Contractions of subcurves of families of log curves

Sebastian Bozlee

REVIEW 1 major objections 5 minor 15 references

Contractions of subcurves of families of log curves

T0 review · 1 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Every mesa curve admits a contraction of its supported subcurve that is compatible with base change and yields controlled singularities, including elliptic Gorenstein singularities.

desk verdict A genuinely new base-change-compatible contraction for log curves, built on a clearly stated acyclicity hypothesis that also marks the honest boundary of the method. read the letter →

arxiv 1908.09733 v3 pith:OUS57U4Q submitted 2019-08-26 math.AG

Sebastian Bozlee This is my paper · ORCID

classification math.AG MSC 14H1014H20

keywords logarithmicgeometrycontractionmapsmoduliofcurvesmesaellipticGorensteinsingularitiestropicalbasechangenodal

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 constructs a general contraction map for families of log curves: given a "mesa curve" (a log curve with a piecewise linear function $\lambda$ on its tropicalization, the dual graph with edge lengths, defining a subcurve $E$), it collapses $E$ in each fiber to a singularity, preserving arithmetic genus, and does so in a way that commutes with arbitrary base change. Earlier contraction constructions relied on choosing a line bundle and taking the projective spectrum of its pushforwards, which does not commute with base change; this paper instead explicitly writes down the structure sheaf of the contracted curve, gaining base-change compatibility and control over the singularities. The resulting singularities include the elliptic Gorenstein singularities (for steep genus-one mesas), and the construction generalizes the centrally aligned curves used in genus-one moduli work. The load-bearing hypothesis is an acyclicity condition $H^1(E, O_E(-\lambda)) = 0$.

What carries the argument

The central object is the mesa curve: a proper log curve $\pi: C \to S$ over a fine-and-saturated (fs) log scheme together with a section $\lambda$ of the characteristic sheaf (the quotient of the log structure by units) whose associated piecewise linear function on the tropicalization has support $E = |\lambda|$ of positive genus, is constant on the core, has slopes $0$ or $-1$ on paths from the core to the complement, and satisfies the acyclicity condition $H^1(E, O_E(-\lambda)) = 0$. The key mechanism is the B-ring $B(U) = \Gamma(U, O_C(-\lambda)) \oplus \Gamma(S, O_S)$ with multiplication $(f,c)\cdot(g,d) = (\lambda(fg) + df + cg, cd)$, quotiented by the ideal generated by the image of $\Gamma(S, O_S(-\rho)) \to B(U)$; this ring is the structure sheaf of the contracted neighborhood, and its flatness and base-change behavior are controlled by a vanishing theorem for higher cohomology of $O_C(-\lambda)$ (Theorem 4.16).

What would settle it

Take the genus-two Gorenstein singularity models discussed in the paper (the topological semistable models that fail the acyclicity condition), compute $H^1(E, O_E(-\lambda))$ for the corresponding $\lambda$, and check whether the explicitly defined B-ring is flat over the base or whether the contraction maps fail to commute with specialization; a nonvanishing $H^1$ that yields non-flatness would show the acyclicity assumption cannot be dropped.

Watch

Extended reading notes

Core claim

The main theorem (Theorem 1.1) states that for any fine-and-saturated (fs) log scheme $S$ and mesa curve $(\pi: C \to S, \lambda)$, there exists a contraction $\tau: C \to \underline{C}$ of $E = |\lambda|$ inside $C$ such that each connected component of $E$ that is the support of a steep mesa of genus one contracts to an elliptic Gorenstein singularity, and for every morphism $T \to S$ the contracted curve pulls back to the contraction of the pullback curve, naturally. The contraction is constructed locally by defining the ring of functions on a neighborhood of the contracted point as $B(U) = \Gamma(U, O_C(-\lambda)) \oplus \Gamma(S, O_S)$, modulo the image of $\Gamma(S, O_S(-\rho))$, with an explicit multiplication; this ring is shown to be flat and its formation to commute with base change, and the contracted curve is obtained as a pushout. The proof verifies the flatness and base-change properties via vanishing of higher cohomology of $O_C(-\lambda)$, which follows from the acyclicity condition, and then gives an explicit description of the local ring near the singularity: functions on the normalization $Z$ whose boundary values $[f(p_i)]$ lie in a codimension-$g$ subspace $V$ of $k^m$ determined by a Mittag-Leffler problem on $E$.

Load-bearing premise

The construction goes through only when the acyclicity condition $H^1(E, O_E(-\lambda)) = 0$ holds; if that cohomology group is nonzero for some input, the flatness and base-change compatibility of the contracted family are no longer guaranteed.

Editorial extensions

If this is right

If the theorem is right, every mesa curve yields a flat family of singular curves with reduced geometric fibers, and the contraction is compatible with arbitrary base change.
The construction induces morphisms between moduli spaces of curves by contracting the universal curve of a logarithmic moduli space.
Steep genus-one mesas contract specifically to elliptic Gorenstein singularities, so cusps, tacnodes, and transverse unions of them appear as allowed singularities in the target moduli.
The explicit description of the local ring near the contracted point turns the singularity type into a linear condition on boundary values, making the resulting curves amenable to further modular classification.

Reading between the lines

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

Because the acyclicity condition excludes the genus-two Gorenstein models discussed in the paper, a natural next step is to seek a modified B-ring whose flatness does not require $H^1(E, O_E(-\lambda)) = 0$; if such a ring exists, the construction might extend to those singularities.
The local-ring description in terms of a codimension-$g$ subspace $V$ suggests the contraction mechanism is really a Mittag-Leffler interpolation condition; this may transfer to other moduli problems where contractions are defined by linear conditions on boundary values, such as weighted stable maps.
Base-change compatibility means the contracted image of an individual curve can be computed directly from the fiber mesa structure, without first building a smoothing family; this could simplify practical checks in moduli computations.

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

You should know two things about this paper before choosing whether to spend time on it. First, the main theorem is real: for a family of log curves equipped with a 'mesa' section λ, it constructs a contraction of the subcurve E=|λ| that commutes with arbitrary base change, and it gives explicit control over the resulting singularities. Second, the paper is honest about what it does not do: the construction needs H^1(E,O_E(−λ))=0, and Section 1.1 concedes that this fails for the topological models of Battistella's genus-two Gorenstein singularities. So the scope is narrower than the introduction's 'any genus' phrasing might suggest, but that is a boundary of the method, not a hidden flaw.

What is actually new is the mesa curve formalism and the direct construction of the contracted curve from a structure sheaf that is visibly compatible with base change. The prior contractions of RSW, Parker, and HH09 are recovered as special cases, and the explicit ring description near the singularity (Proposition 4.28) is useful and convincing. The proof is long but the architecture is sound: Proposition 4.18 turns acyclicity into the higher cohomology vanishing that drives flatness and base change, and Proposition 4.19 assembles those ingredients cleanly. The examples in 4.6—cusps glued transversally, tacnodes, and the variation of log structure in the tacnode family—are concrete and help the reader see what the abstract construction does.

The soft spots are proportionate. Most importantly, the acyclicity assumption is load-bearing: without it, Proposition 4.18 collapses, and with it, the genus-two Gorenstein contractions of Battistella are out of reach. The author says this explicitly, so a referee should not treat it as a hidden assumption, but the introduction should probably lead with the boundary rather than the promise. The omitted proof of Lemma 4.15 is a minor but real gap; it is a standard truncation of a complex of flat modules, and the author should supply it in revision. There is no circularity and no parameter fitting; the cited special cases are used as comparisons, not as premises.

Who is this for? Anyone working on modular compactifications of M_g,n, log geometry, or contraction constructions for families of curves. It deserves a serious referee: the central argument holds up, the contribution is new, and the limitations are stated in the paper itself. Send it to review; ask for the Lemma 4.15 proof and a slightly more candid introduction.

Referee Report

1 major / 5 minor

Summary. The paper introduces the notion of a mesa curve—a log curve equipped with a section λ of the characteristic sheaf whose associated piecewise linear function has a 'mesa' shape—and constructs, for each family of mesa curves, a contraction of the support E=|λ|. The contraction is built by explicitly defining a structure sheaf: locally one takes Spec of a ring B(U) built from Γ(U,O_C(−λ)) and Γ(S,O_S), then glues it to C−E via a pushout. The main theorem (Theorem 1.1) asserts that this contraction is a family of curves, commutes with arbitrary base change, and that steep genus-one components are contracted to elliptic Gorenstein singularities. The proof is divided into reduction to a standard situation, vanishing and flatness via a cohomology-and-base-change argument, well-definedness and independence of auxiliary sections, and an explicit description of the singularity in §4.6. Section 1.1 explicitly notes that the construction does not cover Battistella's genus-two Gorenstein models because they fail the acyclicity condition H^1(E,O_E(−λ))=0.

Significance. If correct, the theorem provides a base-change-compatible contraction for a broad class of subcurves of log curves with explicit control of the resulting singularities, generalizing the genus-one constructions of Ranganathan–Santos-Parker–Wise and Parker and giving a potential route to modular compactifications of M_{g,n} and to desingularizations of stable maps. The paper's method—constructing the structure sheaf directly rather than via a relative Proj—is a useful contribution, and the singularity analysis in §4.6 is concrete and informative. The main result is conditional on the mesa axioms, and the author is transparent about the acyclicity boundary. The proof is long but mostly careful; however, the validity of one of its cornerstones, Lemma 4.15, is doubtful and needs repair before the flatness and base-change claims can be accepted.

major comments (1)

[§4.4, Lemma 4.15 and Theorem 4.16] Lemma 4.15 is stated without proof and is false as stated. Take A=k[ε]/(ε^2) and the cochain complex C• with C^i=A for i≥0 and differentials all equal to multiplication by ε. Each C^i is flat, and H^0(C•)=εA≅k is not flat over A, while H^i(C•)=0 for all i≥1; thus the hypotheses hold with j=0. The usual smart truncation τ≤0C• has degree-0 term ker d^0=εA, which is not flat, and the naive truncation is not quasi-isomorphic to C•. Consequently the assertion in Lemma 4.15 that τ≤jC• is a complex of flat A-modules fails, and the proof of Theorem 4.16, which uses Lemma 4.15 to obtain both base-change compatibility and flatness of H^0, is invalid. Since Theorem 4.16 is used in Proposition 4.18 and again in Proposition 4.19 to establish exactness, flatness, and base change for the B-sequence, this gap is load-bearing for Theorem 1.1. The intended results may be true—they are standard consequences of the cohomology-and-base-change theorem—but the paper needs either a correct replacement lemma with proof or a direct citation of the standard theorem.

minor comments (5)

[Abstract and §1.1] The abstract describes contractions for curves 'of any genus' without mentioning the acyclicity hypothesis H^1(E,O_E(−λ))=0 that is built into Definition 3.2(vii). Because §1.1 explicitly states that this condition fails for the genus-two Gorenstein models of [Bat19], the abstract should state that the theorem concerns acyclic mesa curves, so that readers are not led to expect those contractions.
[§4.2, proof of Proposition 4.23] The sentence 'Since π:E→S is proper and U→S is affine (hence separated)' is inaccurate: the open subscheme U=C∖∪σ_i(S) is not necessarily affine over S. Only separatedness of U→S is used, and it holds because C/S is separated; the wording should be corrected.
[§4.4, proof of Proposition 4.19(i)] The sentence 'We always have (iii) for flat maps T→S' is confusing, since (iii) is part of the proposition being proved. Please clarify that a flat-base-change version of (iii) follows from Proposition 4.18 and Theorem 4.16 and can be invoked at that point.
[Theorem 1.1(ii)] In Theorem 1.1(ii) the two schemes C×_S T and C×_S T are typeset identically, making the statement hard to parse. Please use distinct notation for the contracted curve in the displayed isomorphism.
[Introduction and references] The name 'Battistella and Carrocci' appears in the Introduction while the reference list has 'Battistella and Carocci'; the spelling should be made consistent.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the contraction is constructed explicitly from the mesa axioms; acyclicity is a stated hypothesis, not a hidden consequence.

full rationale

The derivation is self-contained. Theorem 1.1 is proved by an explicit construction of the contracted structure sheaf B(U) in Definition 4.3, followed by independent verification of flatness (Proposition 4.19), base-change compatibility (Proposition 4.19(iii), Corollary 4.21), isomorphism away from E (Proposition 4.23), independence of auxiliary sections (Proposition 4.25), the genus and elliptic Gorenstein properties of the resulting singularity (Proposition 4.28, using Smyth's external criterion Lemma 4.27), and properness (Section 4.7). The acyclicity hypothesis H^1(E, O_E(-lambda)) = 0 is an input axiom in Definition 3.2(vii), not a consequence of the contraction; it is used transparently in Proposition 4.18 to obtain higher cohomology vanishing, and Section 1.1 explicitly identifies it as the boundary that excludes Battistella's genus-two Gorenstein models. That is a scope limitation, not circularity. No fitted parameter is renamed as a prediction, and the cited works [RSW17], [Par17], [Smy11], and Kato's log-curve structure theorem are external background or comparison results rather than unexamined substitutes for the proof. The central claim therefore has independent content.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

No free parameters are fitted to data. The paper's central claim is conditional on the apparatus of log geometry, standard cohomology-and-base-change facts, and the newly introduced mesa axioms, most notably acyclicity H^1(E, O_E(-lambda)) = 0. The mesa is a defined object, not an unexplained entity. The theorem is an existence and construction result, so no empirical constants are used.

assumptions (5)

domain assumption Kato's structure theorem for log curves (Theorem 2.3): the underlying family of a log curve is nodal and admits the stated local charts.
Used throughout to describe the germ of smooth points, marked points, and nodes; taken from [Kat00].
ad hoc to paper Mesa axioms: support conditions, slopes 0 or -1, constancy on the core, non-negative degree on rational components, and acyclicity H^1(E, O_E(-lambda)) = 0.
These conditions define the class of curves for which the contraction is constructed; the acyclicity condition is the one that excludes genus-two Gorenstein contractions.
standard math Standard cohomology and base change formalism for proper morphisms, including the truncated complex criterion in Lemma 4.15.
Required to prove flatness and that the contraction commutes with pullback; Lemma 4.15 is stated without proof.
standard math Stein factorization and properness criteria used in Lemma 5.7 and Proposition 4.33.
Used to prove connectivity of geometric fibers and properness of the contracted family.
standard math Serre duality and evaluation exact sequences on nodal curves used in Section 4.3, Lemmas 4.10 through 4.13.
Establishes the codimension-g value constraints and the genus-one linear condition needed to describe the singularities.

invented entities (1)

mesa curve, consisting of a log curve and a mesa section lambda
purpose: Encodes the subcurve E to be contracted and provides the contraction data through the associated line bundle O_C(-lambda).
Introduced as a definition in Definitions 3.2 and 3.3. It is a mathematical object with explicit axioms, not a postulated physical entity; its usefulness is established by Theorem 1.1.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Contractions of subcurves of families of log curves." pith.science (2026). https://pith.science/paper/OUS57U4Q

@misc{pith2026190809733,
  author       = {Pith},
  title        = {Pith review of: Contractions of subcurves of families of log curves},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OUS57U4Q}},
  note         = {Machine review of arXiv:1908.09733}
}

abstract

Let $C$ be a nodal curve, and let $E$ be a union of semistable subcurves of $C$. We consider the problem of contracting the connected components of $E$ to singularities in a way that preserves the genus of $C$ and makes sense in families, so that this contraction may induce maps between moduli spaces of curves. In order to do this, we introduce the notion of mesa curve, a nodal curve $C$ with a logarithmic structure and a piecewise linear function $\overline{\lambda}$ on the tropicalization of $C$. This piecewise linear function determines a subcurve $E$. We then construct a contraction of $E$ inside of $C$ for families of mesa curves. Resulting singularities include the elliptic Gorenstein singularities.

Figures

Figures reproduced from arXiv: 1908.09733 by the authors.

**Figure 1.** A log curve C, a section λ of its characteristic sheaf, and the support |λ| of that section. dual graph together with these lengths the tropicalization of C. The datum λ is identified with a piecewise linear function on the tropicalization of C, see [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

**Figure 2.** A typical log curve and its tropicalization. We will usually use subscript notation for lengths, i.e. δe. See [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗

**Figure 3.** A log curve C, a section λ of its characteristic sheaf, and the support |λ| of that section. For example, in [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 14 canonical work pages

[1]

M. F. Atiyah and I. G. MacDonald, Introduction to commutative algebra, Westview Press, 1969

work page 1969
[2]

existence of modifications, Annals of Mathematics 91 (1970), no

Michael Artin, Algebraization of formal moduli: Ii. existence of modifications, Annals of Mathematics 91 (1970), no. 1, 88--135

work page 1970
[3]

Luca Battistella , Modular compactifications of M_ 2,n with Gorenstein singularities , arXiv e-prints (2019), arXiv:1906.06367

work page Pith review arXiv 2019
[4]

Renzo Cavalieri , Melody Chan , Martin Ulirsch , and Jonathan Wise , A moduli stack of tropical curves , arXiv e-prints (2017), arXiv:1704.03806

work page Pith review arXiv 2017
[5]

Brendon Hassett, Moduli spaces of weighted pointed stable curves, Advances in Mathematics 173 (2003), 316--352

work page 2003
[6]

8, 4471--4489

Brendon Hassett and Donghoon Hyeon, Log canonical models for the moduli space of curves: first divisorial contraction, Transactions of the American Mathematical Society 361 (2009), no. 8, 4471--4489

work page 2009
[7]

Kazuya Kato, Logarithmic structures of F ontaine- I llusie , Algebraic Analysis, Geometry, and Number Theory (1989), 191--224

work page 1989
[8]

2, 215--232

Fumiharu Kato, Log smooth deformation and moduli of log smooth curves, International Journal of Mathematics 11 (2000), no. 2, 215--232

work page 2000

Show all 15 references

[9]

3-4, 1327--1363

Borne Niels and Angelo Vistoli, Parabolic sheaves on logarithmic schemes, Advances in Mathematics 231 (2012), no. 3-4, 1327--1363

2012
[10]

Arthur Ogus, Lectures on logarithmic algebraic geometry, Cambridge University Press, 2018

2018
[11]

thesis, University of Colorado, 2017

Keli Parker, Semistable modular compactifications of moduli spaces of genus one curves, Ph.D. thesis, University of Colorado, 2017

2017
[12]

Dhruv Ranganathan , Keli Santos-Parker , and Jonathan Wise , Moduli of stable maps in genus one and logarithmic geometry I , arXiv e-prints (2017), arXiv:1708.02359

2017 arXiv
[13]

3, 877--913

David Smyth, Modular compactifications of the space of pointed elliptic curves I , Compositio Mathematica 147 (2011), no. 3, 877--913

2011
[14]

2, 459--503

, Towards a classification of modular compactifications of M _ g,n , Inventiones Mathematicae 192 (2013), no. 2, 459--503

2013
[15]

Ravi Vakil and Aleksey Zinger, A desingularization of the main component of the moduli space of genus-one stable maps into P ^n , Geometry & Topology 12 (2008), 1--95

2008

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

M. F. Atiyah and I. G. MacDonald, Introduction to commutative algebra, Westview Press, 1969

work page 1969

[2] [2]

existence of modifications, Annals of Mathematics 91 (1970), no

Michael Artin, Algebraization of formal moduli: Ii. existence of modifications, Annals of Mathematics 91 (1970), no. 1, 88--135

work page 1970

[3] [3]

Luca Battistella , Modular compactifications of M_ 2,n with Gorenstein singularities , arXiv e-prints (2019), arXiv:1906.06367

work page Pith review arXiv 2019

[4] [4]

Renzo Cavalieri , Melody Chan , Martin Ulirsch , and Jonathan Wise , A moduli stack of tropical curves , arXiv e-prints (2017), arXiv:1704.03806

work page Pith review arXiv 2017

[5] [5]

Brendon Hassett, Moduli spaces of weighted pointed stable curves, Advances in Mathematics 173 (2003), 316--352

work page 2003

[6] [6]

8, 4471--4489

Brendon Hassett and Donghoon Hyeon, Log canonical models for the moduli space of curves: first divisorial contraction, Transactions of the American Mathematical Society 361 (2009), no. 8, 4471--4489

work page 2009

[7] [7]

Kazuya Kato, Logarithmic structures of F ontaine- I llusie , Algebraic Analysis, Geometry, and Number Theory (1989), 191--224

work page 1989

[8] [8]

2, 215--232

Fumiharu Kato, Log smooth deformation and moduli of log smooth curves, International Journal of Mathematics 11 (2000), no. 2, 215--232

work page 2000