Do vacuum fluctuations prevent the creation of closed timelike curves? (chronology protection for wormhole time machines)

Kim and Thorne (1991) asked whether quantum vacuum fluctuations prevent a traversable wormhole, converted into a time machine by the relative motion of its mouths, from developing closed timelike curves (CTCs). They computed the renormalized stress-energy tensor of a massless, conformally coupled scalar field near the resulting chronology horizon and found that it diverges as the horizon is approached, but so weakly that the induced metric perturbation $\delta g\sim (l_P/D)(l_P/\Delta t)$ stays tiny until the proper time $\Delta t$ to the horizon reaches the Planck length $l_P$ ($D$ is the wormhole mouth separation). They conjectured that quantum gravity cuts off the divergence at $\Delta t\sim l_P$, so CTCs form; Hawking countered that semiclassical gravity remains valid until the invariant combination $D\Delta t$ falls to $\sim l_P^2$, where the perturbation reaches order unity, so backreaction destroys the horizon before CTCs appear: the chronology protection conjecture. Later work rigorously proved that the semiclassical framework breaks down at compactly generated chronology horizons (Kay-Radzikowski-Wald), that the divergence is not universal (Visser's Roman ring; Krasnikov's and Ori's time-machine spacetimes), that within Misner space the divergence is robust across couplings, masses, interactions, and vacuum choices, and that holographic and analogue models realize protection in specific settings. Yet the central question - whether the laws of physics, including quantum gravity, prevent the formation of closed timelike curves - remains open, with four concrete surviving facets: the validity of the semiclassical description at and beyond the horizon, the cutoff criterion, field robustness in the wormhole geometry, and the dynamical backreaction computation.

Background

General relativity admits solutions containing closed timelike curves (CTCs): timelike worldlines that return to their own past. If a spacetime develops CTCs only after some moment, in a region initially free of them, it is called a time machine, and the boundary between the region without CTCs and the region with them is a chronology horizon - a special case of a Cauchy horizon, the surface beyond which initial data on an earlier spacelike slice no longer determine the spacetime. A concrete classical route to such a spacetime was identified by Morris, Thorne, and Yurtsever (1988): a traversable wormhole (a short tunnel connecting distant regions, with throat radius $b$ and mouth separation $D\gg b$) requires matter that violates the weak energy condition (the requirement that every local observer measure non-negative energy density), and quantum fields can supply small negative energies (vacuum fluctuations, Casimir-type stresses). Accelerating one wormhole mouth, or setting up a gravitational redshift between the mouths, creates a growing time shift; once the shift exceeds the light-travel time between the mouths, CTCs appear behind a chronology horizon.

Whether quantum effects then destroy the incipient time machine is a question of semiclassical gravity: the classical Einstein equation sourced by the expectation value $\langle T_{\mu\nu}\rangle$ of the stress-energy tensor of quantized fields. This expectation value is defined by renormalization, most commonly by Hadamard point-splitting: the field's two-point function is evaluated at two nearby points, the universal short-distance singular piece shared locally by every physically admissible (Hadamard) state is subtracted, and the points are then collapsed. In an ordinary globally hyperbolic spacetime (one in which initial data on a spacelike slice determine the fields everywhere, so that no Cauchy horizon forms) this leaves a finite result. In a wormhole spacetime, however, two nearby points are connected by many geodesics - one direct one, and others winding through the wormhole $N=1,2,\dots$ times - and each multiply-winding geodesic carries its own singularity that renormalization does not remove. The events that can be joined to themselves by a closed null geodesic winding $N$ times through the wormhole form the $N$th polarized hypersurface $\mathcal{H}_N$; these hypersurfaces nest and accumulate on the chronology horizon $\mathcal{H}_-$.

Kim and Thorne computed the resulting renormalized stress-energy tensor for a massless, conformally coupled scalar field (conformal coupling being the curvature coupling under which a massless scalar is insensitive to conformal stretching of the metric). It diverges as the polarized hypersurfaces are approached, with order of magnitude $T^{\mu\nu}\sim (b/D)^N/[D(\Delta t)^3]$ in units $c=G=\hbar=1$, up to factors of order unity including in the winding exponent: each traversal attenuates the field amplitude by the wormhole's defocusing-lens factor of order $b/D$, so high-winding contributions are exponentially suppressed. The divergence is therefore extremely weak: in the rest frame of the wormhole mouths the induced metric perturbation is $\delta g^{\rm VP}\sim (l_P/D)(l_P/\Delta t)$, where $l_P$ is the Planck length - for a wormhole with $D\sim 1$ m, only of order $10^{-35}$ when one is within a Planck length of the horizon. Kim and Thorne conjectured that the concept of a classical spacetime fails at $\Delta t\lesssim l_P$, so quantum-gravity fluctuations smear out the divergence while it is still negligible and the CTCs form unimpeded. Hawking, in response, conjectured that the spacetime stays classical until the frame-invariant combination $D\Delta t$ falls to $\sim l_P^2$, by which point $\delta g^{\rm VP}\sim 1$; the vacuum polarization would then destroy the horizon. This is the chronology protection conjecture: the laws of physics prevent the appearance of closed timelike curves.

Subsequent rigorous work sharpened the setting. Kay, Radzikowski, and Wald (1997) proved that in any spacetime whose Cauchy horizon is compactly generated (generated by closed null geodesics whose past endpoints lie in a compact set, the situation of the wormhole time machine), no Hadamard state on the initial causal region extends to one that is regular on the horizon: the two-point function is necessarily singular at certain base points of the horizon (points where the horizon generators accumulate from the past), and the renormalized $\langle T_{\mu\nu}\rangle$ is ill-defined there, whether or not it diverges elsewhere. Parallel computations in Misner space - the quotient of flat spacetime by a Lorentz boost, the standard toy model of a chronology horizon - showed that divergences persist for nonconformal and massive fields and cannot be avoided by any vacuum choice once self-interactions are present. These results establish that the semiclassical description fails at the horizon itself, but they do not settle what replaces it: exceptional configurations (a ring of wormholes, no member of which is near chronology violation) make the semiclassical backreaction arbitrarily small arbitrarily close to the horizon, and time-machine spacetimes satisfying all the classical energy conditions have been constructed. What the complete theory - semiclassical gravity plus quantum gravity - does to an incipient time machine therefore remains open, and the cutoff dichotomy posed by Kim and Thorne has become a recognized test question for candidate theories of quantum gravity.

Problem Statement

This is the wormhole-time-machine form of Hawking's chronology protection conjecture, in the specific form left open by the Kim-Thorne analysis.

For a traversable-wormhole time machine - a spacetime with throat radius $b$ and mouth separation $D\gg b$ that develops closed timelike curves behind a compactly generated chronology horizon as a result of the arrangement or motion of its mouths (and for the broader class of spacetimes that develop CTCs from regular initial data behind such a horizon) - determine whether quantum effects prevent the creation of the closed timelike curves. The question presupposes the setting: whether macroscopic traversable wormholes are themselves permitted by the fundamental laws is background context (see Background), not a component of an answer. The single objective decomposes into the following jointly coupled questions - coupled components of the one objective, not alternatives to choose among:

1. Validity of the semiclassical description. Establish a defensible quantization prescription for fields on the non-globally-hyperbolic region at and beyond the chronology horizon, or prove that none exists within semiclassical physics. The Kim-Thorne computation assumes that modes defined at past null infinity propagate uniquely through the spacetime and that Hadamard point-splitting is valid there; this assumption provably fails at the horizon itself (Kay-Radzikowski-Wald) and its status in the region between the polarized hypersurfaces remains contested (Krasnikov).

2. The cutoff criterion. Decide where semiclassical gravity breaks down near the horizon: at proper time to horizon $\Delta t\sim l_P$ in the frame of the wormhole mouths (Kim-Thorne's conjecture, under which the vacuum-polarization divergence is cut off while still tiny, $\delta g^{\rm VP}\sim (l_P/D)(l_P/\Delta t)$, and CTCs form), at the frame-invariant $D\Delta t\sim l_P^2$ (Hawking's conjecture, under which the perturbation reaches order unity and the horizon is destroyed), or at a criterion that supersedes both. Attempts to test these conjectures may produce insight into candidate theories of quantum gravity.

3. Field robustness. Determine whether the conclusion is independent of the quantum field content in the wormhole geometry itself: the Kim-Thorne computation treats a noninteracting, conformally coupled, massless scalar; whether interacting fields, fermions, or gauge fields change the divergence structure or the backreaction in this geometry (not only in Misner-space analogues) is unestablished.

4. Dynamical backreaction. Compute the self-consistent response of the geometry - the semiclassical Einstein equations together with whatever quantum-gravity corrections item 2 dictates - through the epoch of horizon formation, and decide the outcome: destruction of the horizon, its conversion into a singularity, or uneventful passage into the region with closed timelike curves.

Two outcomes of the physics are open: (i) quantum backreaction, within established physics or a candidate theory of quantum gravity, prevents CTC formation in this class of spacetimes (or in a precisely delimited subclass of the wormhole-time-machine class); or (ii) a self-consistent construction within such a framework in which the chronology horizon is crossed and the CTC region forms with well-defined physics throughout. Partial, model-specific answers exist - Misner-space field analyses, holographic and analogue-gravity models, string-theoretic mechanisms - but all are set in models outside the wormhole-time-machine class above, so none settles the question in this class.

References

1. Sung-Won Kim and Kip S. Thorne, 'Do vacuum fluctuations prevent the creation of closed timelike curves?', Physical Review D 43, 3929-3947 (1991), DOI 10.1103/PhysRevD.43.3929, https://doi.org/10.1103/PhysRevD.43.3929
2. Michael S. Morris, Kip S. Thorne and Ulvi Yurtsever, 'Wormholes, Time Machines, and the Weak Energy Condition', Physical Review Letters 61, 1446-1449 (1988), DOI 10.1103/PhysRevLett.61.1446, https://doi.org/10.1103/PhysRevLett.61.1446
3. S. W. Hawking, 'Chronology protection conjecture', Physical Review D 46, 603-611 (1992), DOI 10.1103/PhysRevD.46.603, https://doi.org/10.1103/PhysRevD.46.603
4. John L. Friedman, Nicolas J. Papastamatiou and Jonathan Z. Simon, 'Failure of unitarity for interacting fields on spacetimes with closed timelike curves', Physical Review D 46, 4456-4469 (1992), DOI 10.1103/PhysRevD.46.4456, https://doi.org/10.1103/PhysRevD.46.4456
5. Matt Visser, 'From wormhole to time machine: Remarks on Hawking's chronology protection conjecture', Physical Review D 47, 554-565 (1993), DOI 10.1103/PhysRevD.47.554, https://doi.org/10.1103/PhysRevD.47.554
6. Tsunefumi Tanaka and William A. Hiscock, 'Chronology protection and quantized fields: Nonconformal and massive scalar fields in Misner space', Physical Review D 49, 5240-5245 (1994), DOI 10.1103/PhysRevD.49.5240, https://doi.org/10.1103/PhysRevD.49.5240
7. L. H. Ford and Thomas A. Roman, 'Averaged energy conditions and quantum inequalities', Physical Review D 51, 4277-4286 (1995), DOI 10.1103/PhysRevD.51.4277, https://doi.org/10.1103/PhysRevD.51.4277
8. Eanna E. Flanagan and Robert M. Wald, 'Does back reaction enforce the averaged null energy condition in semiclassical gravity?', Physical Review D 54, 6233-6283 (1996), DOI 10.1103/PhysRevD.54.6233, https://doi.org/10.1103/PhysRevD.54.6233
9. S. V. Krasnikov, 'On the quantum stability of the time machine', Physical Review D 54, 7322-7327 (1996), arXiv:gr-qc/9508038, DOI 10.1103/PhysRevD.54.7322, https://arxiv.org/abs/gr-qc/9508038
10. Bernard S. Kay, Marek J. Radzikowski and Robert M. Wald, 'Quantum field theory on spacetimes with a compactly generated Cauchy horizon', Communications in Mathematical Physics 183, 533-556 (1997), arXiv:gr-qc/9603012, DOI 10.1007/s002200050042, https://arxiv.org/abs/gr-qc/9603012
11. Matt Visser, 'Traversable wormholes: The Roman ring', Physical Review D 55, 5212-5214 (1997), arXiv:gr-qc/9702043, DOI 10.1103/PhysRevD.55.5212, https://arxiv.org/abs/gr-qc/9702043
12. Li-Xin Li and J. Richard Gott III, 'Self-consistent vacuum for Misner space and the chronology protection conjecture', Physical Review Letters 80, 2980-2983 (1998), DOI 10.1103/PhysRevLett.80.2980, https://doi.org/10.1103/PhysRevLett.80.2980
13. Claes R. Cramer and Bernard S. Kay, 'Thermal and two-particle stress-energy must be ill defined on the two-dimensional Misner space chronology horizon', Physical Review D 57, 1052-1056 (1998), arXiv:gr-qc/9708028, DOI 10.1103/PhysRevD.57.1052, https://arxiv.org/abs/gr-qc/9708028
14. S. Krasnikov, 'Quantum field theory and time machines', Physical Review D 59, 024010 (1999), arXiv:gr-qc/9802008, DOI 10.1103/PhysRevD.59.024010, https://arxiv.org/abs/gr-qc/9802008
15. William A. Hiscock, 'Quantized Fields and Chronology Protection', arXiv:gr-qc/0009061 (2000), https://arxiv.org/abs/gr-qc/0009061
16. Matt Visser, 'The quantum physics of chronology protection', in 'The Future of Theoretical Physics and Cosmology' (proceedings in honour of Stephen Hawking's 60th birthday), pp. 161-176 (2002), arXiv:gr-qc/0204022, https://arxiv.org/abs/gr-qc/0204022
17. Amos Ori, 'A class of time-machine solutions with a compact vacuum core', Physical Review Letters 95, 021101 (2005), DOI 10.1103/PhysRevLett.95.021101, https://doi.org/10.1103/PhysRevLett.95.021101
18. Miguel S. Costa, Carlos A. R. Herdeiro, J. Penedones and N. Sousa, 'Hagedorn transition and chronology protection in string theory', Nuclear Physics B 728, 148-178 (2005), arXiv:hep-th/0504102, DOI 10.1016/j.nuclphysb.2005.08.022, https://arxiv.org/abs/hep-th/0504102
19. Amos Ori, 'Formation of closed timelike curves in a composite vacuum/dust asymptotically-flat spacetime', Physical Review D 76, 044002 (2007), DOI 10.1103/PhysRevD.76.044002, https://doi.org/10.1103/PhysRevD.76.044002
20. John Earman, Christopher Smeenk and Christian Wüthrich, 'Do the laws of physics forbid the operation of time machines?', Synthese 169, 91-124 (2009), DOI 10.1007/s11229-008-9338-2, https://doi.org/10.1007/s11229-008-9338-2
21. S. Krasnikov, 'Time machines with the compactly determined Cauchy horizon', Physical Review D 90, 024067 (2014), DOI 10.1103/PhysRevD.90.024067, https://doi.org/10.1103/PhysRevD.90.024067
22. Barak Shoshany, 'Lectures on Faster-than-Light Travel and Time Travel', SciPost Physics Lecture Notes 10 (2019), arXiv:1907.04178, DOI 10.21468/SciPostPhysLectNotes.10, https://arxiv.org/abs/1907.04178
23. Juan Maldacena, Alexey Milekhin and Fedor Popov, 'Traversable wormholes in four dimensions', arXiv:1807.04726 (2018), published as Classical and Quantum Gravity 40, 155016 (2023), DOI 10.1088/1361-6382/acde30, https://arxiv.org/abs/1807.04726
24. Juan Maldacena and Alexey Milekhin, 'Humanly traversable wormholes', Physical Review D 103, 066007 (2021), arXiv:2008.06618, DOI 10.1103/PhysRevD.103.066007, https://arxiv.org/abs/2008.06618
25. Jean-Pierre Luminet, 'Closed Timelike Curves, Singularities and Causality: A Survey from Gödel to Chronological Protection', Universe 7(1), 12 (2021), arXiv:2101.08592, DOI 10.3390/universe7010012, https://arxiv.org/abs/2101.08592
26. Roberto Emparan and Marija Tomašević, 'Holography of time machines', Journal of High Energy Physics 2022(03), 212 (2022), arXiv:2107.14200, DOI 10.1007/JHEP03(2022)212, https://arxiv.org/abs/2107.14200
27. Roberto Emparan and Marija Tomašević, 'Quantum backreaction on chronology horizons', Journal of High Energy Physics 2022(02), 182 (2022), arXiv:2109.03611, DOI 10.1007/JHEP02(2022)182, https://arxiv.org/abs/2109.03611
28. Carlos Barceló, Jokin Eguía Sánchez, Gerardo García-Moreno and Gil Jannes, 'Chronology Protection Implementation in Analogue Gravity', European Physical Journal C 82, 299 (2022), arXiv:2201.11072, DOI 10.1140/epjc/s10052-022-10275-3, https://arxiv.org/abs/2201.11072
29. Valeri P. Frolov, Pavel Krtouš and Andrei Zelnikov, 'Ring wormholes and time machines', Physical Review D 108, 024034 (2023), arXiv:2305.03887, DOI 10.1103/PhysRevD.108.024034, https://arxiv.org/abs/2305.03887
30. Moustafa Ismail and David Mattingly, 'Metric affine gravity with dynamical chronology protection', arXiv:2604.08511 (2026), https://arxiv.org/abs/2604.08511