Weiss derivatives of holomorphic maps

Alexis Aumonier

REVIEW 3 major objections 4 minor 1 references

Weiss derivatives of holomorphic maps

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A finite Weiss tower, whose early layers are explicit configuration spectra, governs the stable homotopy type of holomorphic maps to projective space.

desk verdict A promising framework and a solid continuous-side theorem, but the main holomorphic polynomiality claim rests on a sketch that does not justify the Picard-family embedding. read the letter →

arxiv 2505.03731 v1 pith:6AFSKPJH submitted 2025-05-06 math.AT math.AG

Alexis Aumonier This is my paper · ORCID

classification math.ATmath.AG MSC 55P6555R80

keywords Weisscalculusunitaryholomorphicmapsprojectivespacetowerconfigurationspacesstablesplittingrational

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

The paper claims that the stable homotopy type of the space of holomorphic maps from a smooth projective variety $X$ to projective space is governed by a finite Weiss tower in unitary calculus. Concretely, it proves that the functor $V \mapsto \Sigma^\infty_+ \mathrm{Hol}_\alpha(X, \mathbb{P}(\mathbb{C}^{n+1}\oplus V))$ is $N$-polynomial, with $N=\dim H^0(X,L)$ for a very ample line bundle with $c_1(L)=\alpha$, and it computes the first $d(\alpha)$ layers of the tower explicitly as spectra made from configuration spaces of points on $X$. It also computes the complete Weiss tower for the continuous mapping-space functor. If the paper is right, the unstable homology of holomorphic mapping spaces is reduced to finitely many explicitly known spectra and extension problems between them.

What carries the argument

The load-bearing object is the Weiss tower of a unitary functor $F\colon \mathcal{J}\to \mathrm{Sp}$, whose $k$th layer is a homotopy quotient $(\Theta_kF\otimes S^{\mathbb{C}^k\otimes V})_{hU(k)}$. For holomorphic sections, the argument identifies sections of $L\otimes V$ with linear maps $H^0(X,L)^\vee\to V$ and stratifies by the dimension of the kernel, requiring the projectivised kernel to miss the embedded variety $X$. Each stratum is shown polynomial using a stable splitting of Stiefel manifolds; the same stratification, with the Picard variety as parameter, gives polynomiality of holomorphic maps. For continuous sections, the restriction $V\mapsto S(L\otimes(\mathbb{C}^M\oplus V))$ turns the section-space tower into the Weiss tower, producing the explicit layer spectra. A homology connectivity theorem then identifies the holomorphic and continuous layers for all $k\le d(\alpha)$.

What would settle it

Evaluate both sides of the identification $\Theta_k\simeq \tilde{\Theta}_k$ for $k=d(\alpha)$ on a known variety, such as $X=\mathbb{P}^1$ with degree $d$, and compare their integral homology in degrees below $2(d(\alpha)+1)\dim V+d(\alpha)-2$; the smallest degree where the two spectra differ must lie above that bound, so any disagreement below it would disprove the identification.

Watch

Extended reading notes

Core claim

The central discovery, stated on the paper's own terms, is that the holomorphic unitary functor is $N$-polynomial and its Weiss tower has layers $(\Theta_k \otimes S^{\mathbb{C}^k\otimes V})_{hU(k)} \to T_k(V)\to T_{k-1}(V)$, with $\Theta_k$ for $1\le k\le d(\alpha)$ given by $[\mathrm{Ind}_{U(1)\wr\Sigma_k}^{U(k)} \Sigma^{k(2n+1)} \mathrm{Conf}_k(X,S(L))^{-kT_X}] \times \mathrm{Map}(X,U(1))$. For continuous maps the analogous tower converges and the same layer formula holds for every $k\ge 1$. The mechanism is a rank filtration on the linear maps representing holomorphic sections, a stable splitting of Stiefel manifolds, and a comparison between the section-space tower and Weiss's tower; the two towers agree in the range where holomorphic and continuous mapping spaces have the same homology.

Load-bearing premise

The explicit formulas for the holomorphic layers depend on a connectivity bound, quoted from the author's earlier paper, claiming that holomorphic and continuous maps agree in homology below a stated degree range.

Editorial extensions

If this is right

For $1\le k\le d(\alpha)$, the unstable homology of $\mathrm{Hol}_\alpha(X,\mathbb{P}(\mathbb{C}^{n+1}\oplus V))$ is governed by the explicit configuration spectra $\mathrm{Conf}_k(X,S(L))^{-kT_X}$ with $U(1)\wr\Sigma_k$ and $\mathrm{Map}(X,U(1))$ actions.
The continuous functor $V\mapsto \Sigma^\infty_+\mathrm{Map}_\alpha(X,\mathbb{P}(\mathbb{C}^M\oplus V))$ has a converging Weiss tower, so its stable homotopy type is completely determined by the same layer formula for every $k\ge 1$.
The top derivative $\Theta_N$ is expressed as $\mathrm{ad}_N\otimes \Sigma^\infty_+PU(N)\otimes \Sigma^\infty\Sigma^{\mathrm{un}}|\mathrm{Gr}(X,L)|$, giving the first general description of the top layer.
For pointed degree-$d$ maps $\mathbb{P}^1\to \mathbb{P}(\mathbb{C}^2\oplus V)$, the functor is $d$-polynomial and its Weiss tower splits, which yields a new proof of the classical stable splitting of spaces of rational maps.
The polynomiality degree $N=\dim H^0(X,L)$ is independent of the chosen very ample $L$ under the ampleness hypothesis.

Reading between the lines

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

The explicit bottom layers suggest a practical route to unstable homology: compute a spectral sequence from the layers $\Theta_1,\dots,\Theta_N$ and a few extension problems; the paper does not carry this out, and it is a natural next step.
For $X=\mathbb{P}^n$ with $n\ge 2$, $N=\binom{n+d}{d}$ grows much faster than $d(\alpha)=d$, so the first uncomputed layer $\Theta_{d(\alpha)+1}$ is likely where most of the unstable complexity concentrates; computing it is a concrete open problem suggested by the paper.
The same rank-filtration mechanism should extend to holomorphic maps into other homogeneous varieties, such as Grassmannians, where kernel and span conditions define analogous strata and the same pushout arguments apply.
The appearance of the geometric realisation of the poset of projective subspaces avoiding $X$ in the top derivative points to a direct connection between unstable holomorphic-map topology and the topology of spaces of linear subspaces disjoint from a subvariety, which could be studied independently.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Two things to know. First, this is one of the few papers that actually uses Weiss calculus to say something about holomorphic mapping spaces, and the continuous-side Theorem B looks like a complete, well-documented result. Second, the proof of Theorem A has a real gap: the Picard-parameter step in Theorem 3.10 is explicitly a sketch, and the stress-test concern lands. The hypotheses give one very ample L, not very ampleness of every line bundle in Pic^α(X). Nothing in the text proves the relative evaluation map is an embedding over every point of the Picard variety, and if some [L] is not very ample, the identification with Ψ(V) and the subsequent stratification do not follow. That is load-bearing for the polynomiality claim, not a cosmetic omission.

What is genuinely good: the continuous-case Theorem B is proven carefully via Malkiewich's tower and Goodwillie calculus, with an explicit formula for all layers. The rank-stratification idea is attractive and yields concrete computations in Section 6. The new proof of the CCMM splitting is also a nice application, and it does not depend on the problematic part of Theorem 3.10. The paper is honest about its own sketches, which is to its credit, and the self-citation [Aum24] is an external connectivity input, not a circular restatement.

Soft spots in proportion: Theorem 3.10 is the main one, and it is serious. The proof says 'we will be content with only indicating the most salient modifications' and then handwaves the key point that the family evaluation map is an embedding. The author should either prove that, or restrict the statement to the fixed-line-bundle case (Theorem 3.2), or add an extra hypothesis like every L in Pic^α(X) being very ample. The second soft spot is Proposition 6.8, where the colimit identification is justified 'by inspection'; that is not a proof, though it is a less central part of the paper.

Who this is for: homotopy theorists working on calculus and spaces of holomorphic maps. The paper deserves a serious referee: the continuous-side theorem is likely correct and useful, and the framework is promising. But a referee should press hard on Theorem 3.10 before taking Theorem A as established.

Referee Report

3 major / 4 minor

Summary. The paper develops a unitary- (Weiss-) calculus approach to the stable homotopy type of spaces of holomorphic maps to projective space. For a connected smooth projective complex variety X and a class α admitting a very ample line bundle L with c1(L)=α and c1(L⊗K_X^∨) ample, Theorem A asserts that the functor V ↦ Σ^∞_+ Hol_α(X, P(C^{n+1}⊕V)) is N-polynomial with N = dim H^0(X,L), and that for 1 ≤ k ≤ d(α) the Weiss derivatives are given by explicit Thom spectra over labelled configuration spaces. Theorem B computes the complete Weiss tower for the analogous continuous mapping space. The paper also gives a new proof of the Cohen–Cohen–Mann–Milgram stable splitting for rational maps and computes several examples, including a formula for the top derivative.

Significance. If the proof gaps are closed, this is a valuable and genuinely new organizing framework for the unstable part of the homology of holomorphic mapping spaces. The continuous case (Theorem 4.16) is proved in detail through Malkiewich's and Goodwillie's calculi, and the passage from Goodwillie to Weiss towers is carefully set out. The explicit derivative formulas in Theorem A are concrete, parameter-free, and checkable, and the paper is honest in indicating where it sketches rather than proves. The potential pay-off is high: the unstable range is reframed as a finite extension problem in a Weiss tower.

major comments (3)

[Section 3.3.2, proof of Theorem 3.10] The N-polynomiality of V ↦ Σ^∞_+ Hol_α(X, P(C^{n+1}⊕V)) is the first half of Theorem A, but the proof of Theorem 3.10 is explicitly a sketch: after introducing Ψ(V), the text says 'we will be content with only indicating the most salient modifications.' The parameterized versions of Lemmas 3.4–3.6 are never stated, and the homotopy pushout squares S(ν_r(V)) → D(ν_r(V)) ≃ Ψ_r(V), Ψ_{<r}(V) → Ψ_{≤r}(V) are not verified over the non-discrete base Pic^α(X). In particular, the identification of Hol_α(X,P(C^{n+1}⊕V)) with the C^×-quotient of Ψ(V) relies on the relative evaluation map ϵ: Pic^α(X)×X → P((p_*P)^∨) being an embedding over every [L]; the hypotheses only provide one very ample representative L in the class α, and very ampleness is not a numerical property, so this is not established. Since the explicit formulas for Θ_k in Theorem A are deduced only after Theorem 3.10, this gap is load-bearing and must be repaired.
[Section 6.3, Proposition 6.8] The formula for the top derivative Θ_N is advertised in the introduction, but its proof contains two unproved identifications. The sentence 'By inspection of our constructions, the morphisms between the derivatives are all induced by the natural projections and inclusions' identifies the maps in the Čech diagram (16), and the subsequent 'observe that this colimit is exactly the geometric realisation of the poset Gr(X,L)' identifies the colimit with |Gr(X,L)|; neither is demonstrated. The final passage from the section-space functor (14) to the holomorphic-map functor (20) is also left to the reader. Please provide complete arguments or explicitly mark these statements as conditional.
[Section 5, proof of Lemma 5.1] The proof of Lemma 5.1 states that 'the argument in the proof of Theorem 3.2 will show' d-polynomiality of the pointed rational-map functor, but the rank-filtration argument is not carried out for the Veronese embedding ν_d: P^1 → P^d. The pointed condition changes the linear-algebra model, so this is not a purely formal consequence of Theorem 3.2. Since Theorem 5.3 and its alternative proof via Theorem 5.6 rely on Lemma 5.1, this lemma needs a complete proof.

minor comments (4)

[Introduction, after Theorem A] The phrase 'for 1≤1≤d(α)' should read 'for 1≤k≤d(α)'.
[Introduction] The text contains a duplicated word: 'we are are able' should read 'we are able'.
[Section 6.2, equation (12)] The notation colim_{[K]∈P^3−Q} is applied to terms that do not depend on [K] in two corners of the diagram; please clarify whether these are constant diagrams and how the pushout maps are induced.
[Section 4.7] The paper relies on [Aum24] for the connectivity range that identifies holomorphic and continuous derivatives, but the precise theorem used is not stated. Please state the connectivity result in the form needed here, since the equality Θ_k ≃ eΘ_k for k ≤ d(α) is a direct consequence of it.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: polynomiality is derived from rank filtrations and Miller's splitting, and the self-cited connectivity theorem is an independent prior input, not a restatement of the target result.

full rationale

The paper's derivation chain is not circular. N-polynomiality of the holomorphic-map functor is proved in Section 3 by a rank filtration on linear maps, a stratification into strata whose suspension spectra are shown polynomial via Miller's stable splitting (Lemmas 3.4–3.6, Corollary 3.7), and induction over homotopy pushout squares (Section 3.2); the conclusion is not assumed in the proof. The Weiss derivatives for continuous maps (Theorem 4.16) are computed from Malkiewich's section-space tower, Goodwillie calculus, and the comparison of the two calculi (Propositions 4.5–4.8, Lemmas 4.11–4.14), again without fitting or renaming. The only self-referential input is the cited theorem [Aum24], used in Section 4.7 to identify the holomorphic derivative with the continuous derivative for k ≤ d(α). That is a prior theorem about homological connectivity with stated assumptions, not a definitional restatement of the present polynomiality or derivative claims, so it is independent support rather than circular loading. The passage in Section 3.3.2 stating 'we will be content with only indicating the most salient modifications' does flag an omitted, sketched proof of the parameterized Theorem 3.10; if the relative evaluation embedding fails over Pic^α(X), or the stratum pushouts do not adapt, the proof is incomplete, but incompleteness is not circularity. No equation is used to define its own conclusion, no fitted parameter is renamed as a prediction, and no uniqueness or ansatz is imported from the authors to force the choice.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No numerical constants are fitted to data; the integers M, N, and d(α) are structurally determined by the hypotheses. The derivative spectra Θ_k and tower layers are outputs of Weiss calculus, not postulated entities, so no invented entities are introduced. The main unproved inputs are standard calculus theorems plus the author's prior connectivity result.

assumptions (7)

standard math Weiss's unitary calculus foundations, including polynomiality, homogeneous functors, and the classification of layers
Used throughout Section 2; citations [Wei95], [Wei98], [Tag22].
standard math Malkiewich's tower for section spaces and its layer formula
External result [Mal15, Theorem 3.16 and Corollary 4.2] used in Section 4.2 for the continuous section functor.
standard math Goodwillie calculus: classification of homogeneous functors and analyticity of section spaces
Used in Sections 4.3-4.5; citations [Goo91], [Goo03], [Chi21].
standard math Miller's stable splitting of Stiefel manifolds and its unitary-calculus derivative interpretation
Load-bearing in the induction proving polynomiality (Lemmas 3.4, 3.5, 6.3); citations [Mil85], [Aro00].
domain assumption Homological connectivity of the inclusion Hol_α(X,P) ⊂ Map_α(X,P) with the range * < 2(d(α)+1) dim V + d(α)-2
Author's prior result [Aum24], cited as a black box in Section 4.7 to identify Θ_k with eΘ_k for k ≤ d(α).
standard math Kodaira vanishing and cohomology-and-base-change ensure that p_∗P is a vector bundle
Algebro-geometric input in the proof of Theorem 3.10, Section 3.3.2; citation [Kle05].
domain assumption The rank stratification on the parameterized space Ψ(V) yields Whitney strata with controlled tubular neighborhoods
Used for the Cech diagram computing the top derivative Θ_N in Section 6.3; based on [NTT14], [Mat70].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Weiss derivatives of holomorphic maps." pith.science (2026). https://pith.science/paper/6AFSKPJH

@misc{pith2026250503731,
  author       = {Pith},
  title        = {Pith review of: Weiss derivatives of holomorphic maps},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6AFSKPJH}},
  note         = {Machine review of arXiv:2505.03731}
}

abstract

We propose an orthogonal approach to the stable homotopy type of spaces of holomorphic maps to projective space. We study the Weiss towers of the unitary functors of holomorphic and continuous maps to $\mathbb{P}(V)$, and show that the former is polynomial and completely compute the latter. As an application we give a new proof of a stable splitting of Cohen--Cohen--Mann--Milgram.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

1 extracted references · 1 linked inside Pith

[1]

Goodwillie calculus,

[AC20] G. Arone and M. Ching, “Goodwillie calculus, ” inHandbook of Homotopy Theory, Chapman and Hall/CRC, Jan. 2020, pp. 3–40 (cit. on p. 15). [AF20] D. Ayala and J. Francis, “A factorization homology primer, ” inHandbook of Homotopy Theory, Chapman and Hall/CRC, Jan. 2020, pp. 41–104 (cit. on p. 15). [Aro00] G. Arone, “The Mitchell-Richter filtration of...

arXiv 2024

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Goodwillie calculus,

[AC20] G. Arone and M. Ching, “Goodwillie calculus, ” inHandbook of Homotopy Theory, Chapman and Hall/CRC, Jan. 2020, pp. 3–40 (cit. on p. 15). [AF20] D. Ayala and J. Francis, “A factorization homology primer, ” inHandbook of Homotopy Theory, Chapman and Hall/CRC, Jan. 2020, pp. 41–104 (cit. on p. 15). [Aro00] G. Arone, “The Mitchell-Richter filtration of...

arXiv 2024