Chronology Horizons and Stress–Energy Growth

Renormalized stress often grows near model chronology horizons because repeatedly blueshifted or nearly null image contributions accumulate. It is not, however, a theorem that $\langle T_{ab}\rangle_{\rm ren}$ must diverge everywhere before every compactly generated horizon forms. The sharp result is microlocal: under specific scalar-field and Hadamard-extension hypotheses, the quantum theory becomes singular or ill-defined at base points of such a horizon.

Required background. Quantum Information supplies states, correlators, and reduced algebras.

Helpful background. Quantum Fields in Curved Spacetime supplies Hadamard subtraction; Einstein Equations supplies backreaction; Black-Hole Geometry and Mechanics supplies null generators; Semiclassical Gravity supplies stress validity; and Causality and Global Structure supplies Cauchy horizons and compact generation.

Compact generation and base points

Let $S$ be a closed, achronal, edgeless partial Cauchy surface and $D(S)$ its initial globally hyperbolic region. Its future Cauchy horizon is $H^+(S)$. The horizon is compactly generated if there is a compact set $K$ such that every past-directed null generator of $H^+(S)$ eventually enters and remains in $K$. This recurrence produces a nonempty base-point set $B$: past terminal accumulation points approached arbitrarily closely by horizon generators.

This definition is narrower than “a spacetime containing closed timelike curves.” Noncompactly generated horizons, chronal wormholes without a Cauchy horizon, different asymptotics, and geometries without the stated partial Cauchy surface need separate analysis. Kay, Radzikowski, and Wald give the precise geometry in Kay, Radzikowski, and Wald 1997, §§ 1–2, PDF.

Image sums and blueshift

In quotient models, a candidate two-point function often has an image representation

$$W(x,x')=\sum_{\gamma\in\Gamma}W_0(x,\gamma x').$$

As a chronology horizon is approached, infinitely many images can become nearly null related. A Hadamard singularity behaves locally as $W_0\sim1/\sigma_+$ in four dimensions, where $\sigma$ is Synge’s world function, one half of the squared geodesic interval. Away from the singular support, differentiation gives the schematic structure

$$\nabla_a\nabla_{b'}\frac1\sigma =\frac{2\sigma_{;a}\sigma_{;b'}}{\sigma^3} -\frac{\sigma_{;ab'}}{\sigma^2}.$$

Near coincidence, $\sigma$ scales as a separation squared and $\sigma_{;a}$ as a separation, giving the expected inverse-fourth-power short-distance stress scale. For distinct points approaching null separation, the derivatives of $\sigma$ need not vanish and a cubic inverse power can appear. There is no universal $\sigma^{-2}$ image law. Repeated blueshift and the sum over images make stress growth plausible, but the actual tensor contractions and approach determine its strength.

But the conclusion depends on image weights, field coupling, state, quotient identification, and component or frame. Cancellations can keep a particular renormalized stress component finite on approach. Such finiteness is weaker than local Hadamard regularity: a two-point function may fail the required wavefront-set condition even when one coincident-limit tensor happens to be finite.

Backreaction introduces a further logical step. A large stress on a prescribed chronology-forming metric suggests that the mean geometry will change, but the fixed-background result is not a self-consistent solution showing prevention. One must solve the constraints and causal response before claiming that the horizon cannot form.

Results and open questions through 2026-08-19

Evidence through 2026-08-19 supports strong obstructions for compact generation and several divergent model stresses. Whether stress necessarily diverges at a finite proper time before every candidate chronology horizon remains unresolved; no universal verdict follows.

Evidence and exact locator	Result and assumptions	Limitation or counterevidence
Hawking 1992, pp. 603–610	Blueshift and backreaction arguments motivate chronology protection for manufactured time machines	A conjectural physical argument, not a theorem for every quantum field, state, or causal geometry
Kay, Radzikowski, and Wald 1997, § 1, pp. 3–4, PDF	Natural-state calculations in Misner-like models often find growing or divergent stress near the horizon	The same discussion records special two- and four-dimensional states with finite $\langle T_{ab}\rangle$ on approach
Kay, Radzikowski, and Wald 1997, § 5, Theorem 2 and § 6, PDF	Any distributional bisolution extension of an initial Hadamard Klein–Gordon state fails local Hadamard boundedness at every base point	Establishes singular or ill-defined observables at base points, not divergence everywhere before the horizon
Kontou 2024, §§ 5.1–5.3, PDF	Quantum energy inequalities constrain negative-energy arrangements used in short wormhole time-machine proposals	Field, sampling, and geometric hypotheses do not convert the result into a universal stress-divergence theorem

The conservative microlocal position concludes exactly what the theorem supplies: ordinary local QFT cannot be extended through the base points with both the stated Hadamard behavior and local algebraic structure. The competing dynamical position treats generic blueshift growth as evidence that backreaction prevents the horizon earlier. That expectation is credible in many models, but finite-stress examples and the absence of a universal self-consistent evolution leave the stronger timing claim conjectural.

What a calculation must test

A useful stress calculation records the quotient or global geometry, field equation and coupling, state on $D(S)$, image prescription, point-splitting subtraction, freely falling frame, and approach curve. It must check conservation and state regularity away from the horizon, distinguish a coordinate component from an orthonormal invariant, and compare the stress with curvature and fluctuation scales.

Three independent tests are especially sharp:

1. Verify that the $n=0$ image reproduces the local Hadamard singularity and that subtraction removes only this universal piece.
2. Vary the approach direction to a base point; boundedness along one curve does not establish boundedness in a neighborhood of $(x,x)$.
3. Insert the stress into the semiclassical constraint or linear-response equation and test whether the assumed horizon geometry remains self-consistent before EFT or fluctuation control fails.

Flat spacetime with no identification provides a null limit: the extra image terms disappear and the chosen vacuum stress reduces to its renormalized reference value. A horizon that is not compactly generated requires a separate argument. Compact generation constrains the past tails of generators; it is not the same as compactness of the whole horizon.

Interpretation and open questionss

“Finite stress,” “Hadamard state,” and “extendible quantum theory” are different tests. Conversely, “ill-defined at a base point” does not identify the preceding backreacted spacetime or prove that nature forbids all closed timelike curves. The result marks a failure of the semiclassical QFT construction under its hypotheses; microscopic causal dynamics remain outside it.

The conclusions could change after a theorem extends or limits the compact-generation obstruction, an independently reproduced self-consistent evolution reaches or avoids a chronology horizon, a new Hadamard-state construction changes a finite-stress example, or a corrected image sum changes its asymptotics; a stronger QEI for the relevant sampling class or a verified noncompact counterexample could also change the conclusion.

Exercise

Suppose successive image separations satisfy $\sigma_n=q^n\sigma_0$ with $0\lt q\lt 1$, and each differentiated image contributes $C/\sigma_n^2$ with the same positive $C$. Does the image series converge? Why does this toy result not prove universal stress divergence?

Solution

The magnitudes form

$$\sum_{n=0}^{\infty}\frac{C}{\sigma_n^2} =\frac{C}{\sigma_0^2}\sum_{n=0}^{\infty}q^{-2n}.$$

Because $q^{-2}\gt 1$, individual terms grow and the series diverges. The conclusion assumes positive equal weights, the same tensor component, this geometric separation law, and interchange of differentiation and summation. Real states may have phases, cancellations, different field content, or different horizon generation. The microlocal theorem avoids those model-specific assumptions but gives a different conclusion at base points.

References

• Hawking, S. W. (1992). “Chronology Protection Conjecture.” Physical Review D 46, 603–611. DOI.
• Kay, B. S., Radzikowski, M. J., and Wald, R. M. (1997). “Quantum Field Theory on Spacetimes with a Compactly Generated Cauchy Horizon.” Communications in Mathematical Physics 183, 533–556. DOI. Open PDF.
• Kontou, E.-A. (2024). “Wormhole Restrictions from Quantum Energy Inequalities.” Universe 10, 291. DOI. Open PDF.