Lagrangian interpretation of Abel-Jacobi mappings associated to Fano threefolds

Mark Green; Phillip Griffiths; Rodolfo Aguilar

REVIEW 4 major objections 4 minor 9 references

Lagrangian interpretation of Abel-Jacobi mappings associated to Fano threefolds

T0 review · 4 major / 4 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read For a Fano threefold, the pairing between tangent and normal directions of the curve family on an anti-canonical K3 surface reproduces the differential of the Abel-Jacobi map.

desk verdict A plausible and useful formula for dAJ_X via symplectic geometry, but Section 3's key diagram commutativity is under-proved; worth refereeing but not ready to cite. read the letter →

arxiv 2509.06197 v1 pith:62GLMCCJ submitted 2025-09-07 math.AG

Rodolfo Aguilar , Mark Green , Phillip Griffiths This is my paper · ORCID

classification math.AG MSC 14J4514K3014C3014D07

keywords FanothreefoldsAbel-JacobimappingintermediateJacobianLagrangiansubmanifoldsanti-canonicaldivisorsK3surfacessymmetricproductscubicthreefold

topics The Hodge Conjecture

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper aims to show that the differential of the Abel-Jacobi map from a family of curves on a smooth Fano threefold to its intermediate Jacobian is encoded in the symplectic geometry of the intersections of those curves with any smooth anti-canonical K3 surface. The central formula, Theorem A(ii), states that the symplectic pairing between the tangent direction α(ξ) and a normal direction β(η) in the symmetric product equals the pairing of the Abel-Jacobi differential applied to ξ with the cohomology class η. If true, this gives a new, largely symplectic route to classical results: the tangent bundle theorem for the Fano surface of a cubic threefold and the isomorphism between its Albanese variety and the intermediate Jacobian. The paper also sketches a relative version, Theorem B, in which deformations of the threefold play a role through a Deligne cohomology fibration.

What carries the argument

The load-bearing object is the symmetric product Y^{(d)} of a smooth anti-canonical K3 surface Y, together with its holomorphic symplectic form ψ. The map f sends each curve C_z to the d points C_z ∩ Y. The proof of Theorem A(ii) runs through the relative log complexes of the pair (X, Y) and the differential of the period map; the key mechanism is that the residue of a generator Ψ ∈ H^0(Ω^3_X(log Y)) along Y gives ψ, while the obstruction to lifting Ψ to a section over the deformation space lands in H^1(Ω^2_X), exactly the space of first-order deformations η. The formula ψ(α(ξ), β(η)) = ⟨dAJ_X(ξ), η⟩ is then the statement that the symplectic pairing of tangent and normal directions equals th

What would settle it

Take a smooth cubic threefold and a line L_z. The formula predicts ⟨dAJ_X(ν), ω_P⟩ = ψ(α(ν), β(ω_P)) for every P ∈ L_z^⊥; the left side is the explicit integral ∫_{L_z} ν ⌋ ω_P and the right side is computable from the intersection points L_z∩Y and the deformation of L_z with Y fixed. Checking one line for which both sides are non-zero would confirm or refute the identity.

Watch

Extended reading notes

Core claim

Let X be a smooth Fano threefold, Z the parameter space of a family of degree d curves C_z, and Y ∈ |-K_X| a smooth anti-canonical divisor meeting each curve transversely. Intersection induces a morphism f: Z → Y^{(d)}, and Y^{(d)} is hyperkähler with symplectic form ψ. Theorem A asserts that f(Z) is a maximal-dimensional Lagrangian submanifold of Y^{(d)}, and that for every tangent vector ξ of Z and every first-order deformation class η ∈ H^1(Ω^2_X), the identity ψ(α(ξ), β(η)) = ⟨dAJ_X(ξ), η⟩ holds, where α = f_* and β is a naturally defined map from H^1(Ω^2_X) to the normal space of f(Z) in Y^{(d)}. In words: the infinitesimal Abel-Jacobi map is the symplectic pairing between how the inter

Load-bearing premise

The load-bearing premise is that the symplectic form on the symmetric product pairs every tangent direction of the curve family non-trivially with some normal direction; without this non-degeneracy the map β is not well-defined and the central identity collapses, and the paper assumes rather than proves it (it is equivalent to f(Z) being maximal isotropic, the content of Theorem A(i)).

Editorial extensions

If this is right

For the Fano surface Z of lines on a smooth cubic threefold, dAJ_X is an isomorphism and the induced map Alb(Z) → J(X) is an isomorphism.
The tangent bundle theorem for the cubic threefold — that T_z Z is the restriction of the universal sub-bundle on the Grassmannian — follows directly from the surjectivity of dAJ_X^* together with the pairing formula.
For general conics on prime Fano threefolds of degree 10, the same argument yields that dAJ_X is an isomorphism.
Theorem A gives a symplecto-geometric description of the full infinitesimal Abel-Jacobi map: it is captured by the normal directions β(η) and the symplectic form ψ.
Theorem B predicts, with a proof sketched, that the relative family F(Z) is maximal Lagrangian in Y^{(d)} × D_λ, exactly compensating for the obstruction to deforming curves as X deforms keeping Y fixed.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The identity suggests that for any Fano threefold, the kernel of dAJ_X is characterized by the directions in which β(η) is symplectically orthogonal to the image of α; this could give a purely geometric test for when Abel-Jacobi maps degenerate.
Because the formula only needs a single smooth anti-canonical divisor, it may extend to singular or movable anti-canonical limits if the residue/logarithmic arguments are replaced by a limiting mixed Hodge structure argument.
The non-degeneracy of ψ, assumed rather than proved, may be equivalent to the unobstructedness of the family; if so, examples with obstructed deformations would be natural testing grounds for the formula.
The cubic-threefold argument that β is surjective uses the fact that H^1(Ω^2_X) ≅ V and that global sections of O_X(1) deform the threefold; the same method could be applied to other Fano threefolds with explicit cohomology to identify which curve families yield isomorphisms.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

The thing to know: this is a short note that fits the Donagi–Markman/Markushevich/Iliev–Manivel machinery to a Fano threefold, and the genuinely new output is the formula ψ(α(ξ), β(η)) = ⟨dAJ_X(ξ), η⟩ in Theorem A(ii). I think the formula is probably right, but the proof of Theorem A(ii) is not complete as written, and the paper should not be accepted without a rewritten Section 3.

What is good: the authors are clear about what is theirs. Theorem A(i) is credited to Iliev–Manivel and re-proved by a short, neat residue computation. The application to the Fano surface of a cubic threefold gives a quick route to the tangent bundle theorem and to the isomorphism J(X)→Alb(Z), using only the Lagrangian statement plus surjectivity of β. The genus-six prime Fano case is a natural bonus. The dimension count for Theorem B is clean, and the authors honestly say the rest of the proof will appear elsewhere.

The soft spots, in proportion. The main one is diagram (3.4), which is the load-bearing justification of Theorem A(ii). The paper says 'we will show below that this diagram commutes' and then offers a much smaller square. The map τ is described in words, not defined precisely, and the compatibility [∂,Res]=0 in the C∞ log complex with the period-map connecting map is asserted, not shown. There may well be a sign or a missing term; a referee has to check it line by line. The reader's worry about non-degeneracy of ψ turns out to be less serious: once (i) is proved, maximal isotropy gives the non-degeneracy, so it is an ordering issue rather than a missing hypothesis. Unobstructedness of the curve-family deformations is explicitly assumed, which is fine as a standing hypothesis but should be flagged as such. Theorem B is only a sketch, but that is disclosed.

If the diagram is fixed, this is a solid, useful note for people working on intermediate Jacobians and Fano threefolds. As it stands, it is a worthwhile preprint but not established. I would send it to a serious referee, with instructions to focus on Section 3 and require a complete proof of the diagram's commutativity before acceptance.

Referee Report

4 major / 4 minor

Summary. The paper proposes a Lagrangian interpretation of the differential of Abel-Jacobi maps for families of curves on smooth Fano threefolds. For a curve C_z in a family Z and a smooth anti-canonical divisor Y, the intersection points define a map f: Z -> Y^{(d)}. Theorem A(i) asserts that f(Z) is a maximal-dimension Lagrangian submanifold of the hyperkähler symmetric product Y^{(d)}; Theorem A(ii) asserts the formula psi(alpha(xi), beta(eta)) = <dAJ_X(xi), eta> for xi in T_z Z and eta in H^1(Omega^2_X). The paper applies this to recover the tangent-bundle theorem for the Fano surface of a cubic threefold and to discuss conics on genus-6 Fano threefolds. Theorem B states a Lagrangian statement for the relative family obtained by deforming X while keeping Y fixed; the paper gives only a dimension count for this theorem.

Significance. If Theorem A(ii) is fully established, the paper gives a genuinely new route to classical Abel-Jacobi statements: the differential of the Abel-Jacobi map is encoded in the symplectic geometry of the intersection cycles C_z · Y. The local-coordinate residue proof of Theorem A(i) is a transparent re-verification of the known Lagrangian statement, and the factorization in (4.3)-(4.5) is illuminating. The authors are explicit about the external framework [DM96, Mar08, IM07] and about the assumptions they make; there is no circularity in the main derivation. However, the central formula in Theorem A(ii) rests on a diagram whose commutativity is not actually proved, and Theorem B is explicitly deferred. The paper reads as a research announcement with the main theorem only partially supported, so the significance is real but conditional on completing the proof.

major comments (4)

[§3, diagram (3.4)] The proof of Theorem A(ii) reduces to the commutativity of diagram (3.4), but the manuscript only proves a proper square involving i_*, rho and tau. It does not justify the definitions/identifications needed for the full diagram: the map tau is only described as 'induced by considering a class in H^1(Omega^2_X)|_C as a (1,1) form in A^1(Omega^1_C) whose values are a conormal vector field', the compatibility of the connecting map of the period map with the residue/Gysin maps is not shown, and the upper and lower paths are asserted to equal psi(alpha(xi),beta(eta)) and <∇_eta Psi, AJ_X(xi)> without a complete diagram chase. Since this is the central claim of the paper, a full proof of the commutativity of (3.4) is required.
[§2, non-degeneracy of psi] The paper states 'we crucially use the non-degeneracy in both variables of the symplectically induced bilinear form psi: T_{f(z)}f(Z) ⊗ N_{f(Z)/Y^{(d)}, f(z)} -> C' before proving Theorem A(i). This non-degeneracy is used to identify the normal space with the dual of the tangent space and to define beta. It follows from f(Z) being Lagrangian, which is Theorem A(i), but the logical order and the proof itself are not made explicit. The reader needs either a direct proof of this nondegeneracy or a clear statement that it is a corollary of (i) established immediately after that proof.
[§5, Theorem B] Theorem B is stated as a theorem in the introduction, but Section 5 only proves the dimension formula (1.6). The text then says 'the remainder of the argument may be done by an adaptation of the arguments in [DM96]' and 'the complete proof of a more general version of Theorem B will be given elsewhere.' This is not a proof of the Lagrangian statement. If Theorem B is to be included as a result, the Lagrangian claim must be proved; otherwise the statement should be labelled as a conjecture or a program.
[§4, Example 1] The sentence 'With what was said above this implies that for the Fano surface dAJ_X is an isomorphism' is false as written: dim Z = 2 while dim J(X) = 5, so the differential of Z -> J(X) cannot be an isomorphism. What the subsequent argument needs is that the induced map dAJ_X^*: H^1(Omega^2_X) -> T_z^*Z is surjective (and indeed the next sentence uses exactly this). The wording should be corrected, and the distinction between the differential on Z and the induced Albanese map should be made carefully.

minor comments (4)

[§3, notation] The notation Z is used both for the parameter space and for the universal family in (3.1)-(3.4). The distinction between Z and the sheaf/space \mathcal Z is not clear in the printed text; please define both symbols explicitly.
[§3, equation (3.3)] The diagram (3.3) contains question marks and an unexplained '?' label. The connecting maps and the inclusion i_* should be labelled, and the exact sequence whose connecting map gives ∇Psi should be written out.
[References] Reference [IP99] is misspelled 'Iskoviskivh'; should be Iskovskikh.
[§4, geometric remark] The geometric proof of surjectivity of beta in Example 1 is only sketched. If kept, it should be expanded, especially the claim that the deformations X' obtained from sections P·Q fill out the normal space.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the derivation is based on external frameworks and the central formula is not equivalent to its inputs by construction.

full rationale

The paper's derivation chain is self-contained in the sense relevant to a circularity check. Theorem A(i) is proved directly in Section 2 by residue computations and a Riemann-Roch dimension count, and the non-degeneracy of the induced pairing ψ is not an assumed input equivalent to the conclusion but follows once f(Z) is shown to be isotropic of half the dimension of Y^(d). Theorem A(ii) is derived from the commutativity of diagram (3.4); the paper asserts 'We will show below that this diagram commutes' and verifies a sub-diagram, but the full commutativity — particularly the behavior of τ and the ∂–Res compatibility in the C^∞ log complex — is not completely demonstrated. That is a rigor/completeness gap, not a circularity: the claimed equality ψ(α(ξ),β(η)) = ⟨dAJ_X(ξ),η⟩ is not identical by construction to the definition of either side, and no fitted parameter is presented as a prediction. The only citation to the authors' own prior work is [CG72] (Griffiths is a coauthor), used at the end to name the classical tangent bundle theorem after it has been re-derived; it is not load-bearing. The frameworks of [DM96], [Mar08], and [IM07] are external and are not self-citations. No circular step can be exhibited, so the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper is a theoretical derivation built on external frameworks; it introduces no fitted numerical parameters and no new entities. The axioms are standard background and domain assumptions, with the non-degeneracy of ψ and the unobstructed deformation assumptions being the most load-bearing.

assumptions (5)

domain assumption X is a smooth Fano threefold with H^0(TX) = 0
Stated in Section 1: 'For simplicity of exposition we will assume that H^0(TX) = 0. Most of the Fano varieties that have non-trivial intermediate Jacobians will satisfy this condition.' It is used in the cohomology diagram (2.4).
domain assumption Y ∈ |−K_X| is a smooth anti-canonical divisor and C_z meets Y transversely
Assumed in Section 1 to define f: Z → Y^{(d)}; transversality is needed for the vertical equality in (2.6) and for the local coordinate argument in the proof of Theorem A(i).
domain assumption The deformation theory of the family of curves C_z ⊂ X is unobstructed, and Z, B_Z are smooth with Z → B_Z a smooth fibration
Assumed in Section 1 before the diagram (1.3): 'We assume that this theory is also unobstructed'. The dimension computation in (1.6) depends on this.
domain assumption The symplectically induced bilinear form ψ is non-degenerate in both variables on T_{f(z)} f(Z) ⊗ N_{f(Z)/Y^{(d)}, f(z)}
Explicitly flagged as crucial in Section 2: 'we crucially use the non-degeneracy in both variables of the symplectically induced bilinear form ψ'. No proof is given in the text.
domain assumption Existence of holomorphic symplectic form τ on D_λ and the Lagrangian fibration D_λ → B, obtained by generalizing Donagi-Markman to the relative case
Asserted in Section 1 as 'it may be proved' by generalizing [DM96] to the relative case of (X, Y)'s. Not proven in this paper; cited to [Mar08] and [DM96].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Lagrangian interpretation of Abel-Jacobi mappings associated to Fano threefolds." pith.science (2026). https://pith.science/paper/62GLMCCJ

@misc{pith2026250906197,
  author       = {Pith},
  title        = {Pith review of: Lagrangian interpretation of Abel-Jacobi mappings associated to Fano threefolds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/62GLMCCJ}},
  note         = {Machine review of arXiv:2509.06197}
}

read the original abstract

Using the general framework due to Donagi-Markman \cite{DM} and Markushevich \cite{M} we shall derive an expression for the differential of Abel-Jacobi mappings on Fano threefolds. This formula involves information normal to the Lagrangian submanifolds constructed in \cite{DM} and \cite{M}. It may be applied to give new proofs of a number of classical results about these varieties.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

9 extracted references · 8 canonical work pages

[1]

Torino, Turin, 2004, pp

Arnaud Beauville, Fano threefolds and K3 surfaces , The F ano C onference, Univ. Torino, Turin, 2004, pp. 175--184. 2112574

work page 2004
[2]

Herbert Clemens and Phillip A

C. Herbert Clemens and Phillip A. Griffiths, The intermediate J acobian of the cubic threefold , Ann. of Math. (2) 95 (1972), 281--356. 302652

work page 1972
[3]

Debarre, A

O. Debarre, A. Iliev, and L. Manivel, On the period map for prime F ano threefolds of degree 10 , J. Alg. Geom. 21 (2012), 21--59

work page 2012
[4]

1620, Springer, Berlin, 1996, pp

Ron Donagi and Eyal Markman, Spectral covers, algebraically completely integrable, H amiltonian systems, and moduli of bundles , Integrable systems and quantum groups ( M ontecatini T erme, 1993), Lecture Notes in Math., vol. 1620, Springer, Berlin, 1996, pp. 1--119. 1397273

work page 1993
[5]

Iliev and L

A. Iliev and L. Manivel, Prime F ano threefolds and integrable systems , Math. Ann. 339 (2007), 937--955

work page 2007
[6]

Iskoviskivh and Y

V. Iskoviskivh and Y. Prokhorov, Algebraic G eometry V: F ano V arieties , Springer-Verlag, 1999

work page 1999
[7]

Markushevich, An integrable system of K3 - F ano flags , Math

D. Markushevich, An integrable system of K3 - F ano flags , Math. Ann. 342 (2008), no. 1, 145--156. 2415319

work page 2008
[8]

z M^ [2] #1 #2 [1] split #1 split P enumerate [ ] enumerate [I.] enumerate [ (i) ] enumerate [ (a) ] enumerate [--] enumerate supp Ob Diff div Gy Nil Spec E xt Eff [ , ] * -2pt V Y^ (d) [1]

work page

Show all 9 references

[9]

enumi [1] #1\/ . [1] #1\/ : F_ lim Gor mytheo lemc [equation] Lemma mainth [equation] Main Theorem st [equation] Structure Theorem note Examo Example Lempp [equation] Lemma Definb [equation] Definition * definb Definition tr . equation obs [equation] Observation M U mathx U ma...

2004

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

Torino, Turin, 2004, pp

Arnaud Beauville, Fano threefolds and K3 surfaces , The F ano C onference, Univ. Torino, Turin, 2004, pp. 175--184. 2112574

work page 2004

[2] [2]

Herbert Clemens and Phillip A

C. Herbert Clemens and Phillip A. Griffiths, The intermediate J acobian of the cubic threefold , Ann. of Math. (2) 95 (1972), 281--356. 302652

work page 1972

[3] [3]

Debarre, A

O. Debarre, A. Iliev, and L. Manivel, On the period map for prime F ano threefolds of degree 10 , J. Alg. Geom. 21 (2012), 21--59

work page 2012

[4] [4]

1620, Springer, Berlin, 1996, pp

Ron Donagi and Eyal Markman, Spectral covers, algebraically completely integrable, H amiltonian systems, and moduli of bundles , Integrable systems and quantum groups ( M ontecatini T erme, 1993), Lecture Notes in Math., vol. 1620, Springer, Berlin, 1996, pp. 1--119. 1397273

work page 1993

[5] [5]

Iliev and L

A. Iliev and L. Manivel, Prime F ano threefolds and integrable systems , Math. Ann. 339 (2007), 937--955

work page 2007

[6] [6]

Iskoviskivh and Y

V. Iskoviskivh and Y. Prokhorov, Algebraic G eometry V: F ano V arieties , Springer-Verlag, 1999

work page 1999

[7] [7]

Markushevich, An integrable system of K3 - F ano flags , Math

D. Markushevich, An integrable system of K3 - F ano flags , Math. Ann. 342 (2008), no. 1, 145--156. 2415319

work page 2008

[8] [8]

z M^ [2] #1 #2 [1] split #1 split P enumerate [ ] enumerate [I.] enumerate [ (i) ] enumerate [ (a) ] enumerate [--] enumerate supp Ob Diff div Gy Nil Spec E xt Eff [ , ] * -2pt V Y^ (d) [1]

work page