← All problems

Bartnik’s cosmological splitting conjecture

Let (Mn+1,g)(M^{n+1},g) be a globally hyperbolic, timelike geodesically complete cosmological spacetime, meaning that MM has a compact Cauchy hypersurface. Then (M,g)(M,g) is isometric to a Lorentzian product (R×Σ,−dt2+h)(\mathbb{R}\times\Sigma,-dt^2+h), where Σ\Sigma is a compact manifold and hh is a Riemannian metric on Σ\Sigma.

References

Progress summary

Refreshed
Open · progress reported

A complete proof claim was withdrawn, and newer work only reduces the conjecture to unresolved analytic questions.

Bartnik’s conjecture, proposed in 1988, predicts that the relevant globally hyperbolic, timelike geodesically complete spacetime splits as a product. The general statement remains unsettled.

Recent developments: October–November 2026

  • A claimed positive resolution was withdrawn after its key Proposition 2.2 was shown false, reportedly by a counterexample of Ehrlich and Galloway (1990). The claimed splitting theorem therefore does not establish the conjecture.
  • On October 12, 2026, Bartnik’s splitting conjecture and the cosmological wall proved conditional splitting results and structural regularity for cosmological Busemann functions, but not the conjecture itself.
  • A November 1, 2026 report on Cosmological rigidity via pp-harmonic temporal functions reduces the problem to global weak solvability after proving that suitable functions force splitting; this is claimed partial progress, not a resolution.

Current status (as of November 2026): Bartnik’s cosmological splitting conjecture remains open; the complete proof claim was withdrawn, while the newer conditional and analytic reductions leave a global existence problem unresolved.

Sources

Solutions 0

No solutions have been posted yet.