REVIEW 3 major objections 4 minor 14 references
Holomorphic family of strongly pseudoconvex domains in a K\"ahler manifold
T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read A holomorphic family of strongly pseudoconvex domains in a Kähler manifold has positive variation of Kähler-Einstein metrics whenever the total space is strongly pseudoconvex.
desk verdict A genuine generalization of Schumacher-style positivity to noncompact fibers, but the load-bearing mixed-derivative boundary estimate is cited from the Euclidean setting rather than proved here, so the verdict should wait on that check. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the variation form $\rho$, a smooth d-closed $(1,1)$-form that restricts to the complete Kähler-Einstein metric on every generic fiber. Its positivity is controlled by the geodesic curvature $c(\rho)$, defined by comparing $\rho^{n+1}$ with $\rho^n \wedge i\,ds\wedge d\bar{s}$; this is a smooth function on each fiber, and $\rho$ is positive-definite exactly when $c(\rho)>0$ on every fiber. The argument uses two identities: the elliptic equation $-\Delta c(\rho) + (n+1)c(\rho) = \|\bar{\partial} v_\rho\|^2$ on each fiber, and the boundary comparison $c(\rho)/c(\tau_r)\to 1$ as $x\to\partial D_y$, where $\tau_r = i\partial\bar{\partial}\log(-r)$ is the complete model metric built from the defining function $r$. The comparison turns the known blow-up of $c(\tau_r)$ at the boundary into the same blow-up for $c(\rho)$, which excludes the vanishing alternative supplied by the real-analytic maximum principle.
What would settle it
Compute the boundary asymptotics of the mixed derivatives $\phi_{\alpha s}$ and $\phi_{s\beta}$ for a strongly pseudoconvex domain family in a non-flat Kähler manifold, for instance with a Riemann-surface base carrying a nonzero curvature; if any such derivative fails to be $O(|r|^{-1/2-\varepsilon})$, then the comparison $c(\rho)/c(\tau_r)\to 1$ fails and Theorem 1.1 collapses.
Extended reading notes
Core claim
The central discovery is that the variation of Kähler-Einstein metrics $\rho$, defined on the smooth part $D'$ of the family by $\rho = \frac{1}{n+1}\Theta_{h_{X'/Y'}} + i\partial\bar{\partial}(-\log(-r)+\phi)$, is positive-definite on $D'$ whenever the total domain $D$ is strongly pseudoconvex in $X$. Because $\rho\vert_{D_y}$ equals the fiberwise Kähler-Einstein metric, the theorem is really about the base direction: the geodesic curvature $c(\rho)$, defined by $\rho^{n+1}=c(\rho)\,\rho^n \wedge i\,ds\wedge d\bar{s}$, is shown to satisfy an elliptic equation and to blow up to $+\infty$ at the boundary of every generic fiber. This forces $c(\rho)>0$ by the maximum principle and real analyticity, and hence forces $\rho>0$ on $D'$. A second result extends $\rho$ as a positive current across the singular fibers when the total space admits a complete Kähler metric whose restriction to the fibers has scalar curvature bounded below.
Load-bearing premise
The proof depends on the boundary decay rate for the mixed second derivatives of the Monge-Ampère potentials, $|\phi_{\alpha s}| = O(|r|^{-1/2-\varepsilon})$; if that rate is wrong for general Kähler bases, the ratio $c(\rho)/c(\tau_r)\to 1$ used to force positivity need not hold.
Editorial extensions
If this is right
- The relative canonical bundle $K_{D'/S'}$ acquires a smooth hermitian metric whose curvature form is $(n+1)\rho$, so it is positive in a strong sense.
- The boundary blow-up $c(\rho)\to\infty$ along each fiber gives a quantitative lower bound for the fiberwise Kähler-Einstein metric near the boundary.
- Under the completeness and scalar-curvature hypothesis, $\rho$ extends as a positive current across singular fibers, preserving positivity in a weak sense on the degenerate fibers.
- The result applies to arbitrary surjective holomorphic maps between Kähler manifolds, so curved bases are allowed and not only coordinate projections in $\mathbb{C}^n$.
Reading between the lines
- One consequence the authors leave implicit is that the positivity of $\rho$ can be read as a form of relative negative curvature in the base direction; this suggests a link to hyperbolicity of the family that is not developed here.
- A testable extension would be to prove the mixed-derivative estimate $|\phi_{\alpha s}|=O(|r|^{-1/2-\varepsilon})$ directly from bounded geometry of the Kähler base, removing the present reliance on the Euclidean argument.
- The scalar-curvature bound in the extension theorem is likely stronger than necessary; a local uniform bound on the fiberwise Kähler-Einstein volumes might suffice by itself, and searching for an example without the bound would clarify whether the hypothesis is sharp.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies a holomorphic family p:D→S of bounded strongly pseudoconvex domains in a Kähler manifold X, with the fibers assumed to admit complete Kähler-Einstein metrics. The family of fiberwise Kähler-Einstein metrics induces a smooth (1,1)-form ρ on D′, and the main theorem (Theorem 1.1) asserts that ρ is positive-definite on D′ whenever the total space D is strongly pseudoconvex. The proof follows Schumacher's framework: ρ satisfies an elliptic equation for its geodesic curvature c(ρ), and positivity is obtained by combining Yau's almost maximum principle with a boundary blow-up of c(ρ), proved by comparing c(ρ) with the geodesic curvature of the reference form τ_r=i∂∂(−log(−r)) on each fiber. A second theorem (Theorem 1.2) claims that under an additional uniform scalar-curvature lower bound, ρ extends across the singular fibers as a positive current, following Păun's method with Demailly approximation, Ohsawa-Takegoshi extension, and a Schwarz lemma volume estimate.
Significance. If correct, the result would extend the known positivity of variations of Kähler-Einstein metrics from compact fibers (Schumacher) and from Euclidean-coordinate families (Choi) to families of strongly pseudoconvex domains in an arbitrary Kähler manifold with arbitrary holomorphic base map. The use of Cheng-Yau theory and Schumacher's PDE is appropriate, and the overall strategy—boundary comparison of geodesic curvatures plus an almost maximum principle—is natural and promising. The paper also proposes an extension across singular fibers, which is a useful contribution. However, the analytic core of the main proof is not self-contained: several load-bearing estimates are delegated to earlier papers by the first author in the Euclidean setting, and the necessary modifications for a general Kähler base are not proved here. I see no circularity: the conclusion is not assumed, and the cited results are independent. The contribution would be significant if the missing estimates are supplied.
major comments (3)
- [§4.2, Proposition 4.5] The proof of Proposition 4.5 requires the boundary estimates |φ_{αs}|=O(|r|^{-1/2-ε}) and |φ_{sβ}|=O(|r|^{-1/2-ε}), but these are not proved in the present setting. The paper states that they follow by applying Schauder estimates to φ_s and φ_{\bar s}, with the citation "for detailed proof, see Section 3.3 in [5]". Reference [5] treats holomorphic families of domains in complex Euclidean space under the coordinate projection, where the reference metric has no additional s-dependence. In the present situation, ω0_{r_y}=ω0_y−i∂∂log(−r_y) and F_y in equation (3.1) depend on y through the ambient Kähler form Ric(ω_y) and through the defining function r_y. Differentiating (3.1) in s introduces terms involving ∂_s of the ambient Ricci form and of r, terms that are absent in the Euclidean case. These mixed-derivative estimates control the terms R1 and R2 in the expression for c(ρ)/c(τ_r), and their decay is essential for the claimed limit c(ρ)/c(τ_r)→1. Without a proof valid in the general Kähler setting, Proposition 4.5 is not established, and therefore the boundary blow-up c(ρ)→∞ in Proposition 4.3 and the strict positivity in Theorem 1.1 are unsupported.
- [§4.2, equations before Lemmas 4.6 and 4.7] The comparison in Proposition 4.5 also relies on two matrix identities, h^{βα}−(g0_r)^{βα}=(g0_r)^{βγ}N_{γδ}(g0_r)^{δα} and (g0_r)^{βα}−(gr)^{βα}=(gr)^{βγ}M_{γδ}(gr)^{δα}, together with Lemmas 4.6 and 4.7. These are cited from [4] (equations (5.3) and Lemmas 5.3–5.4), whose proofs are described as "essentially the same". However, [4] works in a Euclidean/coordinate-projection setting, and the general Kähler case introduces additional s-dependence through g0=(1/(n+1))Ψ_U and through ω0_y. In particular, Lemma 4.6 asserts (g0_r)^{βα}=O(|r|) in U∩D_V, and Lemma 4.7 gives the growth of N_y; both are needed to show that R2/c(τ_r)→0. Since the verification is entirely delegated and the transfer is not automatic, this is a second load-bearing gap in the proof of Proposition 4.5.
- [§5, Theorem 1.2] The proof of the extension theorem contains a gap in the volume estimate. After applying the Schwarz lemma (Theorem 5.4), the paper obtains (ωKE_y)^n≤C(~ω_y)^n on D_y and then states that it is enough to show ∫_{U_y}(~ω_y)^n<C, citing Theorem 5.6 (Diederich–Pinchuk) for the uniform boundedness of Euclidean volumes of analytic slices. This does not control the volume with respect to an arbitrary complete Kähler metric ~ω_D on D. In Remark 5.5 the metric is taken to be the Poincaré metric on a neighborhood U biholomorphic to the unit ball, but the restriction of the Poincaré metric to a slice U_y has infinite volume in general, so the claimed reduction cannot work as stated. Thus the uniform bound on VolKE(U_y), which is essential for Demailly's approximation argument, is not proved.
minor comments (4)
- [Throughout] The manuscript contains many typographical errors, including "extensioin", "Holomophic", "F amily", "str ongly", and inconsistent spacing in the title and abstract; a careful proofreading pass is needed.
- [§4.1] The reduction to the case of a one-dimensional base is stated without justification. The authors should explain why positivity of the geodesic curvature on every holomorphic disk pullback implies positive-definiteness of the (1,1)-form ρ on D′.
- [§4.2, equation (4.5)] The estimate |(ϕ_y)_{αβ}|=O(|r_y|^{n−3/2−ε}) is quoted for each fiber, but it is not stated whether the constants are uniform in y∈V; the subsequent Schauder argument would require such uniformity near the boundary.
- [§5, Theorem 5.4] In the statement of the Schwarz lemma, the inequality (i∂∂log V)^n≥K2V needs clarification: for V=(ωKE_y)^n the Ricci form is negative definite, so the sign of the left-hand side depends on the complex dimension n. The authors should state the theorem with absolute values or with the correct sign convention.
Circularity Check
No significant circularity: Theorem 1.1 is derived from Cheng–Yau existence, Schumacher's PDE, and independent boundary estimates, none of which assume the conclusion.
full rationale
The derivation chain for Theorem 1.1 does not presuppose the positivity of the variation of Kähler-Einstein metrics. The form ρ is constructed from the fiberwise Kähler-Einstein metrics produced by the Cheng–Yau theorem, and positivity is proven via Schumacher's elliptic equation (4.3), Yau's almost maximum principle, and the boundary blow-up of the geodesic curvature c(ρ). The key auxiliary estimates, including |φ_{αs}| and |φ_{sβ}| = O(|r|^{-1/2-ε}), are cited from the first author's earlier work [5] (and related estimates from [4]), but those estimates concern the boundary behavior of Monge-Ampère potentials and are not restatements of the target positivity. The cited results are independent, published, parameter-free analytic estimates with stated assumptions that do not include Theorem 1.1; they are load-bearing but not circular. The concern that these estimates may not transfer verbatim from the Euclidean coordinate-projection setting of [5] to a general Kähler base with an arbitrary holomorphic map is a correctness or gap issue, not circularity: the paper does not define strong pseudoconvexity in terms of ρ, does not fit a parameter to the conclusion, and does not rely on a self-citation that itself assumes the positivity of ρ. The self-citations are disclosed and their use is as sources of analytic lemmas, not as the sole justification of the main theorem. Therefore, no circular step is exhibited, and the appropriate finding is no significant circularity.
Assumptions & free parameters
assumptions (8)
- standard math Cheng-Yau existence and regularity for the complex Monge-Ampère equation on strongly pseudoconvex domains in Kähler manifolds (Theorem 2.1).
- standard math Schumacher's elliptic equation for the geodesic curvature c(ρ) (Proposition 4.1).
- standard math Yau's almost maximum principle and the real-analytic strong maximum principle for elliptic equations (Proposition 4.2).
- standard math Boundary asymptotics of the Monge-Ampère solution from Cheng-Yau and van Coevering, including |φ_{αβ}| = O(|r|^{n-3/2-ε}) and the Lee-Melrose optimal estimate (Theorem 2.3 and Remark 2.4).
- domain assumption Mixed-derivative boundary estimates |φ_{αs}|,|φ_{sβ}| = O(|r|^{-1/2-ε}), imported from Section 3.3 of [5] and asserted to remain valid in the general Kähler setting.
- standard math Proposition 4.4 and Lemmas 4.6 and 4.7 on boundary blow-up of c(τ_r) and regularity of the matrices N and M, taken from [4] (Remarks and Lemmas 5.3-5.4 there).
- standard math Mok-Yau Schwarz lemma for volume forms and Diederich-Pinchuk uniform volume estimates for holomorphic families of analytic sets (Theorems 5.4 and 5.6).
- domain assumption Standing hypotheses: for every generic fiber D_y, Ric(ω_y)<0; D is a bounded strongly pseudoconvex domain; p is proper on D; W is contained in an analytic subset; and D admits a complete Kähler metric ω~_D with fiberwise scalar curvature uniformly bounded from below.
Cite this review
Pith. "Pith review of Holomorphic family of strongly pseudoconvex domains in a K\"ahler manifold." pith.science (2026). https://pith.science/paper/ZMP7B2TA
@misc{pith2026190805842,
author = {Pith},
title = {Pith review of: Holomorphic family of strongly pseudoconvex domains in a K\"ahler manifold},
year = {2026},
howpublished = {\url{https://pith.science/paper/ZMP7B2TA}},
note = {Machine review of arXiv:1908.05842}
}
abstract
Let $p:X\rightarrow Y$ be a surjective holomorphic mapping between K\"ahler manifolds. Let $D$ be a bounded smooth domain in $X$ such that every generic fiber $D_y:=D\cap p^{-1}(y)$ for $y\in Y$ is a strongly pseudoconvex domain in $X_y:=p^{-1}(y)$, which admits the complete K\"ahler-Einstein metric. This family of K\"ahler-Einstein metrics induces a smooth $(1,1)$-form $\rho$ on $D$. In this paper, we prove that $\rho$ is positive-definite on $D$ if $D$ is strongly pseudoconvex. We also discuss the extensioin of $\rho$ as a positive current across singular fibers.
Reference graph
Works this paper leans on
-
[5]
Choi, A study of variations of pseudoconvex domains via K¨ ahler-E instein metrics , Math
Y.-J. Choi, A study of variations of pseudoconvex domains via K¨ ahler-E instein metrics , Math. Z. 281 (2015), 299–314
work page 2015
-
[4]
Choi, Variations of K¨ ahler-Einstein metrics on stronlgy pseudoconvex domains, Math
Y.-J. Choi, Variations of K¨ ahler-Einstein metrics on stronlgy pseudoconvex domains, Math. Ann. 362 (2015), 121–146
work page 2015
-
[1]
B. Berndtsson, M. Pun, Quantitative extensions of pluricanonical forms and close d positive currents, Nagoya Math. J. 205 (2012), 25–65
work page 2012
-
[2]
J. Cao, H. Guenancia, M. P˘ aun, Variation of singular K¨ ahler-Einstein metrics , arXiv:1710.01825
-
[3]
S.-Y. Cheng, S.-T. Yau, On the existence of a complete K¨ ahler metric on noncompact complex manifolds and the regularity of Fefferman’s equatio n, Comm. Pure Appl. Math. 33 (1980), no. 4, 507–544
work page 1980
-
[6]
Y.-J. Choi, G. Schumacher Extension of the curvature form of the relative canonical li ne bundle on family of Calabi-Yau manifolds and applications , arXiv:1804.00895
-
[7]
van Coevering, K¨ ahler-Einstein metrics on strictly pseudoconvex domain s, Ann
C. van Coevering, K¨ ahler-Einstein metrics on strictly pseudoconvex domain s, Ann. Glob. Anal. Geom. 42 (2012), 287–315
work page 2012
-
[8]
K. Diederich, S. Pinchuk Uniform Volume Estimates for Holomorphic Families of Analy tic Sets, Tr. Mat. Inst. Steklova. 235 (2001), 52–56
work page 2001
Show all 14 references
-
[9]
J. M. Lee, R. Melrose, Boundary behaviour of the complex Monge-Amp` ere equation , Acta Math. 148 (1982), 159–192
1982
-
[10]
Mok, S.-T
N. Mok, S.-T. Yau, Completeness of the K¨ ahler-Einstein metric on bounded dom ains and the characterization of domains of holomorphy by curvature conditions , The mathematical heritage of Henri Poincar´ e, Part 1 (Bloomington, Ind., 1980), 41 -59, Proc. Sympos. Pure Math., 39...
1980
-
[11]
M. P˘ aun,Relative adjoint transcendental classes and Albanese map o f compact K¨ ahler man- ifolds with nef Ricci curvature , Higher Dimensional Algebraic Geometry: In honour of Pro- fessor Yujiro Kawamata’s sixtieth birthday. Mathematical Society of Japan, 2017
2017
-
[12]
Saeki, Topology of Singular Fibers of Differentiable Maps , Lecture notes in Mathematics, Springer, 2004
O. Saeki, Topology of Singular Fibers of Differentiable Maps , Lecture notes in Mathematics, Springer, 2004
2004
-
[13]
Schumacher, Positivity of relative canonical bundles and applications , Invent
G. Schumacher, Positivity of relative canonical bundles and applications , Invent. Math. 190 (2012), no. 1, 1–56
2012
-
[14]
Y-.T. Siu, Curvature of the Weil-Petersson metric in the moduli space o f compact K¨ ahlerx- Einstein manifolds of negative first Chern class , Contri- butions to several complex variables, Hon. W. Stoll, Proc. Conf. Complex Analysis, Notre Dame/Indiana 19 84, Aspects Math. E9,...
1986
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.