Black-Hole Perturbation Validity

A black-hole perturbation is controlled only when its amplitude, perturbative order, gauge, boundary data, and observation time remain compatible. No single component of $h_{ab}$ can certify validity, and a successful linear calculation does not by itself establish nonlinear stability.

Helpful background. Differential equations for relativity supplies the initial-value, boundary-value, and asymptotic language used below.

A validity vector, not one amplitude

Begin with a declared one-parameter family and background,

$$g_{ab}(\lambda) =g^{(0)}_{ab}+\lambda h^{(1)}_{ab} +\lambda^2h^{(2)}_{ab}+O(\lambda^3).$$

The parameter fixes what “first” and “second” order mean. Comparing fields on different members of the family also requires an identification map; changing that map is a gauge transformation. Consequently, a small coordinate component is not a gauge-invariant remainder estimate. The background, perturbative parameter, identification map, and observable must travel together, as formalized in Bruni et al. 1997, §§ 2–3 and 4.1–4.2, Props. 2–3 and Eqs. (4.7)–(4.10).

A practical validity statement supplies the following vector of checks.

Check	What must be declared	Typical failure signal
Invariant size and order	a dimensionless norm or observable, retained order $N$, region, and expected $O(\lambda^{N+1})$ remainder	the next-order correction is not smaller in the stated norm
Background and gauge	mass and spin assigned to $g^{(0)}_{ab}$, identification map, residual gauge class, and gauge-invariant or gauge-controlled output	a “small” field is removed or enlarged by an admissible gauge change, or a parameter shift is counted as radiation
Constraints and boundaries	constraint residuals, compatible initial data, future-horizon regularity, and the chosen condition at infinity	constraint growth, incoming contamination, or a singular horizon representation
Flux and balance	energy and angular-momentum fluxes with the same normalization and perturbative order	balance residuals exceed the claimed remainder or accumulated losses move the background appreciably
Time and frequency scales	observation time, damping or orbital scales, detuning from resonance, and any multiscale variables	secular terms or small denominators make $\lambda t/T$ order unity
Singular geometric limits	distance from extremality and any throat, surface-gravity, or mode-separation scale	the expansion is nonuniform as a second scale tends to zero
Nonlinear coupling	quadratic sources, mode coupling, and backreaction retained at the requested accuracy	$h^{(1)}h^{(1)}$ effects compete with the retained linear response

For a small body, validity is additionally regional. Inner and outer expansions must overlap in a buffer region, use common variables and gauge, and agree order-by-order; local smallness outside that region proves nothing about the global solution. This matched-expansion structure is given in Pound and Wardell 2024, § 5.1, printed pp. 40–42.

Forced response and stability are different questions

A free perturbation solves the homogeneous linearized equations from initial data. Its growth or decay tests linear stability for a specified norm, gauge quotient, and boundary condition. A forced perturbation has a source or prescribed driving; even when every free mode is bounded, resonant or nearly resonant forcing can produce a large response or secular accumulation. Source conservation, scaling, and switch-on behavior are therefore part of the approximation.

Linear stability means that solutions of the linearized problem remain controlled, often after removing pure gauge pieces and infinitesimal changes of background parameters. Nonlinear stability is stronger: solutions of the full equations that start close to the background must remain close in a specified sense while nonlinear mode coupling and backreaction are controlled. Neither statement follows merely from a mode-frequency calculation, and this card makes no mutable verdict about a particular near-extremal or nonlinear problem.

Stop or reorganize the expansion

Re-expand around an evolving background when accumulated flux changes its mass or spin at the retained order. Introduce multiple scales when a nominally small term grows secularly, and use a resonance-specific expansion when detuning becomes as small as the forcing correction. Near extremality, test uniformity in both the perturbation amplitude and the vanishing geometric scale. If constraints, boundary regularity, or the next-order term fail their target size, the perturbation series has left its declared domain even if selected waveform components still look smooth.

The derivations, gauge constructions, and formulation-specific diagnostics belong to Perturbative General Relativity and Gauge Invariance.

Common pitfalls

Small coordinate components imply validity. Component size depends on gauge and coordinates. Use a declared invariant or gauge-controlled norm and verify the next-order remainder.

Mode stability implies nonlinear stability. A bounded homogeneous linear mode does not control forced resonance, secular drift, quadratic coupling, or background backreaction. State precisely which stability problem was solved.

References

• Bruni, Marco, Sabino Matarrese, Silvia Mollerach, and Sebastiano Sonego. “Perturbations of Spacetime: Gauge Transformations and Gauge Invariance at Second Order and Beyond.” Classical and Quantum Gravity 14 (1997): 2585–2606. DOI
• Pound, Adam, and Barry Wardell. “Black Hole Perturbation Theory and Gravitational Self-Force.” In Handbook of Gravitational Wave Astronomy. Springer, 2024. DOI