Reissner–Nordström and Kerr Interiors

On their charged and rotating subextremal branches, the exact Reissner–Nordström and Kerr solutions contain smooth inner horizons in their maximal analytic extensions. The Schwarzschild endpoint does not. Those horizons are valuable reference geometries, but they are not reliable pictures of a generic dynamical interior: radiation that is harmless outside is infinitely blueshifted along the Cauchy horizon. The stationary metrics therefore define the background, the causal vocabulary, and the quantities whose stability must be tested—not the generic endpoint.

Helpful background. Black-Hole Geometry and Mechanics supplies stationary horizons and Kerr geometry; Black-Hole Perturbation Theory supplies exterior decay and mode language; Formulations of General Relativity supplies initial-value variables; Causality and Global Structure supplies domains of dependence and Cauchy horizons; Computational Relativity supplies numerical validation concepts.

Charged and rotating root structure

In geometrized coordinates for this local discussion, set $G=c=1$, take $M\gt0$, and restore $G$ in the invariant length scale at the end. For four-dimensional asymptotically flat Einstein–Maxwell theory with zero cosmological constant, Reissner–Nordström has

$$ds^2=-f(r)\,dt^2+f(r)^{-1}dr^2+r^2d\Omega^2, \qquad f(r)=1-\frac{2M}{r}+\frac{Q^2}{r^2},$$

with roots

$$r_\pm=M\pm\sqrt{M^2-Q^2}.$$

For $0\lt\lvert Q\rvert\lt M$, $r_+$ is the event horizon and $r_-$ is an inner, or Cauchy, horizon. Between them $f\lt0$; on the future black-hole branch, decreasing $r$ is as compulsory as increasing time in an exterior region. Inside $r_-$, $f\gt0$ again and the exact extension contains a timelike curvature singularity at $r=0$. At $Q=0$, the formal lower root is $r_-=0$, the Schwarzschild central curvature singularity rather than a horizon. The repeated extension shown by the eternal solution is an analytic property of an exactly stationary electrovacuum metric, not a prediction that a collapsed charged star opens passages to other exteriors.

Kerr replaces charge with angular momentum. With $a=J/M$,

$$\Delta=r^2-2Mr+a^2=(r-r_+)(r-r_-), \qquad r_\pm=M\pm\sqrt{M^2-a^2}.$$

For $0\lt\lvert a\rvert\lt M$, the outer and inner Killing horizons occur at $r_+$ and $r_-$, respectively. At $a=0$, the formal lower root $r_-=0$ is the Schwarzschild central curvature singularity, not an inner horizon. On the rotating branch $a\ne0$, the Kerr singular set $\Sigma=r^2+a^2\cos^2\theta=0$ is a ring in the analytic chart, and the extension contains regions with closed timelike curves. At $a=0$ the singularity is instead the central locus $r=0$. These statements describe the maximal analytic extension; the physically relevant question is which portion remains after perturbing admissible initial data. The causal structures and their distinction from dynamical collapse are set out in Chandrasekhar 1983, Chapters 5–7, pp. 204–358.

Restoring dimensions, the gravitational length is $GM/c^2$, and the dimensionless parameters are $Q/(\sqrt{G}M)$ in compatible electromagnetic units and $\chi=cJ/(GM^2)$. Nothing physical depends on the Boyer–Lindquist labels themselves.

Why the inner horizon is a Cauchy horizon

Let $\Sigma_0$ be a complete initial hypersurface for the exterior and black-hole region. Its future domain of dependence $D^+(\Sigma_0)$ contains precisely those events whose past-directed causal curves must meet $\Sigma_0$. The future boundary of that domain inside the stationary black hole is the Cauchy horizon. Beyond it, the data on $\Sigma_0$ no longer determine a unique extension.

This loss of global hyperbolicity is logically separate from curvature blowup. On the charged and rotating subextremal branches of exact Reissner–Nordström and Kerr, regular coordinates cross the inner horizon and the metric is analytic there. The same null surface is nevertheless an infinite-blueshift surface. Along late infall, an advanced-time interval at the event horizon is compressed exponentially relative to an observer’s proper time near the inner horizon. That competing combination—metric regularity in the exact solution and unstable radiation transport—is the seed of mass inflation.

Dafermos and Luk formulate the nonsymmetric vacuum problem in double-null variables and prove persistence of a nontrivial continuously extendible Cauchy-horizon segment for a specified class of data close to Kerr; this is not a theorem that the curvature remains bounded or that all of the Kerr extension survives Dafermos and Luk 2025, §§1.2–1.5, pp. 9–33, and §4, pp. 50–65, PDF. In spherical Einstein–Maxwell–scalar models, the relation between exterior decay, the inner horizon, and mass inflation can be proved with much more global control Dafermos 2005, §§7–12, pp. 23–51, PDF.

What survives perturbation

Three statements must be kept distinct.

• The stationary inner horizon is a smooth null hypersurface in the exact metric.
• A perturbed metric may extend continuously across a null boundary even while its connection, Hawking mass, or curvature diverges.
• Inextendibility depends on the demanded regularity: failure of a $C^2$ extension does not imply failure of every $C^0$ extension.

The 2026 exterior result now supplies a narrower rigorous bridge: Hintz reports that its decay estimates meet the Dafermos–Luk hypotheses, giving $C^0$ stability of the Cauchy horizon for the stated small vacuum perturbations of rotating subextremal Kerr Hintz 2026, v2, §1.1, consequence (1), p. 3, PDF. This current preprint does not prove generic curvature blowup, classify the complete singular boundary, or cover arbitrary collapse data.

A dependable mental picture is therefore layered: Kerr or Reissner–Nordström fixes the zeroth-order causal map; exterior tails provide incoming radiation; counterstreaming near the inner horizon amplifies energy; and the resulting boundary must be classified by invariant curvature and extension regularity, not by resemblance to the eternal diagram.

Checks and limitations

The Schwarzschild limits $Q\to0$ in Reissner–Nordström and $a\to0$ in Kerr give $r_-\to0$, where the endpoint is the central curvature singularity, so there is no finite-area inner horizon to perturb. The extremal limit makes $r_+=r_-$ and invalidates estimates that use a nonzero inner-horizon surface gravity. Reissner–Nordström is a spherical charged model, whereas astrophysical black holes are expected to be nearly neutral and rotating; it is used because it isolates the mechanism, not because its matter content is typical.

The exact solutions also do not decide whether a late boundary is null everywhere, becomes spacelike, or develops both types of pieces. Those are dynamical questions addressed on the following pages.

Exercise

For Reissner–Nordström with $Q=4M/5$, find both horizon radii and the signed quantities $f'(r_\pm)/2$ associated with the static Killing-field normalization. Compare their magnitudes. Why must the inner-horizon blueshift rate use a positive magnitude?

Solution

The square root is $3M/5$, so $r_+=8M/5$ and $r_-=2M/5$. Factoring $f=(r-r_+)(r-r_-)/r^2$ gives

$$\frac12 f'(r_\pm) =\frac{r_\pm-r_\mp}{2r_\pm^2}.$$

Thus $f'(r_+)/2=15/(64M)$ and $f'(r_-)/2=-15/(4M)$. The inner magnitude is sixteen times larger. The negative inner sign means that an outgoing radial null ray with $dr/dv=f/2$ approaches $r_-$ exponentially as advanced time increases. Consequently the compression rate used in a blueshift factor is $\kappa_-=\lvert f'(r_-)\rvert/2\gt0$. Both rates have units of inverse length in geometrized units; neither is a statement about curvature divergence in the exact solution.

References

• Chandrasekhar, Subrahmanyan, The Mathematical Theory of Black Holes, Oxford University Press, 1983.
• Dafermos, Mihalis, “The Interior of Charged Black Holes and the Problem of Uniqueness in General Relativity”, Communications on Pure and Applied Mathematics 58, 445–504, 2005. Open PDF.
• Dafermos, Mihalis, and Jonathan Luk, “The Interior of Dynamical Vacuum Black Holes I: The C⁰-Stability of the Kerr Cauchy Horizon”, Annals of Mathematics 202, 309–630, 2025. Open PDF.
• Hintz, Peter, “Nonlinear Stability of Subextremal Kerr Black Holes”, arXiv:2606.28253v2, 2026. Open PDF.