Pith. sign in

REVIEW 1 major objections 5 minor 3 references

Sheaf homology of hyperplane arrangements, Boolean covers and exterior powers

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 For an essential hyperplane arrangement, the sheaf homology with coefficients in any exterior power of the natural sheaf is supported only on the diagonal i+j = rk L − 1, with dimensions given by derivatives of the characteristic…

desk verdict A solid computation of exterior-power sheaf homology for arrangement lattices, with a fixable gap in the deletion-restriction proof that a referee should ask to be filled. read the letter →

arxiv 1908.04500 v2 pith:QBO5LGKI submitted 2019-08-13 math.AT math.COmath.GR

Brent Everitt , Paul Turner This is my paper · ORCID
classification math.ATmath.COmath.GR MSC 05E4552C3555N30
keywords sheafhomologyhyperplanearrangementsintersectionlatticeBooleancoverexteriorpowerscharacteristicpolynomialcellulardeletion-restriction
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 computes the full sheaf homology of the intersection lattice of a hyperplane arrangement when the coefficient sheaf is any exterior power of the natural sheaf. The result is that the bi-graded homology H_i(L\setminus 0;\Lambda^j F) vanishes except on the diagonal i+j = \operatorname{rk}L-1, plus a few low-degree exceptions, and on that diagonal its dimension is a derivative of the characteristic polynomial. This turns the coefficients of the characteristic polynomial into homology groups, a categorification the paper notes but does not pursue. The argument avoids the lattice itself and works on a Boolean cover, where the homology becomes cellular and is governed by a deletion–restriction long exact sequence.

What carries the argument

The Boolean cover of L is the lattice of all subsets of the atoms of L, with the canonical map sending a subset to its join in L; sheaf homology of L is isomorphic to sheaf homology of this cover, and on Boolean lattices sheaf homology is isomorphic to a cellular homology computed from the rank filtration. The paper's main technical tool is a deletion–restriction long exact sequence for this cellular homology, which mirrors the classical deletion–restriction relation for characteristic polynomials and drives the induction. Also essential is a splitting theorem showing that the exterior-power sheaf Λ^jF on the Boolean cover of an essential arrangement is decomposable, so its cellular homology can be reduced to that of a restriction lattice.

What would settle it

For the essential rank-2 braid arrangement (three lines through the origin in the plane), the paper predicts dim H_cell_1(~L;$Λ^{1}$F) = 1 and dim H_cell_0(~L;$Λ^{2}$F) = 1, with all other cellular homology groups zero. Writing down the cellular complex of the Boolean cover on three atoms and direct computation of the exterior-power coefficient maps would confirm or refute the theorem in this case.

Watch

Extended reading notes

Core claim

The central discovery is that for an essential hyperplane arrangement with intersection lattice L, the sheaf homology groups H_i(L\setminus 0;\Lambda^j F) are supported exactly where i+j = \operatorname{rk}L - 1 (with specified H_0 exceptions), and in that range $$ \dim H_i(L\setminus 0;\Lambda^j F) = \frac{(-1)^{i+1}}{j!}\$chi_L^{{(j)}}$(1), $$ where $\chi_L^{(j)}$ is the $j$-th derivative of the characteristic polynomial. The proof first establishes the analogous statement for the cellular homology of the Boolean cover, using induction and a newly derived deletion–restriction long exact sequence, then transfers the result back to sheaf homology. For non-essential arrangements, the same machinery yields explicit formulas in terms of the essentialisation of the natural sheaf and the dimension of the common intersection of all hyperplanes.

Load-bearing premise

The induction assumes that deleting atoms one at a time in the Boolean cover, after identified atoms have been collapsed, does not change the cellular homology; the paper asserts this reduction rather than proving each deletion step in detail.

Editorial extensions

If this is right

  • The bi-graded homology H_*(L\setminus 0;\Lambda^\bullet F) is completely determined by the characteristic polynomial and its derivatives, so no other invariants of the arrangement enter.
  • The graded Euler characteristic identity χ_q H_*(L\setminus 0;\Lambda^\bullet F) = -χ_L(1+q) + (1+q)^{\dim V} recovers the characteristic polynomial from the homology, making the homology a categorification of χ_L.
  • For non-essential arrangements, the homology is described through the essentialisation of the natural sheaf: the support lies in the strip rk L ≤ i+j ≤ dim V, with explicit binomial coefficients involving the dimension of the common intersection.
  • The j=1 case reproduces the previously known natural-sheaf homology, so the result contains and extends the earlier computation as a special case.
  • The cellular homology of the Boolean cover is computed first and has a single non-zero diagonal i+j = rk L = dim V, which shifts to the diagonal i+j = rk L - 1 for sheaf homology.

Reading between the lines

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

  • A natural extension not pursued in the paper is to study the multiplicative structure: the exterior-algebra product on Λ^\bullet F likely gives H_*(L\setminus 0;\Lambda^\bullet F) the structure of a module over an exterior algebra, which could distinguish arrangements that share the same characteristic polynomial.
  • The deletion–restriction long exact sequence for Boolean covers is a general machine that could apply to other coefficient sheaves such as symmetric powers or tensor powers, where diagonal support may fail but the recursion would still constrain the answer.
  • The proofs work over a field, so an immediate testable extension is to compute integral homology with exterior-power coefficients; the explicit cellular complex of the Boolean cover could reveal torsion not visible in the field 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.

Referee Report

1 major / 5 minor

Summary. The paper computes the sheaf homology of the intersection lattice L of a hyperplane arrangement with coefficients in the graded exterior sheaf Λ^•F of the natural sheaf F. The main technical contribution is a machinery of Boolean covers: for a graded atomic lattice L, the Boolean cover ~L is the Boolean lattice on the atoms of L, and the paper shows that the sheaf homology H_*(L\0;F) is isomorphic to a cellular homology H_cell_*(~L;F) (Theorems 3 and 4, plus Corollary 2). The core new tool is a deletion-restriction long exact sequence for cellular homology of Boolean covers (Theorem 7), obtained from a short exact sequence of cellular chain complexes and an identification, proved in Theorem 6, of the cellular homology of a sub-Boolean with that of the Boolean cover of the restriction. This sequence is then used in an induction on the number of hyperplanes to prove Theorem 8, which gives the complete cellular homology of ~L with exterior-power coefficients for essential arrangements; Theorem 9 converts this to the desired sheaf homology. Theorems 10 and 11 extend the computation to non-essential arrangements by an essentialisation argument.

Significance. If the main results are correct, the paper provides an explicit, closed-form description of the full bi-graded sheaf homology of arrangement lattices with exterior-power coefficients, a genuine extension of the j=1 result of the authors' earlier paper and of Lusztig's theorem. The dimension formulas are expressed directly in terms of derivatives of the characteristic polynomial, so the paper gives a concrete categorification of χ_L(1+q) in the sense of graded Euler characteristics. The paper also introduces a useful set of tools — Boolean covers, cellular homology of Boolean lattices, splitting/doubling lemmas, and the deletion-restriction long exact sequence — that are likely to be of independent interest. The exposition is generally clear, the Euler-characteristic computations are parameter-free, and the claimed formulas are precise and falsifiable. The main concern is a gap in the proof of Theorem 6, on which the deletion-restriction sequence and the entire induction rest.

major comments (1)
  1. [Section 3.6, proof of Theorem 6] The reduction 'We may now repeat this process by taking a sequence of deletions of B until we arrive at ~L^a' is asserted rather than proved. The proof establishes that a single deletion of an atom s identified with another atom s' under the Boolean-cover map does not change cellular homology, by showing that the corresponding restriction sub-Boolean is a double and applying Proposition 5 together with the short exact sequence of chain complexes. To iterate this, one must verify that after deleting s, the remaining atoms of B_s that are identified in pairs under the map to L^a still satisfy the same double condition with respect to the appropriate new atom α = u∨u', and that the sheaf on the final Boolean lattice obtained by choosing one representative from each fiber is isomorphic to the induced sheaf on the Boolean cover of L^a. This is a formal induction on the number of surplus atoms; it is not supplied. Since Theorem 7 (the deletion-restriction long exact sequence) and hence the inductive proof of Theorems 8 and 9 depend on this identification, the main computation is conditional on filling this gap. Please add the missing induction, or give an alternative direct argument that the iterated deletions preserve cellular homology and terminate at ~L^a.
minor comments (5)
  1. [Section 4.2, proof of Theorem 8, base case |A|=3] The base case |A|=3 is not fully demonstrated: for j=1 the proof says only 'Two applications of Theorem 5 give the required result', and for j=2 the verification is a single sentence. Since this case anchors the induction on |A|, please include the explicit chain-level computation or a small table of the cellular chain groups for the braid arrangement.
  2. [Section 3.6] The notation for deletion and restriction sub-Booleans is easy to confuse: in the proof of Theorem 6 the same symbol B_s appears to be used for both the deletion (the sub-Boolean not containing s) and the restriction (the interval [s,1]) in different sentences. Please use distinct, consistently defined symbols such as B_s^del and B_s^res, or define them explicitly at the start of the proof.
  3. [Proposition 6] The title reads 'The graded Euler chracteristic' — 'chracteristic' should be 'characteristic'.
  4. [Abstract and Theorem 9] In the statement of Theorem 9 (and the abstract), the phrase 'or, i = 0 and or j = rkL− 1' contains a repeated 'or'; please correct to 'or i = 0 and j = rkL− 1'.
  5. [Section 4.3] In the proof of Theorem 10, the phrase 'we have that dim H_cell_i(~L; Λ^tF⊥) = 0 unless t = rkL− i' is stated for F⊥ as an essential sheaf; it may help the reader to remind them that Theorem 8 applies because the essentialisation has the same arrangement lattice L with rank rkL and is essential in U⊥V.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the homology dimensions are derived from vanishing arguments plus an independent Euler-characteristic identity, not from the target result.

full rationale

The paper's central computation is not circular. The dimension formulas in Theorems 8–11 are obtained in two stages: first, vanishing is proved by induction using the deletion-restriction long exact sequence (Theorem 7), whose proof is based on the short exact sequence (4), Proposition 5, and the identification in Theorem 6; second, the single non-vanishing group in each bidegree is computed from the graded Euler characteristic identity χ_q H_cell(∼L; Λ•F) = χ_L(1+q) (Proposition 6). Proposition 6 itself is a direct calculation from Corollary 4, which follows from the Euler characteristic of the cellular chain complex and the Möbius-function definition of the characteristic polynomial. The characteristic polynomial is an input invariant of the arrangement, not a fitted parameter, and the Euler characteristic relation does not presuppose the vanishing or the dimension formulas being proved. The reliance on the authors' earlier work [ET] and [ET15] is technical and not load-bearing: [ET15] supplies the cellular-vs-sheaf comparison adapted in Theorem 4, and the j=1 case of the final theorem is a special case of the new computation rather than an assumed input. The proof of Theorem 6 contains an asserted repetition step ('We may now repeat this process by taking a sequence of deletions of B until we arrive at ∼L^a') that is not fully formalized; this is a correctness gap or incompleteness in the written proof, but it is not circularity, because no equation is being used as its own conclusion and no predicted quantity is built into the input by construction. Overall, no step reduces the claimed derivation to its own inputs.

Assumptions & free parameters 0 free parameters · 6 assumptions · 1 invented entities

The paper introduces no free parameters: the characteristic polynomial of the arrangement is an input invariant, and no constants are fitted to data. The main background assumptions are standard tools of homological algebra (derived functors, spectral sequences, universal coefficient theorem) and standard facts about hyperplane arrangements and geometric lattices. The Boolean cover is a new mathematical construction introduced as a tool, not an empirically testable entity.

assumptions (6)
  • standard math Higher colimits compute sheaf homology of posets: H_*(P;F) = L_* lim_P F.
    Used throughout; definition in Section 2.1 with reference to [GZ67, Appendix II].
  • standard math Leray-Serre spectral sequence for poset maps (Theorem 1) and its nerve-covering version (Theorem 2).
    Adapted from [GZ67, Appendix II, Theorem 3.6] and the authors' [ET, Section 2.3]; used to prove Theorem 3.
  • standard math Cellular cohomology of posets from [ET15, Theorem 2] adapts to give cellular homology of Boolean lattices computing sheaf homology.
    Used in Theorem 4; the proof is adapted from [ET15].
  • domain assumption Intersection lattices of hyperplane arrangements are geometric lattices with atoms the hyperplanes.
    Section 1.1; this is the setting of all main theorems.
  • domain assumption In the non-essential case, working over a subfield of C, the orthogonal decomposition V = U ⊕ U⊥ and the identity U⊥(B∩C) = U⊥B∩U⊥C hold.
    Section 4.3, Lemma 2; needed for the essentialisation decomposition F=ΔU⊕F⊥.
  • standard math Lusztig's theorem: for a graded atomic lattice, H_*(L\0;F) is isomorphic to H_*(~L\0;F).
    Theorem 3, stated as 'also appears in [Lus74, §1.2]' and proved in Section 2.4 via the covering spectral sequence.
invented entities (1)
  • Boolean cover ~L of a graded atomic lattice L
    purpose: A lattice on the subsets of the atoms of L with a map to L, used to compute sheaf homology cellularly and to obtain the deletion-restriction long exact sequence.
    Introduced in Section 1.4 as the main computational tool. It is a mathematical construction with proven properties (Theorems 3 and 7); there is no empirically testable handle outside the paper, but it is not a physically postulated entity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sheaf homology of hyperplane arrangements, Boolean covers and exterior powers." pith.science (2026). https://pith.science/paper/QBO5LGKI

@misc{pith2026190804500,
  author       = {Pith},
  title        = {Pith review of: Sheaf homology of hyperplane arrangements, Boolean covers and exterior powers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QBO5LGKI}},
  note         = {Machine review of arXiv:1908.04500}
}
read the original abstract

We compute the sheaf homology of the intersection lattice of a hyperplane arrangement with coefficients in the graded exterior sheaf of the natural sheaf. This builds on the results of our previous paper, where this homology was computed for the natural sheaf, itself a generalisation of an old result of Lusztig. The computational machinery we develop in this paper is quite different though: sheaf homology is lifted to what we call Boolean covers, where we instead compute homology cellularly. A number of tools are given for the cellular homology of these Boolean covers, including a deletion-restriction long exact sequence.

Figures

Figures reproduced from arXiv: 1908.04500 by the authors.

Figure 1
Figure 1. The arrangement lattices L(A) where |A| ≤ 3. rk(0) = 0. The atoms of P are the elements of rank 1. A poset map f : Q → P is a set map such that f x ≤ f y ∈ P if x ≤ y ∈ Q. A subset K ⊂ P is upper convex if x ∈ K and x ≤ y implies y ∈ K. If x ≤ y, the interval [x, y] consists of those z ∈ P such that x ≤ z ≤ y; if x ∈ P the interval P≥x consists of those z ∈ P such that z ≥ x; one defines P≤x, P>x and P<x similarly. … view at source ↗
Figure 2
Figure 2. The decomposition of L into the deletion La and restriction L a for a dependent atom a. In any graded atomic lattice, a set S ⊂ A of atoms is independent if W T < W S for all proper subsets T of S , and dependent otherwise. An atom a in a dependent set of atoms S with the property that W S \ {a} = W S is called a dependent atom. It is well known (see for instance [Bir79, EF13]) that the only graded atomic lattices w… view at source ↗
Figure 3
Figure 3. A double. Doubling: Let F be a sheaf on B for which there is an atom a ∈ A such that for all x ∈ Ba the structure map F x∨a x : F(x ∨ a) → F(x) (6) is the identity. The restrictions of F to Ba and B a are consequently exactly the same sheaf and so we call (B; F) a double (see [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Support of H cell i (eL; Λ jF) for L essential in the (i, j)-plane, with the dimension of the red square given in Theorem 8. Proposition 7. Let L be a Boolean intersection lattice of an essential hyperplane arrangement in a space V, let F be the natural sheaf on L and …
Figure 5
Figure 5. Figure 5: Support of H cell i (eL; Λ jF) in the (i, j)-plane, with the dimension of the red square given by Theorem 10. Applying, Corollary 3 gives H cell i (eL; Λ jF) = M s+t=j H cell i (eL; ∆ΛsU ⊗ Λ tF ⊥ ). Recalling that we are working in characteristic zero, the universal co…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

3 extracted references · 3 canonical work pages

  1. [12]

    Ser., vol

    MR2868112 [Sta07] , An introduction to hyperplane arrangements, Geometric combinatorics, IAS /Park City Math. Ser., vol. 13, Amer. Math. Soc., Providence, RI, 2007, pp. 389–496 . [Y uz91] Sergey Y uzvinsky, Cohomology of local sheaves on arrangement lattices , Proc. Amer. Math. Soc. 112 (1991), no. 4, 1207–1217

    work page 1991

  2. [1979]

    Deletion-restriction for sheaf homology of graded atomic lattices

    [Bj¨ o82] Anders Bj¨ orner,On the homology of geometric lattices , Algebra Universalis 14 (1982), no. 1, 107–128. [DL15] Zsuzsanna Dansco and Anthony Licata, Odd Khovanov homology for hyperplane arrangements , J. Algebra 436 (2015), 102–144. [EF13] Brent Everitt and John Fountain, Partial mirror symmetry, lattice presentations and algebr aic monoids , Pro...

    work page Pith review arXiv 1982

  3. [1992]

    [Qui78] Daniel Quillen, Homotopy properties of the poset of nontrivial p-subgroups of a group , Adv. in Math. 28 (1978), no. 2, 101–128. [Sta12] Richard P . Stanley, Enumerative combinatorics. V olume 1, 2nd ed., Cambridge Studies in Advanced Math- ematics, vol. 49, Cambridge University Press, Cambridge, 2

    work page 1978

Pith tools

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