The cubic threefold is symplectically irrational

Jiaji Cai

REVIEW 3 major objections 4 minor 18 references

The cubic threefold is symplectically irrational

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The cubic threefold cannot be symplectically rational: a sixth-root monodromy obstruction blocks every symplectic blow-up path to projective space.

desk verdict A clean fractional-power computation for the cubic threefold, but the contradiction step leans on an unproved assertion about z-ramification from the author's companion paper. read the letter →

arxiv 2608.01577 v1 pith:RMXTMDNJ submitted 2026-08-03 math.SG math.AG

Jiaji Cai This is my paper · ORCID

classification math.SGmath.AG MSC 53D4514J3014N35

keywords symplecticrationalitycubicthreefoldquantumconnectionformalmonodromyirregularsingularityblow-upGromov-Witteninvariantsalgebraicirrationality

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 proves that smooth cubic threefolds are symplectically irrational: no finite sequence of symplectic blow-ups, blow-downs, and deformations can connect such a threefold to complex projective space. The obstruction is read off from the quantum connection near its irregular singularity: solutions of the horizontality equation contain fractional powers $z^{-1/6}$ and $z^{-5/6}$, while the quantum connections that can appear from blowing up along positive-genus Riemann surfaces produce only $z^{-1/2}$. The blow-up decomposition theorem for quantum connections preserves connections up to formal isomorphisms involving only integer powers of $z$, so the sixth-integer and half-integer power patterns cannot be identified. The theorem recovers the classical algebraic irrationality of cubic threefolds from a symplectic invariant and answers the open question posed in [14].

What carries the argument

The central object is the quantum connection $\nabla_z = \partial_z - z^{-2}(E_X \star) + z^{-1} \mu_X$, a meromorphic flat connection on the even cohomology of $X$ built from the genus-0 quantum product. The paper studies its formal solutions near the irregular singularity $z=0$ and records the fractional powers $z^{\rho}$ appearing in them, which encode the formal monodromy around the singularity. The structural input that makes these powers invariant under symplectic rationality is the blow-up decomposition theorem for quantum connections: under blow-up, the connection splits so that the piece preserving the multiplicity-2 Jordan block is formally isomorphic to the original, with transformations involving only integer powers of $z$. The proof also uses a companion classification saying such a block in a rational threefold must come from a positive-genus Riemann surface.

What would settle it

A concrete computation that would break the proof is to find a symplectically rational threefold whose quantum connection has a multiplicity-2 eigenvalue but no positive-genus Riemann surface origin, or to find a positive-genus Riemann surface quantum connection with a formal solution containing a factor $z^{-1/6}$ rather than only $z^{\pm 1/2}$; either outcome would falsify the classification the contradiction relies on.

Watch

Extended reading notes

Core claim

The paper's central claim is that the cubic threefold is symplectically irrational, meaning it cannot be related to complex projective space by any finite sequence of symplectic blow-ups, blow-downs, and deformations. The proof works with the small and then the big quantum connection, computes the formal solutions near the irregular singularity $z=0$, and isolates the fractional powers of $z$ that appear. For the cubic threefold, a rank-2 Jordan block of the connection forces solutions with factors $z^{\rho}$, $\rho \equiv \pm 1/6 \bmod \mathbb{Z}$. For a Riemann surface of positive genus, the same computation yields only solutions with factors $z^{\pm 1/2}$. Since a symplectically rational threefold would have its multiplicity-2 eigenvalue governed by the quantum connection of a positive-genus Riemann surface, and since the formal isomorphisms supplied by the blow-up decomposition theorem only involve integer powers of $z$, the two fractional-power patterns cannot coincide. This contradiction is Theorem 1.

Load-bearing premise

The proof depends on the companion classification that a multiplicity-2 eigenvalue in the quantum connection of any symplectically rational threefold must come from a positive-genus Riemann surface, with the corresponding part of the connection formally isomorphic to that surface's quantum connection; if that classification is not valid, the mismatch between the cubic's sixth-integer powers and the Riemann surface's half-integer powers does not force a contradiction.

Editorial extensions

If this is right

The cubic threefold cannot be symplectically rational, so no sequence of symplectic blow-ups, blow-downs, and deformations relates it to complex projective space.
The classical algebraic irrationality of smooth cubic threefolds is recovered from the symplectic statement; moreover the proof uses only even-degree cohomology with quantum information, not the middle Hodge structure that the classical proof centers on.
Together with the companion quartic-threefold result, the paper completes the symplectic rationality classification for smooth projective hypersurfaces of complex dimension at most three.
Formal monodromy of the quantum connection at the irregular singularity is exhibited as a workable symplectic irrationality obstruction: it distinguishes the cubic's sixth-integer fractional powers from the half-integer powers forced by positive-genus Riemann surfaces.

Reading between the lines

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

Extension: the same fractional-power test could be run on other threefolds with an irregular quantum connection; any manifold whose solutions show fractional powers not congruent to $\pm 1/2 \bmod \mathbb{Z}$ would be a candidate for symplectic irrationality.
Extension: because the proof ignores middle cohomology entirely, it suggests the even part of the quantum connection may already determine this rationality obstruction; checking formal monodromy for other rational threefolds and seeing only integral or half-integral powers would test that.
Extension: the sixth-root factors should have a mirror-theoretic echo; one could compute the B-model period integrals or oscillatory integrals of the mirror family and look for the same $z^{-1/6}$ monodromy as an independent check of the invariant.

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,

This paper answers a real question: it proves the cubic threefold is symplectically irrational via formal monodromy, recovering Clemens–Griffiths. The genuinely new piece is Section 3, where the small quantum connection is explicitly reduced and the fractional powers −1/6 and −5/6 emerge. That computation is transparent and, as far as I can tell, correct. The comparison with Riemann surfaces (Section 4) is also solid: the ±1/2 powers. If the proof goes through, it completes the threefold hypersurface classification and gives a new symplectic invariant.

The soft spot is precisely where the stress-test note lands. The contradiction in Section 5 depends on [1, Theorem 3.5] plus the claim that the formal isomorphism 'only involves integer powers of z' and that mirror maps and base change do not touch z. The paper neither states nor proves this. A ramified base change like z = w^3 would identify −1/6 with −1/2 mod Z, killing the invariant. This is not a pedantic worry; it's a specific alternative that the cited theorem needs to exclude, and the paper doesn't show that it does. Since the theorem comes from an unreviewed preprint, the circularity burden is real. I don't think the paper is wrong—I suspect the unramifiedness follows from the structure of formal monodromy in this setting—but the author needs to spell it out.

The rest of the manuscript is lean and readable, with a few typos and some under-explained passages in the big-to-small quantum connection step, which is sketched but probably fine. No code or machine-checked proofs are provided, but the Section 3 computation is reproducible by hand.

Who is this for? Anyone working on symplectic rationality, quantum connections, or mirror symmetry. It's a short, serious contribution that repays a close read.

My recommendation: send it to peer review. Not because I'm convinced—I'm not, until the unramified assertion is proved—but because the core idea is good and the computation deserves referee time. A referee should ask for a precise statement of [1, Thm 3.5] and a proof that the base change is unramified in z. That's a fixable gap if the author is right, and a fatal one if not. The paper is worth the engagement.

Referee Report

3 major / 4 minor

Summary. The paper claims to prove that smooth cubic threefolds are symplectically irrational. The argument studies formal solutions of the quantum connection in a ring containing formal fractional powers z^rho, and uses the fractional powers appearing in the formal monodromy as an obstruction. For the cubic threefold, an explicit computation in Section 3 yields solutions with fractional powers congruent to -1/6 and -5/6 modulo Z, attached to a rank-2 Jordan block. For a Riemann surface of positive genus, Section 4 yields fractional powers congruent to -1/2 and +1/2 modulo Z. Section 5 then invokes a classification theorem from the author's companion paper [1]: in a symplectically rational threefold, a multiplicity-2 eigenvalue must come from a positive-genus Riemann surface, and the corresponding Jordan block is formally isomorphic to that surface's quantum connection, with the isomorphism involving only integer powers of z. This is claimed to contradict the fractional-power computations, yielding Theorem 1 and recovering the Clemens-Griffiths irrationality theorem.

Significance. If the companion result [1, Theorem 3.5] is valid, this is a significant result: it gives a quantum-cohomology obstruction to symplectic rationality that uses only even-degree cohomology, and it recovers the classical Clemens-Griffiths theorem. The Section 3 computation is explicit and checkable, and the paper is clear about where it relies on [1]. The main weakness is that the decisive formal-isomorphism property is neither proved nor precisely stated in this manuscript, so the main theorem is not self-contained at the key step. The paper would be substantially strengthened by stating [1, Theorem 3.5] in full and proving the unramified-in-z property that carries the fractional-power invariant.

major comments (3)

[Section 5, proof of Theorem 1] The sentence 'the mirror maps and the base change do not involve the variable z' is the load-bearing step that turns the computations of Sections 3 and 4 into a contradiction, but it is asserted without proof or without a precise statement from [1]. This matters because a ramified base change z = w^3 would send z^{-1/6} to w^{-1/2} and z^{-5/6} to w^{-5/2} ≡ w^{-1/2} mod Z, so the fractional-power invariant would no longer distinguish the cubic Jordan block from the Riemann-surface Jordan block. The paper must state the exact formal-isomorphism theorem from [1] and prove that the isomorphism is unramified in z, or explicitly include this as a hypothesis with a proof.
[Section 5, proof of Theorem 1] The main theorem depends entirely on [1, Theorem 3.5] and on the companion paper's eigenvalue multiplicity classification, neither of which is proved or even stated in sufficient detail here. Since the fractional-power contradiction only follows if [1, Theorem 3.5] preserves the fractional part of rho modulo Z, this is not a peripheral citation but the central logical dependence. The manuscript should either reproduce the needed statement with a proof sketch or clearly separate it as an external input whose status is fully disclosed.
[Section 3, paragraph 'We claim that the operator ...'] The passage from the small quantum connection to the big quantum connection uses a gauge transformation S = M eS with M = I + sum M_n t^n, where the coefficients lie in the ring K. The paper does not show that M involves only integer powers of z, nor does it prove the recursive construction of M_n converges in the relevant completed ring. Since Proposition 6 is the input to Theorem 1, the preservation of rho mod Z under S = M eS needs to be justified explicitly; if M were allowed to contain fractional powers of z, the invariant could change.

minor comments (4)

[Section 1] There are typos in the introduction: 'fully anwers' should be 'fully answers', 'Ackowledgements' should be 'Acknowledgements', and 'Morevoer' in Proposition 6 should be 'Moreover'.
[Section 3] The 'direct computation' that yields the indicial equation rho^2 + rho + 5/36 = 0 is not shown. Since this equation is the only numerical input to the contradiction, displaying the reduced 2x2 system or the relevant recurrence would make the paper easier to verify.
[Section 4, Proposition 7] The statement that 'all solutions' have the displayed form should specify the coefficient field for a and b, since the solutions are elements of the tensor product with the formal solution ring R.
[References] Reference [12], arXiv:2308.13567, is dated 2026 in the bibliography, but the arXiv identifier suggests 2023; please check and correct the year.

Circularity Check

1 steps flagged · score 6.0 of 10

Main theorem is gated by an unproved, self-cited classification theorem in [1]; the crucial 'no z in the base change' assertion is exactly what is needed for the fractional-power contradiction.

uniqueness imported from authors [Section 5, proof of Theorem 1 (announced in Section 1)]
"By [1, Theorem 3.5], the Jordan block of A^{-1}∇_z A corresponding to this eigenvalue must be formally isomorphic to the quantum connection of Σ_g, possibly after pull-back along mirror maps and base change, on both sides. However, the mirror maps and the base change do not involve the variable z. The formal isomorphism in the end at most involves integer powers of z. Therefore, the solution S in Proposition 6 must also be a solution for Σ_g, possibly after formal gauge transformation. This contradicts Proposition 7."

The contradiction between the cubic's fractional powers ±1/6 mod Z (Prop. 6) and the Riemann surface's ±1/2 mod Z (Prop. 7) holds only if the formal isomorphism is unramified in the formal variable z. That is precisely the sentence 'the mirror maps and the base change do not involve the variable z; the formal isomorphism ... at most involves integer powers of z.' The paper neither states nor proves [1, Theorem 3.5], and that theorem is from the author's own companion paper, so the central premise is justified only by a self-citation. Moreover, a ramified base change z=w^3 would send z^{-1/6} to w^{-1/2} and z^{-5/6} to w^{-5/2} ≡ w^{-1/2} mod Z, so without the unproved no-z property the fractional-power invariant does not distinguish the two Jordan blocks.

full rationale

Sections 3 and 4 contain genuine, self-contained computations: the cubic small-quantum connection is explicitly decoupled and the fractional powers ±1/6 mod Z are derived from the indicial equation; the Riemann-surface computation is explicit. No fitted-input or self-definitional circularity appears there. The circularity is in the bridge: Section 5 invokes [1, Theorem 3.5] to assert a formal isomorphism with integer-power-only dependence on z, and Section 3 also transfers Jordan structure to the big quantum connection 'by the arguments of [1]'. Both are self-citations, and the z-unramified property is the exact condition needed for the contradiction; without it, a ramified variable change z=w^3 would collapse the fractional-power invariant. The paper does not reproduce or prove [1, Theorem 3.5], nor does it offer an independent verification. If [1, Theorem 3.5] is a valid theorem with the stated property, the argument can be completed, but as presented the central claim is gated by a load-bearing self-citation chain. Score 6 reflects partial circularity: the computations are independent, but the main conclusion is forced by the self-cited classification.

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

The proof introduces no fitted parameters; q is a formal Novikov variable. The main assumptions are the standard formal reduction theory for ODEs, the quantum cohomology formalism, Iritani's blow-up theorem, and the author's earlier results, with the latter being the most fragile. No invented entities are postulated.

assumptions (5)

standard math Turrittin-Wasow formal reduction: irregular ODE systems have formal solutions with fractional powers z^ρ and formal gauge transformations using integer powers of z.
Used in Section 3 to justify the decoupling of the ODE system and the form of the solutions.
domain assumption The quantum connection and the Gromov-Witten invariants satisfy Dubrovin's formalism, including the flatness and the commutation of ∇z with ∇ti.
Assumed throughout Sections 2 and 3; no proof given in the paper.
domain assumption Iritani's blow-up decomposition theorem (arXiv:2307.13555) applies to the quantum connection and holds on the level of formal monodromy.
Invoked in Section 5 to relate the quantum connection of a rational manifold to its blow-up components.
ad hoc to paper The results of the author's companion paper [1]: an eigenvalue of multiplicity 2 in a symplectically rational threefold must come from a Riemann surface of positive genus, and the corresponding Jordan blocks are formally isomorphic to that surface's quantum connection (Theorem 3.5).
This is the load-bearing step in the proof of Theorem 1. It is cited from an unreviewed preprint by the same author and is not independently established here.
domain assumption The big quantum connection has the same fractional powers and Jordan structure as the small quantum connection; the gauge transformation M and the formal diagonalization exist.
Asserted in Section 3; the construction is sketched but not carried out in detail.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The cubic threefold is symplectically irrational." pith.science (2026). https://pith.science/paper/RMXTMDNJ

@misc{pith2026260801577,
  author       = {Pith},
  title        = {Pith review of: The cubic threefold is symplectically irrational},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RMXTMDNJ}},
  note         = {Machine review of arXiv:2608.01577}
}

read the original abstract

We prove that smooth cubic threefolds are symplectically irrational. This recovers the classical irrationality theorem of Clemens-Griffiths. The obstruction is given by the formal monodromy of the quantum connection around the irregular singularity.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 14 canonical work pages

[1]

Jiaji Cai,The quartic threefold is symplectically irrational, 2026, arXiv:2605.29143

work page Pith review arXiv 2026
[2]

Zihong Chen,On the exponential type conjecture, 2024, arXiv:2409.03922

arXiv 2024
[3]

Herbert Clemens and Phillip A

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

work page 1972
[4]

1620, Springer, Berlin, 1996, pp

Boris Dubrovin,Geometry of2d topological field theories, Integrable systems and quan- tum groups (Montecatini Terme, 1993), Lecture Notes in Math., vol. 1620, Springer, Berlin, 1996, pp. 120–348. MR 1397274

work page 1993
[5]

J.165(2016), no

Sergey Galkin, Vasily Golyshev, and Hiroshi Iritani,Gamma classes and quantum cohomology of Fano manifolds: Gamma conjectures, Duke Math. J.165(2016), no. 11, 2005–2077. MR 3536989

work page 2016
[6]

V. V. Golyshev and D. Zagir,Proof of the gamma conjecture for Fano 3-folds with a Picard lattice of rank one, Izv. Ross. Akad. Nauk Ser. Mat.80(2016), no. 1, 27–54. MR 3462676

work page 2016
[7]

Math.172(2008), no

Jianxun Hu, Tian-Jun Li, and Yongbin Ruan,Birational cobordism invariance of uniruled symplectic manifolds, Invent. Math.172(2008), no. 2, 231–275. MR 2390285

work page 2008
[8]

III, Mem

Masuo Hukuhara,Sur les points singuliers des ´ equations diff´ erentielles lin´ eaires. III, Mem. Fac. Sci. Ky¯ usy¯ u Imp. Univ. A2(1942), 125–137. MR 21644

work page 1942

Show all 18 references

[9]

Hiroshi Iritani,Quantum cohomology of blowups, 2023, arXiv:2307.13555

2023 arXiv
[10]

Ludmil Katzarkov, Maxim Kontsevich, Tony Pantev, and Tony Yue YU,Birational invariants from Hodge structures and quantum multiplication, 2025, arXiv:2508.05105

2025
[11]

Malmquist,Sur l’´ etude analytique des solutions d’un syst` eme d’´ equations diff´ erentielles dans le voisinage d’un point singulier d’ind´ etermination

J. Malmquist,Sur l’´ etude analytique des solutions d’un syst` eme d’´ equations diff´ erentielles dans le voisinage d’un point singulier d’ind´ etermination. I, Acta Math. 73(1940), 87–129. MR 3897

1940
[12]

Daniel Pomerleano and Paul Seidel,The quantum connection, Fourier-Laplace trans- form, and families of A-infinity-categories, 2026, arXiv:2308.13567

2026 arXiv
[13]

Yasutaka Sibuya,Sur r´ eduction analytique d’un syst` eme d’´ equations diff´ erentielles ordinaires lin´ eaires contentant un param` etre, J. Fac. Sci. Univ. Tokyo Sect. I7(1958), 527–540. MR 96016

1958
[14]

Ivan Smith,Irrationality and monodromy for cubic threefolds, Ann. Sc. Norm. Super. Pisa Cl. Sci. (5)24(2023), no. 3, 1257–1284. MR 4675959

2023
[15]

W. J. Trjitzinsky,Analytic theory of linear differential equations, Acta Math.62 (1933), no. 1, 167–226. MR 1555383

1933
[16]

H. L. Turrittin,Asymptotic expansions of solutions of systems of ordinary linear dif- ferential equations containing a parameter, Contributions to the Theory of Nonlin- ear Oscillations, vol. II, Princeton Univ. Press, Princeton, NJ, 1952, pp. 81–116. MR 50754

1952
[17]

Singer,Galois theory of linear differential equa- tions, Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol

Marius van der Put and Michael F. Singer,Galois theory of linear differential equa- tions, Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol. 328, Springer-Verlag, Berlin, 2003. MR 1960772

2003
[18]

Wolfgang Wasow,Asymptotic expansions for ordinary differential equations, Pure and Applied Mathematics, vol. Vol. XIV, Interscience Publishers John Wiley & Sons, Inc., New York-London-Sydney, 1965. MR 203188

1965

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Jiaji Cai,The quartic threefold is symplectically irrational, 2026, arXiv:2605.29143

work page Pith review arXiv 2026

[2] [2]

Zihong Chen,On the exponential type conjecture, 2024, arXiv:2409.03922

arXiv 2024

[3] [3]

Herbert Clemens and Phillip A

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

work page 1972

[4] [4]

1620, Springer, Berlin, 1996, pp

Boris Dubrovin,Geometry of2d topological field theories, Integrable systems and quan- tum groups (Montecatini Terme, 1993), Lecture Notes in Math., vol. 1620, Springer, Berlin, 1996, pp. 120–348. MR 1397274

work page 1993

[5] [5]

J.165(2016), no

Sergey Galkin, Vasily Golyshev, and Hiroshi Iritani,Gamma classes and quantum cohomology of Fano manifolds: Gamma conjectures, Duke Math. J.165(2016), no. 11, 2005–2077. MR 3536989

work page 2016

[6] [6]

V. V. Golyshev and D. Zagir,Proof of the gamma conjecture for Fano 3-folds with a Picard lattice of rank one, Izv. Ross. Akad. Nauk Ser. Mat.80(2016), no. 1, 27–54. MR 3462676

work page 2016

[7] [7]

Math.172(2008), no

Jianxun Hu, Tian-Jun Li, and Yongbin Ruan,Birational cobordism invariance of uniruled symplectic manifolds, Invent. Math.172(2008), no. 2, 231–275. MR 2390285

work page 2008

[8] [8]

III, Mem

Masuo Hukuhara,Sur les points singuliers des ´ equations diff´ erentielles lin´ eaires. III, Mem. Fac. Sci. Ky¯ usy¯ u Imp. Univ. A2(1942), 125–137. MR 21644

work page 1942