Pith's one-line read
For every m-subharmonic function, the normalized ball maximum at a point always equals the m-Lelong number, and a new capacity-decay condition yields the conjectured L^p range for a large class of such functions.
desk verdict
Solid paper with real results, but the proof of the main maximum-identity theorem leans on an unstated tangent-uniqueness theorem that needs verification before publication.
read the letter →
A machine-rendered reading of the paper's core claim, the
machinery that carries it, and where it could break.
The reading
This paper establishes that for every m-subharmonic function — the class between subharmonic and plurisubharmonic in C^n, defined by a Hessian eigenvalue condition — the normalized limit of the maximum value on shrinking balls equals the m-Lelong number, a measure of singularity strength. It also introduces a scale of local Hessian-capacity conditions: if the capacity of deep sublevel sets decays like t^{-(m+δ)}, then the function is locally integrable with every power s < (m+δ)n/(n-m). At the critical case δ=0, this condition is satisfied by compact singularities, local energy classes, and radial germs, giving precisely the strict subcritical range predicted by the long-standing sharp integrability conjecture for these classes. The paper further gives explicit radial examples showing that the direct strong-openness analogue and the direct exponential-integrability analogue from the plurisubharmonic case both fail when m
What carries the argument
The main device is a scale of local Hessian-capacity conditions C_{m,δ}. For a compact set K inside a bounded m-hyperconvex domain D, the relative Hessian capacity Cap_m(E,D) is defined by a supremum of Hessian masses of comparison functions; the condition asks that the capacity of K∩{u<A-t} decays at least like t^{-(m+δ)}. Combined with a volume-capacity inequality and the layer-cake formula, this decay forces L^s integrability with the stated range. The second load-bearing object is the ball-maximum limit ℓ_u(a)=2 lim M(u,a,r)/φ_m(r), which is shown to equal the m-Lelong number by a tangent-rescaling argument. The radial Hessian test and the classification of power-logarithmic singularitie
What would settle it
Take any local m-subharmonic germ with positive m-Lelong number at 0 and compute both 2 lim_{r↓0} M(u,0,r)/φ_m(r) and the spherical-mean limit defining the m-Lelong number. The theorem predicts they are exactly equal; any germ where these differ — for instance one whose rescalings r^{2q}u(rw) converge to a non-radial limit with the same spherical mean but different maximum — would falsify it. Equivalently, checking whether the rescaled sequence always has an L1_loc limit of the form -ν/(2q)|w|^{-2q} settles the issue.
The paper's central identity is ν_u(a) = 2 lim_{r↓0} M(u,a,r)/φ_m(r), valid for every m-subharmonic function u at a point a, where M is the maximum over the ball of radius r and φ_m is the model function (-(r^{2q})/(q))^{-1} with q=(n-m)/m. The proof rescales u by r^{2q}, using a strong uniqueness theorem for tangents to force the rescaled functions to converge in L1_loc to the model singularity -ν/(2q)|w|^{-2q}; then the spherical-mean formula and a maximum principle squeeze the normalized maximum to the same limit. The paper's second main result is the capacity-decay criterion: if u satisfies C_{m,δ}, namely Cap_m(K∩{u<A-t},D) ≤ C t^{-(m+δ)} locally, then u ∈ L^s_loc for every s < (m+δ)n/(
Load-bearing premise
The proof of the ball-maximum identity relies on a strong uniqueness theorem for tangents of m-subharmonic functions whose precise hypotheses are not reproduced; if some local germ admits a non-radial tangent outside that theorem, the equality could fail.
Editorial extensions
If this is right
The equality between ball-maximum limit and m-Lelong number gives a practical way to read singularity strength from supremum asymptotics, without computing spherical means.
For radial m-subharmonic germs, the maximum identity turns into a direct verification of C_m, yielding the full strict range p < nm/(n-m) for all such germs.
Energy classes of finite Hessian mass automatically satisfy C_{m,p}, recovering known Sobolev-type integrability exponents and going beyond the critical range.
The counterexamples show that the classical route from bounded exponents to exponential integrability is blocked when m<n; a substitute must involve capacity or tangent profiles rather than only the m-Lelong number.
If the critical capacity condition C_m were proved for every local m-subharmonic germ, the long-standing sharp integrability conjecture would follow for all m between 1 and n.
Reading between the lines
Editorial extensions of the paper, not claims the author makes directly.
The capacity scale may be nearly necessary: it appears plausible that a germ satisfying the sharp L^p range with quantitative control on tails forces sublevel capacity decay of order at least m, so C_m could characterize the conjecture's range (a testable direction).
The failure of lower semicontinuity in the L1_loc topology is tied to the topology; stronger metrics based on capacities or energies might restore semicontinuity for the integrability-exponent functional and could be worth investigating.
Because radial germs satisfy C_m, a promising test toward the full conjecture is whether sums or convex combinations of radial singularities still satisfy C_m; a counterexample there would reveal an intrinsically non-radial obstruction.
The ball-maximum identity may extend to directional refinements — replacing balls with sectors or ellipsoids could yield directional m-Lelong numbers and sharp directional integrability, though the radial rigidity of tangents would be lost.