REVIEW 5 minor 18 references
A degeneration's limiting discriminant comes from its Whitney strata
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
T0 review · deepseek-v4-flash
2026-08-01 16:31 UTC pith:LXFX2MTU
load-bearing objection Genuinely new (ω,−ω) conormal degeneration result; Whitney/Sabbah framework sound and well-attributed. Referee it, condition on cleanup of Example 3.21 and equation (9).
Degenerating Discriminants
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
For a Gröbner degeneration of a projective variety with weight vector ω, the relative conormal space—the closure of tangent hyperplanes to smooth points of the nearby fibers—is itself the Gröbner degeneration of the conormal variety with weight (ω,−ω). Over t≠0 its fibers are exactly the conormal varieties of the general fibers. At t=0, every irreducible component of the reduced special fiber is the conormal variety of some stratum of the minimal Whitney stratification of the total space compatible with the special fiber, so the limiting conormal cycle is a sum Σ m_S[N_S] over such strata. The multiplicities m_S are determined by a triangular system involving Euler characteristics of local M
What carries the argument
The central object is the relative conormal space N_{X/A¹}, the closure of tangent hyperplanes to smooth points of the fibers of a flat family. The argument runs through the minimal Whitney stratification of the total space compatible with the special fiber: this stratification is shown to satisfy the relative Thom condition for Gröbner degenerations, which controls the limits of tangent spaces of strata and thereby forces every component of the limiting conormal cycle to be the conormal variety of a stratum contained in the special fiber. Multiplicities are then assigned by a triangular system from the classical theory of specialization of conormal cycles, using local Milnor fibers and Eule
Load-bearing premise
The central claim collapses if the minimal Whitney stratification compatible with the special fiber fails to be preserved by the rescaling action on the family, since then the tangent-space limits used to identify components are not controlled.
What would settle it
Compute the flat limit of the relative conormal ideal, using the explicit ideal in Remark 3.2, for a Gröbner degeneration whose minimal Whitney stratification is known; if the primary decomposition of (N_{X/A¹})₀ contains a component not equal to the conormal variety of a stratum contained in X₀, or omits a stratum that does appear, Theorem 3.11 is false. A concrete test is to vary the weights in the family X=V(y₁²y₂−y₀²y₃+t²(y₃³+y₂³)) and check whether the limit still matches the stratum list.
If this is right
- The limiting dual hypersurface of any Gröbner degeneration factorizes into the duals of its Whitney strata, with strata contributing only when they are non-defective.
- For hypersurfaces with isolated singularities, the extra components are points whose multiplicities are μⁿ_p + μⁿ⁻¹_p, recovering the classical degree formula for dual hypersurfaces.
- For degenerations of complete intersections into transverse unions of smooth components, multiplicities are block products ∏(|J_i|−ε_i), computable as lengths of explicit local algebras.
- For reciprocal linear spaces, the dual degeneration splits into deletion and contraction cones with multiplicities 1 and 2, yielding the matroid beta-invariant degree formula by deletion–contraction induction.
- Higher associated hypersurfaces (Chow and Hurwitz forms) degenerate under induced Plücker weights, so limits of Hurwitz forms can be read off from the same strata; normal toric special fibers produce no extra Chow factors.
Where Pith is reading between the lines
- The same identification should hold for any flat projective family admitting a Whitney stratification of the map; the Gröbner assumption is used only to guarantee equivariance and the relative Thom condition, so non-Gröbner flat families may require finer strata or produce components not conormal to strata.
- Since multiplicities are local Milnor data, one could compute limiting discriminants numerically from Euler obstruction algorithms without first computing the full Whitney stratification.
- The Gelfand–Tsetlin examples suggest a general combinatorial recipe for toric degenerations: refine torus-orbit strata by the ω-tail hypersurfaces of the defining relations, then read off conormal multiplicities; verifying this would give enumerative formulas for Hurwitz degrees of Grassmannians from polytope data.
- The generalized Cayley trick connects the results to tropical initial degenerations of sparse resultants: initial forms of mixed discriminants factor into A-discriminants times duals of zero-dimensional strata, offering a recursive way to compute sparse discriminants.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies how projective dual varieties and discriminants behave under flat degenerations, focusing on Gröbner degenerations. The central construction (Section 3) shows that for a Gröbner degeneration X→A¹ with weight ω, the relative conormal space N_{X/A¹} is the Gröbner degeneration of the conormal variety N_X with weight (ω,−ω) (Prop. 3.3). Theorem 3.11 identifies every irreducible component of the reduced special fiber of N_{X/A¹} with the conormal variety of a stratum in the minimal Whitney stratification of the total space compatible with X₀. Theorem 3.16, following Sabbah/Kleiman, computes the multiplicities of these components via a triangular system involving Milnor fibers and local Euler obstructions. Proposition 3.20 then gives the factorization of the limiting dual hypersurface. Section 4 applies these results to hypersurfaces with isolated singularities, generic complete intersections, and reciprocal linear spaces; Section 5 extends to higher associated hypersurfaces via the Cayley trick; Section 6 connects to mixed discriminants of horizontally parametrized polynomial systems. The paper includes reproducible code for its examples.
Significance. If the main theorems are correct, the paper provides a systematic and computable description of how dual varieties and discriminants degenerate, combining classical results of Kleiman, Sabbah, Teissier, Lê, and BMM with Gröbner degeneration techniques. The central Proposition 3.3 is clean and the rank identity (8) is correct; Theorem 3.11 and Theorem 3.16 are coherent adaptations with honest attribution. A clear strength is that the paper ships reproducible code accompanying all examples, including the Macaulay2 computations of Whitney stratifications and Euler obstruction matrices. The applications are varied and illustrate the theory well. The main caveat is that some parts of the later sections (especially Lemma 4.7 and Theorem 5.2) are presented more sketchily, but these are extensions rather than the core Section 3 results.
minor comments (5)
- [Example 3.21] The computation of χ(F_{p1}) via the 'exceptional locus E' is not correct as written. Over the small Milnor-ball neighborhood of p1, the double cover a² = λb² + t²(1+λ³) is unramified: the equation a=0 has no solution with small (b,λ) when t≠0. The global curve E=V(λb²−t²(1+λ³)) is a plane cubic (indeed its affine part has χ=−3), not a local circle λb²=t² with χ=0. The final value χ(F_{p1})=2 is nevertheless correct—the local Milnor fiber consists of two disjoint disks—but the explanation should be replaced by the local double-cover argument.
- [Theorem 3.16] In equations (12) and (13), the summation condition 'S⊆T' should be 'S ⊆ \overline{T}' (equivalently, x_S ∈ \overline{T}). As written, the condition is vacuous for distinct strata, since strata are disjoint.
- [Lemma 4.7] The proof states 'One can check Thom's a_π-condition directly from the tangent space equations of these local models' without providing the check. Since this lemma is used to justify the complete intersection multiplicity formula, a few sentences or a reference for the local normal forms would be helpful.
- [Theorem 5.2 / Proposition 5.8] The passage from the flat degeneration of Y^∨ in dual Stiefel coordinates to the induced degeneration of Z_i(X) in Plücker coordinates is asserted without much detail. Because the Plücker map is only a rational map with indeterminacy locus, it is not immediate that the image of the flat limit equals the flat limit of the images; a fuller justification or a precise reference is needed.
- [Corollary 5.5] The displayed formula for the special fiber of the Hurwitz form is missing the exponent in 'Ch(Z) 2δp(C0)'; it should read Ch(Z)^{2δ_p(C0)}. Also, in the proof of Theorem 4.10, 'By Lemma 4.9' should refer to Proposition 4.9.
Circularity Check
No significant circularity: the central conormal-specialization theorem is proved by direct computation and external classical results.
full rationale
The main chain (Prop. 3.3 → Lemma 3.9 → Thm. 3.11 → Thm. 3.16 → Prop. 3.20) does not reduce to its inputs. Prop. 3.3 proves Con(X_t^sm) = (t^ω, t^{-ω})·Con(X^sm) by explicit Jacobian rescaling (Eq. 8), so the identification of the relative conormal with the Gröbner degeneration is not assumed. Lemma 3.9 proves the equivariance of the minimal compatible Whitney stratification from uniqueness together with connectedness of C^×, citing only standard sources [Gib+76], [Whi65/Tei82/Mat12] for existence/uniqueness. Thm. 3.11 then uses Thom's a_π condition from [BMM94] (external), and Thm. 3.16 is Sabbah's formula as presented in [Kle84] (external). The only self-citation that could look load-bearing is [BKW26, Prop. 2.8] in Thm. 4.10, but the same birational conormal parametrization is explicitly also attributed to independent work [MHT26, Prop. 4.1], and the deletion/contraction multiplicities are checked by direct factorization of the initial form. Self-citations such as [BB26] and [Bor+24] occur only in illustrative examples and are backed by Macaulay2 computations or external Khovanskii-basis results; they are not used to derive the general theorems. No fitted parameter is renamed as a prediction, and no ansatz is smuggled in via a self-citation. Therefore no circular step is exhibited.
Axiom & Free-Parameter Ledger
axioms (9)
- standard math Flatness of Gröbner degenerations and the flat-limit construction via scheme-theoretic closure
- domain assumption Existence and uniqueness of the minimal Whitney stratification of complex algebraic varieties, and existence of Whitney stratifications of maps
- domain assumption A Whitney stratification of the flat map π to A¹ satisfies Thom's a_π condition (via Thom's isotopy lemma and [BMM94, Thm 4.2.1])
- domain assumption Sabbah's formula (12) computing limiting conormal multiplicities from Milnor fibers and local Euler obstructions
- standard math Cayley trick: ρ^{-1}(Z_i(X)) = Y^∨ for Y = σ_i(X × P^{d−i})
- domain assumption Milnor's bouquet theorem for isolated hypersurface singularities and the Lê-Teissier formula Eu_{X0}(p) = 1 + (−1)^n μ^{n−1}
- standard math Deletion-contraction identity for the beta invariant of matroids
- domain assumption Universal Gröbner bases for reciprocal linear space ideals (circuit polynomials) and the birational conormal parametrization (x,y) ↦ x⋆x⋆y
- domain assumption Normality of Gelfand-Tsetlin toric varieties, Kushnirenko's theorem, and Khovanskii's genus formula
read the original abstract
We study the behavior of projective dual varieties and discriminants under flat degenerations. For Gr\"obner degenerations, we show that the conormal variety admits a Gr\"obner degeneration with opposite weights on the dual coordinates. Using Whitney stratifications and Sabbah's formula, we describe the irreducible components and multiplicities of the special fiber of this degeneration, and hence of the limiting dual hypersurface. We then extend these results to higher associated hypersurfaces in Grassmannians. As applications, we recover classical formulas for hypersurfaces with isolated singularities, analyze degenerations of generic complete intersections and reciprocal linear spaces, and relate the theory to mixed discriminants of parametrized polynomial systems.
Reference graph
Works this paper leans on
-
[1]
Gelfand-Tsetlin polytopes and Feigin- Fourier-Littelmann polytopes as marked poset polytopes
[ABS11] F. Ardila, T. Bliem, and D. Salazar. “Gelfand-Tsetlin polytopes and Feigin- Fourier-Littelmann polytopes as marked poset polytopes”. In:Journal of Combi- natorial Theory, Series A118.8 (2011), pp. 2407–2428. [Bas77] R. Bassein. “On smoothable curve singularities: local methods”. In:Mathematis- che Annalen230 (1977), pp. 273–277. [BB25] B. Betti an...
2011
-
[9]
Conormal spaces and Whitney stratifications
[HN22] M. Helmer and V. Nanda. “Conormal spaces and Whitney stratifications”. In: Foundations of Computational Mathematics(2022). [HS95] B. Huber and B. Sturmfels. “A Polyhedral Method for Solving Sparse Polynomial Systems”. In:Mathematics of Computation64.212 (1995), pp. 1541–1555. [HSS98] B. Huber, F. Sottile, and B. Sturmfels. “Numerical Schubert calcu...
2022
-
[12]
arXiv: 2604.24667. [Mil68] J. Milnor.Singular Points of Complex Hypersurfaces. Vol
-
[14]
Quelques remarques sur la g´ eom´ etrie des espaces conormaux
arXiv:2602.19563. [Sab85] C. Sabbah. “Quelques remarques sur la g´ eom´ etrie des espaces conormaux”. In: Diff´ erents aspects de la positivit´ e. Vol
-
[18]
Tangents to an analytic variety
[Whi65] H. Whitney. “Tangents to an analytic variety”. In:Annals of Mathematics81.3 (1965), pp. 496–549. [Zha23] Y. Zhang.Degeneration of Dual Varieties of Hypersurfaces
1965
-
[130]
A matroid invariant via the K-theory of the Grassmannian
Ast´ erisque. Soc. Math. France, 1985, pp. 161–192. [Spe09] D. E. Speyer. “A matroid invariant via the K-theory of the Grassmannian”. In: Advances in Mathematics221.3 (2009), pp. 882–913. [Sta86] R. P. Stanley. “Two poset polytopes”. In:Discrete & Computational Geometry 1.1 (1986), pp. 9–23. [Stu17] B. Sturmfels. “The Hurwitz Form of a Projective Variety”...
1985
-
[961]
Ensembles et morphismes stratifi´ es
Lecture Notes in Mathematics. Springer, 1982, pp. 314–491. [Tho69] R. Thom. “Ensembles et morphismes stratifi´ es”. In:Bulletin of the American Mathematical Society75.2 (1969), pp. 240–284. [Vak25] R. Vakil.The Rising Sea: Foundations of Algebraic Geometry. Princeton, NJ: Princeton University Press,
1982
-
[1976]
Springer Netherlands, 1977, pp. 199–265. [HM24] M. Helmer and R. Mohr.A New Algorithm for Whitney Stratification of Varieties. hal-04663472
1977
-
[1992]
Introduction to Chow Forms
[DS95] J. Dalbec and B. Sturmfels. “Introduction to Chow Forms”. In:Invariant Methods in Discrete and Computational Geometry. Ed. by N. L. White. Dordrecht: Springer Netherlands, 1995, pp. 37–58. [EH16] D. Eisenbud and J. Harris.3264 and All That: A Second Course in Algebraic Geometry. Cambridge: Cambridge University Press,
1995
-
[2002]
Limits of Tangents, Whitney Stratifications and a Pl¨ ucker Type Formula
[LT23] D. T. Lˆ e and B. Teissier. “Limits of Tangents, Whitney Stratifications and a Pl¨ ucker Type Formula”. In:Handbook of Geometry and Topology of Singularities IV. Cham: Springer, 2023, pp. 1–44. [LT88] D. T. Lˆ e and B. Teissier. “Limites d’espaces tangents en g´ eom´ etrie analytique”. In:Commentarii Mathematici Helvetici63.4 (1988), pp. 540–578. [...
2023
-
[2005]
Multigraded Cayley-Chow Forms
[OT19] B. Osserman and M. Trager. “Multigraded Cayley-Chow Forms”. In:Advances in Mathematics348 (2019), pp. 583–606. [PS06] N. Proudfoot and D. Speyer. “A broken circuit ring”. In:Beitr¨ age zur Algebra und Geometrie47.1 (2006), pp. 161–166. [PSS25] E. Pratt, L. Sodomaco, and B. Sturmfels.Multigraded Hurwitz Forms
2019
-
[2006]
Vari´ et´ es polaires. II. Multiplicit´ es polaires, sections planes, et condi- tions de Whitney
arXiv:math/0602052. [Tei82] B. Teissier. “Vari´ et´ es polaires. II. Multiplicit´ es polaires, sections planes, et condi- tions de Whitney”. In:Algebraic Geometry, La R´ abida
-
[2008]
arXiv:0811.1373. [GM88] M. Goresky and R. MacPherson. “Stratified Morse Theory”. In:Stratified Morse Theory. Berlin, Heidelberg: Springer, 1988, pp. 3–22. [Goe+23] K. Goel, V. Mukundan, S. Roy, and J. Verma. “Algorithms for computing mixed multiplicities, mixed volumes, and sectional Milnor numbers”. In:Journal of Software for Algebra and Geometry13 (2023...
Pith/arXiv arXiv 1988
-
[2016]
Stratification of projection maps from toric varieties
[EHHT25] B. El Hilany, M. Helmer, and E. Tsigaridas. “Stratification of projection maps from toric varieties”. In:Journal of Symbolic Computation(2025), p. 102547. [Eis03] D. Eisenbud.Commutative Algebra with a View Toward Algebraic Geometry. Vol
2025
-
[2022]
Notes on topological stability
arXiv:2205.08537. 32 [Mat12] J. N. Mather. “Notes on topological stability”. In:Bulletin of the American Mathematical Society49.4 (2012), pp. 475–506. [MHT26] S.-J. Matsubara-Heo and S. Telen.Principal Matroid Determinants
Pith/arXiv arXiv 2012
-
[2024]
Distributive lattices, affine semigroup rings and algebras with straight- ening laws
arXiv:2409.13288. [Hib87] T. Hibi. “Distributive lattices, affine semigroup rings and algebras with straight- ening laws”. In:Commutative Algebra and Combinatorics. Vol
-
[2025]
The Milnor number and deformations of complex curve singularities
arXiv:2506.06264. [BB26] V. Borovik and C. Briand.Degenerating Discriminants: Supplementary material. 2026.url:https://zenodo.org/records/21398038. [BG80] R.-O. Buchweitz and G.-M. Greuel. “The Milnor number and deformations of complex curve singularities”. In:Inventiones Mathematicae58 (1980), pp. 241–
Pith/arXiv arXiv 2026
-
[2026]
Singular loci of Grassmann-Hibi toric varieties
arXiv: 2606.26281. [BL10] J. A. Brown and V. Lakshmibai. “Singular loci of Grassmann-Hibi toric varieties”. In:Michigan Mathematical Journal59.2 (2010), pp. 243–267. [BMM94] J. Briancon, P. Maisonobe, and M. Merle. “Localisation de syst` emes diff´ erentiels, stratifications de Whitney et condition de Thom”. In:Inventiones Mathematicae 117.3 (1994), pp. 5...
Pith/arXiv arXiv 2010
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.