity: every sufficiently small compatible asymptotic state is attained by a unique solution modulo gauge;
4. quantitative completeness: the maps and inverses are continuous, or smoother, without unexplained loss of derivatives.

An energy equality supports item 2 but does not prove item 3. A stability theorem supports item 1 and may give detailed late-time expansions, but it does not automatically parametrize all admissible radiation states.

Scattering results through 2026-08-12

Each row fixes the equation, dimension, asymptotics, data class, regularity, gauge, boundary, norm, and conclusion; moving between rows is a formulation choice, not a corollary.

Regime	Data and boundary	Conclusion or claim through 2026-08-12	Status and boundary of inference
Scalar wave on a fixed subextremal Kerr exterior	Finite degenerate-energy data on past/future horizon and null infinity	Bounded scattering map, existence and uniqueness of scattering states, and asymptotic completeness	Peer-reviewed theorem; linear field on a fixed background, not Einstein backreaction
Vacuum Einstein near Minkowski, spatial dimension $n\ge4$	Small weighted $b$-Sobolev Cauchy and radiation data in harmonic gauge	The nonlinear radiation-field map is a local isomorphism	Theorem/preprint version; excludes physical $n=3$ and its long-range logarithms
Vacuum Einstein near Minkowski in $3+1$	Small asymptotically flat Cauchy data	Global completeness, detailed forward asymptotics, and total-energy radiation statements	Peer-reviewed forward results; a general local bijection with freely prescribed full null data is not supplied by these statements
Vacuum Einstein near a subextremal Kerr black hole	Small polyhomogeneous perturbations with specified gauge-source and constraint conditions	Forward nonlinear stability and settlement to nearby Kerr are claimed for the full subextremal family	Hintz arXiv:2606.28253v2, recent 339-page preprint; not itself a two-sided surjective scattering theorem
Vacuum Einstein near de Sitter, even spatial $n\ge4$	Small data on spacelike past and future conformal boundaries in Sobolev-type norms	Existence, uniqueness, asymptotic completeness, and locally invertible locally Lipschitz scattering map are claimed	Cicortas arXiv:2410.01558v2; positive cosmological constant and spacelike infinity, not asymptotically flat null infinity

The first row is proved in Dafermos, Rodnianski, and Shlapentokh-Rothman 2018, Theorems 1 and 2, pp. 379–391. The higher-dimensional asymptotically flat row is Fang Wang’s isomorphism theorem Wang 2014, Theorems 1.1–1.3 and §7. For physical $3+1$ vacuum data, Lindblad proves wave-like null asymptotics and that radiated energy equals the initial mass in the stated small-data class Lindblad 2017, Theorems 1.1–1.2 and §§6–7. Those are strong forward conclusions, but logical care forbids silently upgrading them to a bijective map for arbitrary compatible data on all of $\mathscr I$.

The two newest rows require especially cautious reading. Hintz’s v2 preprint claims full-subextremal Kerr stability with a finite polyhomogeneous initial expansion, a remainder of order $r^{-3-\epsilon_0}$, finite-dimensional gauge source terms, companion linear estimates, and a Nash–Moser iteration Hintz 2026, Theorems 1.1 and 13.1, pp. 2–3 and 318–320, arXiv v2. Cicortas’s v2 preprint claims a sharp nonlinear de Sitter scattering theory in all even spatial dimensions at least four Cicortas 2024, Theorem 1.1, pp. 3–4, arXiv v2. These are theorem claims in current preprints, not interchangeable evidence for physical asymptotically flat $3+1$ scattering.

Why the physical problem remains difficult

In $3+1$ dimensions the Einstein system has long-range Coulomb terms and only the weak null structure in common wave gauges. The leading asymptotic profile can modify the optical function, and logarithms can enter expansions. Free null data are also constrained: shear, mass aspect, angular momentum aspect, charges, endpoint behavior near spatial infinity, and BMS frame cannot be chosen independently.

A black-hole final state adds two boundaries and a finite-dimensional family of stationary endpoints. A satisfactory scattering map must track flux to $\mathscr I^+$ and $\mathcal H^+$, final mass and angular momentum, gauge modulation, tails, superradiance, and the backward blue-shift at the horizon. Near extremality, uniform constants are another issue. Large data can form trapped surfaces and singularities, so no unrestricted asymptotic-completeness conjecture can be true without selecting an exterior or a cosmic-censorship class.

Consequently the central open question is not merely whether solutions decay. It is whether one can characterize a natural, gauge-invariant range of physical radiation data and prove a locally invertible nonlinear map for $3+1$ asymptotically flat vacuum gravity—near Minkowski or a black-hole family—with all endpoint conditions and asymptotic ends controlled. This remains unresolved at the evidence cutoff.

Diagnostic: identify the missing arrow

Assume a theorem proves that every small Cauchy datum produces a complete solution, that a future radiation field exists, and that its flux equals the initial energy. Which part of asymptotic completeness is still absent?

Solution

The theorem gives a well-defined, norm-controlling forward map $\mathscr F^+$. It does not show that every compatible future radiation state lies in the range, nor that an inverse solution is unique modulo gauge. Surjectivity requires solving from infinity with the asymptotic constraints and endpoint conditions; injectivity requires ruling out a nonradiating remainder or pure-gauge ambiguity. Thus energy exhaustion alone is not a bijective scattering theory.

References

• Cicortas, Serban. “Nonlinear Scattering Theory for Asymptotically de Sitter Vacuum Solutions in All Even Spatial Dimensions.” arXiv:2410.01558v2 (2024). arXiv v2.
• Dafermos, Mihalis, Igor Rodnianski, and Yakov Shlapentokh-Rothman. “A Scattering Theory for the Wave Equation on Kerr Black Hole Exteriors.” Annales Scientifiques de l’École Normale Supérieure 51 (2018): 371–486. DOI.
• Hintz, Peter. “Nonlinear Stability of Subextremal Kerr Black Holes.” arXiv:2606.28253v2 (2026). arXiv v2.
• Lindblad, Hans. “On the Asymptotic Behavior of Solutions to Einstein’s Vacuum Equations in Wave Coordinates.” Communications in Mathematical Physics 353 (2017): 135–184. DOI.
• Wang, Fang. “Radiation Field for Einstein Vacuum Equations with Spacial Dimension $n\ge4$.” arXiv:1304.0407v2 (2014). arXiv.