Theoretical analysis reveals black hole unitarity does not refute conformal cyclic cosmology, indicating classical curvature and entanglement entropy remain fundamentally distinct.
Roger Penrose's Conformal Cyclic Cosmology (CCC) is now widely held to have been superseded by the discovery, over the past five years, that black hole evaporation is unitary. The argument for this claim is almost never stated in full; when it is, it turns out to rest on an equivocation. CCC's crossover mechanism requires the vanishing, at each aeon boundary, of a specific classical curvature invariant — the Weyl curvature — and not, as is generally assumed, the vanishing of entropy as such. The islands program and its associated replica-wormhole calculations establish a bound on a different quantity entirely: the fine-grained, quantum-informational entanglement entropy of Hawking radiation. Nothing in the definitions of these two quantities forces them to coincide. Penrose's own informal gloss on his hypothesis does not, in fact, keep them apart — he has stated in print that his mechanism requires genuine, quantum-mechanical information loss, siding openly against the modern unitary consensus — but this paper argues that the conflation is his, not his geometry's, and that the two notions can be prised apart without thereby rescuing or refuting CCC. This paper develops the distinction into a full argument, tests it against the natural objection that entanglement entropy might nonetheless backreact on classical curvature, and uses standard effective-field-theory reasoning to show that any such backreaction is suppressed by upwards of ninety orders of magnitude across nearly the entire evaporation history of a realistic black hole. What survives this analysis is not a defense of CCC but a relocation of the difficulty: the one regime in which the equivocation might still bite is precisely the regime of the black hole remnant problem, a dispute in quantum gravity considerably older than the islands program itself, and one CCC's own construction shows independent signs of having already registered without confronting directly.
No takes yet. Share an insight, caveat, or question.
Usman Aslam (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: