The lock principle for scalar curvature

We prove a Riemannian positive mass theorem for asymptotically flat spin manifolds with hypersurface singularities. Unlike earlier results, some components of the singular set may be mean-concave, provided that other components of the singular set are sufficiently mean-convex. Our proof uses initial data sets where a suitably chosen second fundamental form transfers convexity defects between different singularity components.

💡 Research Summary

The paper “The lock principle for scalar curvature” establishes a new version of the Riemannian positive mass theorem that allows hypersurface singularities with mixed mean‑convex and mean‑concave behavior. Classical results (e.g., Shi‑Tam 2002, Miao 2002) require each singular hypersurface Σ to be mean‑convex, meaning the jump in mean curvature H⁺−H⁻ is non‑positive. The authors relax this by considering a decomposition of the manifold M into regions M₀,…,M_N separated by smooth compact hypersurfaces Σ₁,…,Σ_N. Each region carries a smooth metric g_i with non‑negative scalar curvature, and the outermost region M_N is asymptotically flat.

The main theorem assumes that the mean curvature jumps satisfy a global quadratic inequality
∑_{i=1}^N (H_i⁻)² − (H_i⁺)² ≥ 0,
and that there exists an index Λ such that for i ≤ Λ the jumps are “bad” (H_i⁻ ≤ H_i⁺, i.e., mean‑concave) while for i > Λ they are “good” (H_i⁻ ≥ H_i⁺, i.e., mean‑convex). Under these hypotheses the ADM mass of (M,g) is non‑negative.

The proof introduces the “lock principle”. One defines numbers d_ℓ = Σ_{i=1}^ℓ (H_i⁺² − H_i⁻²) and c_ℓ = max{d_ℓ,0}^{1/2}. On each region M_ℓ a symmetric (0,2)‑tensor k_ℓ = c_ℓ g_ℓ is placed; thus k is piecewise a constant multiple of the metric and vanishes on the asymptotically flat end. This tensor plays the role of a “lock”: it converts an unfavorable mean‑curvature jump into a jump of k, transports that defect across the adjacent region, and finally absorbs it at a later hypersurface where the mean‑curvature jump is favorable.

A key technical ingredient is Lemma 2, a hyperbolic‑rotation construction. Given smooth functions H⁻, H⁺ on a hypersurface Σ and a real number a satisfying H⁺² − H⁻² + a² ≥ 0, one defines modified curvatures
Ĥ⁻ = √(H⁻² − a²), Ĥ⁺ = √(H⁺² − H⁻² + a²).
Choosing a hyperbolic angle ϑ appropriately yields a vector
X = F_ϑ(Ĥ⁻ − Ĥ⁺) = (X₁, X₂)
with X₁ ≥ |X₂| pointwise on Σ, where F_ϑ is the 2×2 matrix (cosh ϑ − sinh ϑ; −sinh ϑ cosh ϑ). This inequality is precisely the condition (1.7) in the work of Kazaras‑Khuri‑Lin (2025) that guarantees the dominant energy condition (DEC) holds weakly across Σ for the initial data set (M,g,k).

Applying Lemma 2 to each Σ_i with a = c_i−1 (or a = 0 when c_i = 0) produces the required hyperbolic angles ϑ_i, establishing X₁ ≥ |X₂| on every interface. Consequently the pair (M,g,k) satisfies the weak DEC across all Σ_i. The authors then invoke the positive mass theorem for creased initial data proved by Kazaras, Khuri, and Lin, which asserts that any asymptotically flat initial data set obeying the weak DEC has non‑negative ADM mass. Hence the ADM mass of the original Riemannian manifold (M,g) is non‑negative.

The paper concludes with several speculative extensions: (i) the “lock” could be used to absorb unfavorable k‑terms at infinity via an asymptotically hyperboloidal end, a technique relevant to Gromov’s total mean curvature conjecture; (ii) the positivity of the mean curvatures H_i^{±} is not essential, and allowing sign changes leads to a slightly stronger version of the quadratic condition; (iii) incorporating distances between singular hypersurfaces could weaken the global inequality, echoing recent work on positive mass theorems with shields and arbitrary ends; (iv) by suitable choices of k, any hypersurface Σ_i can be turned into a marginally outer trapped surface or a trapped surface, opening the possibility of combining the lock principle with Penrose‑type inequalities.

In summary, the authors present a novel “lock principle” that transfers curvature defects between singular hypersurfaces via a carefully constructed second fundamental form k. This global averaging mechanism enables a positive mass theorem that tolerates mean‑concave singularities, extending the scope of scalar curvature rigidity results beyond what local deformation techniques can achieve. The work bridges Riemannian geometry, spinor methods, and general relativity, and suggests a rich landscape of further applications.