Hawking Radiation Is Not Explained by a Literal Pair Story

The familiar picture of two particles appearing beside a horizon—one escaping and one falling in—is neither Hawking’s derivation nor a literal spacetime history. The controlled result uses field modes, a state, and a detector or asymptotic-flux prescription on a declared background. The pair story is only a mnemonic for correlations or energy balance; it cannot locate particle creation, predict a freely falling detector, or explain information recovery.

Helpful background. QFT in Curved Spacetime supplies mode, state, and Bogoliubov-transformation language; Semiclassical Gravity supplies renormalized stress tensors and the backreaction boundary; and Quantum Information supplies the entanglement, purity, and channel language needed to formulate the information problem.

Collapse mode mixing gives the thermal factor

For the worked calculation here, take a neutral, nonrotating Schwarzschild black hole formed by collapse. Trace a late outgoing null ray, labeled by retarded time $u$ at future null infinity, backward toward the body. Near the last escaping ray, a regular past null coordinate $U$ has the asymptotic relation

$$U=U_H-Ae^{-\kappa u}+\cdots, \qquad A\gt0.$$

Here $\kappa$ is the surface gravity normalized by the asymptotic time translation. Positive-frequency modes in $U$ are not purely positive frequency in $u$. Their Bogoliubov coefficients obey the thermal ratio

$$\frac{\lvert\beta_{\omega\omega'}\rvert^2} {\lvert\alpha_{\omega\omega'}\rvert^2} =e^{-2\pi\omega/\kappa}, \qquad T_H=\frac{\kappa}{2\pi}$$

in the site’s units. With inverse-length $\kappa$, restoring constants gives $k_{\mathrm B}T_H=\hbar c\kappa/(2\pi)$.

For a stationary rotating or electrically charged hole, take the field mode $e^{-i\omega t+im\phi}$ and $D_a=\nabla_a-i e_{\mathrm f}A_a$. Here $m$ is the azimuthal integer; $e_{\mathrm f}$ is the emitted field’s gauge-charge eigenvalue, not the background geometric charge $q$. With the normalized generator $\chi^a=t^a+\Omega_H\phi^a$, use the rationalized Maxwell convention and horizon-to-infinity potential in Horizon Area, Surface Gravity, Angular Velocity, and Potentials, with gauge $A_a\to0$ at the selected infinity. The horizon-generator energy is

$$\widetilde\omega =\omega-m\Omega_H-e_{\mathrm f}\Phi_H.$$

Neither $e_{\mathrm f}$ nor $\Phi_H$ transfers alone between Maxwell normalizations; their product is the convention-consistent chemical-potential term. Hawking distinguishes field charge from hole charge in Hawking 1975, §3, printed p. 213, Eqs. (3.1)–(3.3), PDF. Superradiant regimes consequently need separate care. The result still depends on the state, normalization, asymptotic time, and late-time approximation, and selects no local pair event. See Hawking 1975, §§2–3, printed pp. 202–214, PDF. The full derivation is developed in Hawking Radiation from Gravitational Collapse.

Why the pair cartoon is not a local mechanism

In curved-spacetime quantum field theory, “particle” means a mode or detector notion tied to a state and an observer. There is generally no coordinate- and observer-independent local particle-number density. Near a large, regular future horizon, a freely falling local frame is approximately inertial; the horizon is not a material surface that must emit pairs into that frame.

This immediately exposes three failures of a literal reading:

1. no invariant particle worldlines are born at one point;
2. moving the imagined creation point across the horizon leaves the asymptotic calculation unchanged; and
3. the distant outgoing quantum is a global mode tied to a measurement protocol.

Fredenhagen and Haag derive the late detector response from the field’s short-distance structure and horizon scaling limit, without assigning a local pair-production event Fredenhagen and Haag 1990, §§2–4, printed pp. 275–284. Their result is operational and fixed-background: it neither makes local particle number invariant nor evolves the metric.

The phrase “negative-energy partner falls in” can summarize a balance law, but it is not a unique microscopic trajectory. With energy defined by the asymptotic stationary generator, a positive outward flux is accompanied in the semiclassical stress tensor by negative Killing-energy flux across the future horizon. That statement concerns a renormalized expectation value and a chosen generator. It does not turn one local member of an invariant pair into a negative-energy particle for every observer.

State, detector, and propagation are separate layers

The Boulware, Hartle–Hawking, and Unruh states encode different boundary conditions. The collapse state has no incoming thermal bath from past null infinity and gives late outgoing flux. For eternal Schwarzschild, the Hartle–Hawking state is regular on both horizons and describes thermal equilibrium, while the Boulware state is horizon-singular. This does not extend naively to asymptotically flat Kerr: bosonic superradiance obstructs a globally regular, isometry-invariant Hartle–Hawking state of the same kind. These distinctions are collected in Boulware, Hartle–Hawking, and Unruh States.

A detector supplies another layer. Its response depends on the Wightman function pulled back to its worldline, its energy gap, switching, and duration. An accelerated detector can respond in the Minkowski vacuum, while a freely falling detector near a large horizon need not see a local bath corresponding to the asymptotic Hawking temperature. Unruh’s collapse-equivalent boundary conditions and detector analysis make this observer dependence explicit Unruh 1976, complete printed pp. 870–892.

Propagation from the near-horizon region to infinity is a third layer. The curvature potential partly transmits and partly reflects each angular mode. For a fixed field species and polarization $a$ of spin $s_a$ on Schwarzschild, the asymptotic number flux has the schematic form

$$\frac{dN_a}{dt\,d\omega} =\frac{1}{2\pi}\sum_{\ell,m} \frac{\Gamma^{(a)}_{\ell}(\omega)} {e^{\omega/T_H}-(-1)^{2s_a}},$$

where $\Gamma^{(a)}_{\ell}$ is that species-polarization’s transmission probability; the sum over species $a$ is separate. Integer $s_a$ gives the Bose denominator and half-integer $s_a$ the Fermi denominator. Energy flux adds a factor of $\omega$. Page’s calculation shows explicitly that the horizon thermal factor is filtered into a spin- and mode-dependent spectrum at infinity Page 1976, §§II–IV, printed pp. 199–206, especially Eq. (4), Eqs. (11)–(12), and the transmission tables. The full propagation map is Greybody Factors and Asymptotic Flux.

Radiation, backreaction, and information are different questions

The fixed-background calculation establishes late particle or detector response and asymptotic flux on a specified collapsing geometry. It does not solve the semiclassical Einstein equation for the evolving metric, hence does not determine the full evaporation history, endpoint, interior, or breakdown under large curvature and backreaction.

The information problem instead asks whether complete quantum evolution, including gravity and the late geometry, maps an initial pure state to a final pure state and where accessible correlations reside. Approximate thermality of low-order modes does not prove exact factorization; the pair cartoon supplies no unitary recovery mechanism. Current recovery proposals add assumptions and dynamics beyond this stable card and require dated assessment.

The limits of the approximation are explained in Fixed-Background Limits and Backreaction. The surrounding QFT construction is Black-Hole Quantum Fields and Hawking Radiation.

Common pitfalls

A horizon is treated as a hot material surface. Hawking temperature uses asymptotic normalization and quantum correlations; the equivalence principle still holds at a regular large horizon.

Thermal means exactly blackbody at infinity. Greybody factors filter the spectrum and depend on field spin, angular mode, frequency, and boundary normalization.

Hawking radiation already solves evaporation. A flux on a fixed background is an input to backreaction, not the backreacted spacetime or its endpoint.

The pair story proves information loss or recovery. Either conclusion requires a global quantum state, a complete dynamical model, and explicit assumptions about the late geometry.

References

• Fredenhagen, Klaus, and Rudolf Haag. “On the Derivation of Hawking Radiation Associated with the Formation of a Black Hole.” Communications in Mathematical Physics 127 (1990): 273–284. DOI. Open PDF.
• Hawking, Stephen W. “Particle Creation by Black Holes.” Communications in Mathematical Physics 43 (1975): 199–220. DOI. Open PDF.
• Page, Don N. “Particle Emission Rates from a Black Hole: Massless Particles from an Uncharged, Nonrotating Hole.” Physical Review D 13 (1976): 198–206. DOI.
• Unruh, William G. “Notes on Black-Hole Evaporation.” Physical Review D 14 (1976): 870–892. DOI.