Near-AdS2 Black Holes Beyond the Universal Sector

JT gravity and the Schwarzian describe a universal leading sector only when every omitted throat mode remains parametrically heavier than the temperature and the excitation amplitude is small enough to suppress nonlinear mixing. Real black holes also contain irrelevant deformations that join the throat to the exterior, Kaluza–Klein towers, gauge and matter fields, and higher-derivative interactions. This page turns “beyond JT” into a hierarchy of calculable corrections rather than a vague warning.

Required background. Gauge Theory and String-Theory Preparation provides effective-field-theory and operator-dimension conventions.

Helpful background. Quantum Gravity supplies truncation and ultraviolet matching logic; Quantum Information supplies spectral diagnostics; Semiclassical Gravity provides near-horizon thermodynamics; and Black-Hole Information motivates late-time sensitivity to small corrections.

The hierarchy behind the universal sector

Let $E_{\rm th}\sim T$ be the throat excitation scale, $E_{\rm gap}$ the lightest omitted Kaluza–Klein or matter scale, and $E_{\rm match}$ the scale at which the AdS2 region joins the exterior. A controlled one-field description requires

$$T\ll E_{\rm gap},\qquad T\ll E_{\rm match}, \qquad \frac{\delta\phi}{\phi_{\rm bg}}\ll1 .$$

The last condition constrains amplitude rather than frequency. A low-frequency perturbation can still cause large dilaton backreaction if inserted for long enough or with large enough energy. Conversely, an excitation with small amplitude but frequency near $E_{\rm gap}$ resolves an omitted mode and invalidates the truncation.

An irrelevant boundary operator of dimension $\Delta\gt1$ with coupling $g_\Delta$ produces corrections organized by a dimensionless quantity such as

$$\varepsilon_\Delta(T)=g_\Delta T^{\Delta-1}.$$

Its coefficient and even the first allowed $\Delta$ depend on the ultraviolet black hole, compactification, charges, and symmetries. Universality predicts the leading Schwarzian form; matching determines the irrelevant data.

Results and limitations through 2026-08-20

The formulation axes are the throat gap, operator dimensions, compactification spectrum, charge sector, matching radius, perturbation amplitude, and order of limits.

Evidence or position	Controlled result	Limitation or adverse discriminator
Maldacena, Stanford, and Yang 2016, §§3–5, pp. 6–18	The Schwarzian governs the leading boundary response and correlator dressing below the throat gap	The derivation assumes a single relevant soft sector and fixed ultraviolet boundary conditions
Maldacena, Stanford, and Yang 2016, §7, pp. 23–24	The paper identifies extra light fields and short throats as explicit limits on universality	It does not classify the complete Kaluza–Klein spectrum of a generic compactification
Almheiri and Polchinski 2015, §§3.2–3.3 and 5, pp. 8–9 and 14–19	Matter backreaction can be solved in concrete two-dimensional dilaton models	Boundary conditions and potentials are model-dependent, and sufficiently strong stress destroys the rigid-throat approximation
Ultraviolet matching program	Irrelevant couplings can be extracted by reducing a specified higher-dimensional solution and matching in an overlap region	Without an overlap or a complete mode basis, “small correction” is uncontrolled and mode mixing remains unresolved

The controlled evidence supports a universal leading thermodynamic and soft-mode sector under an explicit hierarchy. Competing positions disagree about how broad that universality is: some regard most deviations as perturbative irrelevant corrections, whereas others emphasize additional nearly gapless fields, fragmentation, or nonlinear throat dynamics. Extrapolation to all observables is model-dependent and remains conjectural.

A representative correction and its check

Suppose a heavy scalar $\psi$ of mass $m$ couples to a light throat operator $\mathcal O$ through $\lambda\psi\mathcal O$. At tree level, solving its equation gives schematically

$$\psi=\frac{\lambda}{m^2-\nabla^2}{\cal O} =\frac{\lambda}{m^2}{\cal O} +\frac{\lambda}{m^4}\nabla^2{\cal O}+\cdots .$$

Substitution produces local operators only when the covariant derivative expansion is controlled: the relevant proper frequencies, gradients, and curvature scale must be small compared with $m$, with the same boundary conditions used in the inverse operator. In AdS2, $T\ll m$ in boundary-time units is insufficient. Redshift permits low boundary frequencies even for fields of nonzero bulk mass; a local large-mass expansion also requires $mL_2\gg1$. Compare the expanded response with the full Green function while varying these ratios. Failure near a pole or an unsuppressed boundary contribution means that the field must be retained or its nonlocal response matched explicitly. A consistent truncation of background solutions is a separate way to obtain pure JT thermodynamics; it does not prove that every correlator has a gap to all omitted fields.

Another correction comes from finite throat length. A solution purely ingoing at a near-horizon boundary must be matched to the exterior potential, generating frequency-dependent response coefficients. These terms can be small in thermodynamics yet decisive for a narrow resonance or a late-time tail. A page that quotes only the local AdS2 Green function has not completed this matching.

Breakdown by light fields and nonlinear response

If an additional field has dimension near one, $\varepsilon_\Delta$ decreases too slowly with temperature to be safely ignored. Gauge zero modes, moduli, and nearly massless compactification fields are common candidates. Their boundary conditions and ensemble matter: fixing a charge can remove one fluctuation that is present in a grand-canonical calculation.

Nonlinear backreaction supplies a distinct failure case. The Schwarzian approximation expands around one monotonic boundary trajectory. Large stress can make the dilaton vanish, produce competing saddles, or invalidate the chosen matching surface. This is not cured by lowering the frequency alone. The open question is which higher-dimensional black holes admit a parametrically broad window in both energy and amplitude, and which observables remain dominated by that window at late times.

What could change these conclusions

The conclusions could change if a theorem classifies all light modes for a major near-extremal compactification, if an independent dimensional reduction or numerical solution reproduces or contradicts the predicted correction hierarchy, or if a counterexample exhibits fragmentation or nonlinear breakdown inside the claimed parameter regime. The conclusions also depend on whether new precision data, a changed boundary condition, or an ultraviolet matching calculation moves the lightest operator dimension or the throat gap.

Exercises

Determine when integrating out a throat field of mass $m$ is justified, and explain how you would detect the failure using one response function.

Solution

The derivative expansion requires every relevant frequency and momentum scale to be small compared with $m$, and the induced field amplitude must remain perturbative. Expanding the exact propagator gives corrections in powers of the scale squared divided by $m^2$. To test it, compute the same retarded response once with the full propagator and once with the first local operators. Their difference should decrease predictably as $T/m$ is lowered. A nearby pole, nonuniform error, or growing residual shows that the field cannot be integrated out even if the leading thermodynamic correction looks small.

References

• Almheiri, Ahmed, and Joseph Polchinski. “Models of AdS2 Backreaction and Holography.” Journal of High Energy Physics 11 (2015) 014. arXiv:1402.6334.
• Maldacena, Juan, Douglas Stanford, and Zhenbin Yang. “Conformal Symmetry and Its Breaking in Two Dimensional Nearly Anti-de-Sitter Space.” Progress of Theoretical and Experimental Physics 2016 (2016) 12C104. arXiv:1606.01857.