Cohomology of Burnside Rings

Benen Harrington

REVIEW 2 major objections 4 minor 7 references

Cohomology of Burnside Rings

T0 review · 2 major / 4 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 square-free group orders, Ext between mark modules is 2-periodic; non-square-free orders give unbounded rank.

desk verdict Useful B-ring framework and believable claims, but the central square-free periodicity theorem is stated without proof and the block-decomposition gap in Lemma 7 is a missing one-liner. read the letter →

arxiv 1908.06156 v1 pith:BNLB75HG submitted 2019-08-16 math.RA math.ACmath.GR

Benen Harrington This is my paper · ORCID

classification math.RAmath.ACmath.GR MSC 16E3019A2216E40

keywords BurnsideringmarkhomomorphismExtandTorB-ringghostsquare-freeorderperiodiccohomologymod-pblockdecomposition

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 Burnside ring $A(G)$ organizes the isomorphism classes of finite $G$-sets, and for each subgroup $H$ the mark module $\mathbb{Z}_H$ is the integers with $A(G)$ acting through the number of $H$-fixed points. The paper proves that when $|G|$ is square-free, the homological algebra of these modules is completely determined: for every $H,J$ and every $l \geq 1$, $\mathrm{Ext}^l_{A(G)}(\mathbb{Z}_H,\mathbb{Z}_J) \cong \mathrm{Ext}^{l+2}_{A(G)}(\mathbb{Z}_H,\mathbb{Z}_J)$, and the same two-step periodicity governs $\mathrm{Tor}$. Concretely, each prime $p$ dividing $|G|$ contributes a single copy of $\mathbb{Z}/p\mathbb{Z}$, appearing only in degrees of one fixed parity, and all other $p$-parts vanish; which parity occurs is read from whether $H$ and $J$ form a two-element equivalence class under agreement modulo $p$. The paper also establishes the converse: if $|G|$ is not square-free, some $\mathrm{Ext}$ and $\mathrm{Tor}$ groups have unbounded rank as $l$ grows, so no finite table can describe them. The payoff is a closed formula for an entire family of finite groups where explicit computations are otherwise intractable.

What carries the argument

The argument is carried by B-rings: subrings $R$ of a product of copies of $\mathbb{Z}$ (the ghost ring) with the property that any two distinct coordinates can be separated by an element of $R$ that is nonzero in one coordinate and zero in the other. The Burnside ring embeds as a B-ring through its mark homomorphisms. For a prime $p$, two coordinates $i,j$ are related by $i \sim_p j$ when every element of $R$ takes congruent values at $i$ and $j$ modulo $p$; these classes are the blocks of the mod-$p$ algebra. The main computational tool is the long exact sequence coming from multiplication by $p$ on $0 \to \mathbb{Z}_j \to \mathbb{Z}_j \to k_j \to 0$, which yields recurrences $a_{l+1}=b_l-a_l$ and $z_l=y_{l+1}-z_{l+1}$ on the $p$-ranks of Ext and Tor. When $|G|$ is square-free, each $\sim_p$-class has size at most two, the relevant mod-$p$ block is a two-dimensional local algebra, and the recurrences force the period-2 pattern; conversely, when $p^2$ divides $|G|$, a standard criterion on local rings implies unbounded dimensions in the mod-$p$ algebra, which carry back to unbounded ranks.

What would settle it

Work with $G=S_3$ and the two non-conjugate subgroups of order $2$; compute $\mathrm{Ext}^1$ and $\mathrm{Ext}^3$ between the corresponding mark modules and check whether they are isomorphic as $A(G)$-modules with the predicted $\mathbb{Z}/2\mathbb{Z}$ value. A mismatch in rank or module structure would refute Theorem 20. Alternatively, test Lemma 7 directly on a B-ring in which a $\sim_p$-class has more than one member: if the constructed product does not vanish modulo $p$ on an unselected member of a complementary class, Proposition 8 would need another proof.

Watch

Extended reading notes

Core claim

The central discovery is that the Ext and Tor groups between mark modules of a Burnside ring are 2-periodic exactly when the group order is square-free. Theorem 20 states that for $|G|$ square-free, $\mathrm{Ext}^l_{A(G)}(\mathbb{Z}_H,\mathbb{Z}_J) \cong \mathrm{Ext}^{l+2}_{A(G)}(\mathbb{Z}_H,\mathbb{Z}_J)$ as $A(G)$-modules for all $H,J$ and all $l \geq 1$; the comparison in Corollary 11 then transfers the periodicity to $\mathrm{Tor}$. Corollary 19 sharpens this to a complete computation: the $p$-primary part is $\mathbb{Z}/p\mathbb{Z}$ precisely when the conjugacy classes of $H$ and $J$ form a two-element class under the relation of agreeing modulo $p$, with even-degree self-Ext and odd-degree mixed Ext, and zero in all other degrees. When $|G|$ is not square-free, the paper proves the opposite behavior: there exist $H,J$ for which the groups $\mathrm{Ext}^l_{A(G)}(\mathbb{Z}_H,\mathbb{Z}_J)$ and $\mathrm{Tor}^{A(G)}_l(\mathbb{Z}_H,\mathbb{Z}_J)$ have unbounded rank.

Load-bearing premise

The block decomposition that drives the whole proof assumes that for each prime $p$ and each equivalence class of subgroups agreeing modulo $p$, some element of $A(G)$ is congruent to $1$ mod $p$ on the entire class and to $0$ mod $p$ on every other class; the construction given verifies the zero condition only at one chosen subgroup in each complementary class, not at all of them.

Editorial extensions

If this is right

For every square-free group $G$, all Ext and Tor groups between mark modules are determined by a parity check: the computation reduces to counting two-element $\sim_p$ classes rather than resolving modules.
The $p$-torsion in the square-free case is always exactly $\mathbb{Z}/p\mathbb{Z}$; no $p^2$ or higher torsion appears.
For non-square-free orders, the unbounded-rank result rules out any finite description of the full family, so the square-free case is not just one example but the boundary of tractability.
Because the periodicity is an isomorphism of $A(G)$-modules and not merely of abelian groups, the multiplicative action of the Burnside ring is also periodic, which is additional structure for applications.

Reading between the lines

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

I would conjecture that the period-2 pattern holds for any B-ring whose pairwise differences $d(i,j)$ are all square-free, not only for Burnside rings; the Burnside-ring proof uses the group only to control the size of the $\sim_p$ classes, so the framework is directly testable on other subrings of ghost rings.
If the missing vanishing in Lemma 7 cannot be supplied, the square-free description might still survive for groups in which each $\sim_p$ class is 'generated' by a single subgroup; failures would first appear in groups with larger classes, and one could hunt for them by explicit computation in small B-rings.
The unboundedness half raises a quantitative question the paper leaves open: for a fixed non-square-free $G$, how fast does the rank of $\mathrm{Ext}^l$ grow, and is the growth rate determined by the radical filtration of the mod-$p$ Burnside algebra?

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

Colleague,

The takeaway: the paper gives a clean B-ring framework for Ext and Tor between mark modules, and the main square-free description is probably correct, but the paper does not actually prove its central periodicity theorem (Theorem 20) — it states it and then moves on to the converse. The other gap, Lemma 7, is smaller: a missing one-sentence justification. Both look fillable.

What is genuinely new: the recurrence a_{l+1}=b_l-a_l and z_l = a_{l+1} relating p-ranks of Ext/Tor to the mod-p block algebra, plus the use of Dress's theorem to identify the equivalence classes with O^p(H) conjugacy. The consequence — square-free order gives a full description of Ext^l and Tor_l, alternating between 0 and Z/pZ depending on parity and whether the two conjugacy classes form a 2-element block — is a nice structural result. The non-square-free unboundedness via Gustafson and Gulliksen is also well argued and gives a clean contrast.

Soft spots, in order. First, Theorem 20 is announced with no proof; the text immediately says 'In the remainder we establish the converse.' The intended argument must assemble the p-primary decompositions from Corollary 19 and show the shift by 2 is an A(G)-module isomorphism. That is not formal: the module action is via the mark homomorphisms, and the parity pattern in Corollary 19 gives it for each prime, but you have to verify compatibility across primes and with the A(G)-action. This is the main gap and it is load-bearing. Second, Lemma 7: the product r_E' is normalized only at a chosen representative j of each complementary class, and it doesn't state why it vanishes on the rest of E'. That follows immediately from the definition of d(i,j) (all elements of a block are congruent mod p), so this is a one-line fix, but as written Proposition 8's surjectivity is not fully justified. Third, the notation O^p(H) is written as Op(H), and with subscripts floating it looks like O_p(H); that's a minor readability issue, not substance.

I checked the C_p example myself and the parity pattern from Corollary 14 is right. The recurrences are derived, not fitted. The citations look appropriate; no circularity.

Overall: this is a credible paper with one real missing proof and one near-miss. Readers working on Burnside rings or equivariant cohomology will want the result. A serious referee should see it — the author needs to fill the gap in Theorem 20 and clarify Lemma 7 before it is published. I would accept it for peer review with that expectation.

Referee Report

2 major / 4 minor

Summary. The paper studies Ext and Tor groups between the mark modules Z_H over the Burnside ring A(G) of a finite group G. It first develops the framework of B-rings, i.e. subrings of the ghost ring Gh(I), and proves a structural theorem modulo a prime p: the mod-p quotient R = R/pR decomposes as a direct sum of local k-algebras whose simple modules correspond to equivalence classes of the relation defined by p | d(i,j). Two recurrences relate the p-ranks of Ext^l_R(Z_i,Z_j) and Tor^R_l(Z_i,Z_j) to the dimensions of Ext/Tor over the mod-p algebra, and Corollary 11 derives the clean identity z_l = a_{l+1}. Applying this to the Burnside ring via the embedding into the ghost ring and Dress's congruence criterion, the paper claims that for square-free |G| all Ext and Tor groups are periodic in l with period 2, with p-primary parts equal to Z/pZ on a parity determined by the ∼_p-class structure (Corollary 19 and Theorem 20). For non-square-free |G|, it claims that Ext and Tor have unbounded rank (Theorem 23), proved using Gustafson's theorem that A(G)⊗F_p is not symmetric and Gulliksen's theorem on the homology of local rings.

Significance. If the missing proof of Theorem 20 is supplied, the paper would give a complete, explicit computation of Ext and Tor between mark modules for all finite groups of square-free order, and a sharp unboundedness result in the complementary case. The B-ring framework and the two-term recurrences are elegant and of independent interest; Corollary 11 (z_l = a_{l+1}) is a particularly clean structural fact. The use of Dress's characterization to read the p-divisibility of d(H,J) directly from the subgroup lattice is a strong idea, and the non-square-free argument correctly imports Gustafson's and Gulliksen's theorems. The claims are stated with full precision and are falsifiable.

major comments (2)

[Theorem 20] The central square-free periodicity theorem is stated without proof. After the statement the text reads 'In the remainder we establish the converse,' and the proof never returns to Theorem 20. The author should supply a proof that the p-primary pattern of Corollary 19 assembles into an A(G)-module isomorphism Ext^l_{A(G)}(Z_H,Z_J) ≅ Ext^{l+2}_{A(G)}(Z_H,Z_J) for each l ≥ 1. This requires checking that the A(G)-module structure on each non-zero Z/pZ summand (given by the mark homomorphism at H or at J) is the same at l and at l+2, and that the direct sum over primes p of the p-primary parts is a direct sum of A(G)-submodules. Without this argument the abstract's 'complete description' in the square-free case is not established.
[Lemma 7] The proof of Lemma 7 constructs r as the product over the complementary equivalence classes E' of elements r_{E'} that are normalized only at a chosen representative j of E': the proof ensures r_{E'}(j) = 0 and r_{E'}(i) ≡ 1 mod p, but it does not justify that r_{E'}(j') ≡ 0 mod p for all remaining j' ∈ E'. This follows from the definition of ∼_p, because every element of R is constant modulo p on each equivalence class, so the vanishing at j propagates to all of E'. Since this vanishing is needed for the surjectivity of θ in Proposition 8, and hence for the block decomposition used in Corollaries 11, 14, and 19, the one-line justification should be added.

minor comments (4)

[Proposition 3] In the converse direction of Proposition 3, the claim that the coordinate projections of S' land in Z would benefit from a one-sentence justification: a subring of Q that is finitely generated as a Z-module is contained in Z, and this applies factorwise.
[Theorem 23] Theorem 23 proves unbounded rank only for Ext, whereas the abstract also promises unbounded Tor. The author should add the one-line deduction from Corollary 11 (z_l = a_{l+1}) so that the stated result matches the abstract.
[B-rings section] The term 'rank' is used for the p-rank (dimension over F_p) when defining a_l and z_l, while the same word elsewhere refers to the Z-rank of finite abelian groups; clarifying this would avoid ambiguity.
[Affiliation and diagrams] There are minor typographical issues: the affiliation line reads 'York YO10 5D D' with an extra space, and the displayed long exact sequence after the short exact sequence (†) is poorly aligned. These do not affect the mathematics.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the derivation is self-contained, with external theorems used only for the unbounded case; Theorem 20 is under-proved but that is a gap, not a circular reduction.

full rationale

No load-bearing circular step appears in the derivation chain. The ring-theoretic invariant d(i,j) is read off the ring by Definition 5, not fitted to the target Ext/Tor groups. The p-rank recurrences a_{l+1}=b_l-a_l and z_l=y_{l+1}-z_{l+1} are derived from the long exact sequence coming from (†) and Lemma 6, and then solved from boundary values in Corollaries 11 and 14, so the claimed periodicity is not imposed by construction. The block decomposition in Proposition 8 is argued from Lemma 7 and a nilpotence argument; Lemma 7 is terse, but it is not circular because each r_{E'} is chosen to separate the fixed class E from E' and the congruence classes make the normalization hold on whole equivalence classes. The square-free description in Corollary 19 combines Dress's criterion, the two-dimensional block algebra k[x]/(x^2), and Lemma 18 to identify the p-primary parts, and the unbounded-rank direction invokes the external theorems of Gustafson and Gulliksen. Theorem 20 is indeed stated without proof before the text moves to the converse; the assembly of the p-primary data into an A(G)-module isomorphism is not written out. That is a missing argument about the A(G)-module structure, not a circular reduction: neither the theorem nor Corollary 19 is fed back into the assumptions, no fitted parameter is renamed as a prediction, and no load-bearing self-citation is present.

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

The central results rest on standard homological algebra and three cited external theorems (Dress, Gustafson, Gulliksen). The B-ring framework is proven internally; no free parameters are fitted. The paper's own derivations (recurrences, Corollary 11) are self-contained.

assumptions (5)

domain assumption Standard facts on Burnside rings and the ghost ring embedding (tom Dieck, Chapter 1).
Used in Preliminaries and Proposition 15; assumes the mark homomorphisms define an embedding into the ghost ring and that transitive G-sets correspond to conjugacy classes of subgroups.
domain assumption Dress's congruence theorem: π_H(a)≡π_J(a) mod p for all a iff O^p(H) is conjugate to O^p(J) (Proposition 16, cited [2]).
This is the load-bearing external characterization that identifies the equivalence classes ∼_p on ccs(G).
domain assumption Gustafson's theorem: if p^2 | |G| then A(G)⊗F is not symmetric over a field of characteristic p (Theorem 21, cited [5]).
Used in Theorem 23 to get a non-symmetric block.
domain assumption Gulliksen's theorem: boundedness of dim Tor^S_l(k,k) is characterized by d(S) ≥ dim M/M^2 - 1 (Theorem 22, cited [3]).
Used in Theorem 23 to infer unbounded Betti numbers.
standard math Weibel Proposition 3.3.10: base change of Ext/Tor commutes with localization (Lemma 4, cited [7]).
Used to show Ext/Tor are torsion and finite.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Cohomology of Burnside Rings." pith.science (2026). https://pith.science/paper/BNLB75HG

@misc{pith2026190806156,
  author       = {Pith},
  title        = {Pith review of: Cohomology of Burnside Rings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BNLB75HG}},
  note         = {Machine review of arXiv:1908.06156}
}

abstract

Let $G$ be a finite group and $A(G)$ its Burnside ring. For $H \subset G$ let $\mathbb{Z}_H$ denote the $A(G)$-module corresponding to the mark homomorphism associated to $H$. When the order of $G$ is square-free we give a complete description of the $A(G)$-modules $\textrm{Ext}^l_{A(G)}(\mathbb{Z}_H, \mathbb{Z}_J)$ and $\textrm{Tor}^{A(G)}_l(\mathbb{Z}_H, \mathbb{Z}_J)$ for any $H, J \subset G$ and $l \geq 0$. We show that if the order of $G$ is not square-free then there exist $H, J \subset G$ such that $\textrm{Ext}^l_{A(G)}(\mathbb{Z}_H, \mathbb{Z}_J)$ and $\textrm{Tor}^{A(G)}_l(\mathbb{Z}_H, \mathbb{Z}_J)$ have unbounded rank as finite groups.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

7 extracted references · 7 canonical work pages

[1]

Auslander, I

M. Auslander, I. Reiten, S. O. Smalo, Representation The ory of Artin Algebras, Cambridge University Press, 1997

work page 1997
[2]

A. W. M. Dress, A characterisation of solvable groups, Ma th. Z., 110:213-217, 1969

work page 1969
[3]

T. H. Gulliksen, A note on the homology of local rings, Mat h. Scand. 21 (1967) 296–300

work page 1967
[4]

T. H. Gulliksen, G. Levin, Homology of local rings, Queen ’s papers in pure and applied math- ematics no. 20, 1969

work page 1969
[5]

W. H. Gustafson, Burnside rings which are Gorenstein, Co mm. Algebra 5 (1977) 1–15

work page 1977
[6]

tom Dieck, Transformation Groups and Representation Theory, Lecture Notes in Math., vol

T. tom Dieck, Transformation Groups and Representation Theory, Lecture Notes in Math., vol. 766, Springer, Berlin, 1979

work page 1979
[7]

Weibel, An Introduction to Homological Algebra, Camb ridge University Press, 1994

C. Weibel, An Introduction to Homological Algebra, Camb ridge University Press, 1994

work page 1994

Pith tools

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

[1] [1]

Auslander, I

M. Auslander, I. Reiten, S. O. Smalo, Representation The ory of Artin Algebras, Cambridge University Press, 1997

work page 1997

[2] [2]

A. W. M. Dress, A characterisation of solvable groups, Ma th. Z., 110:213-217, 1969

work page 1969

[3] [3]

T. H. Gulliksen, A note on the homology of local rings, Mat h. Scand. 21 (1967) 296–300

work page 1967

[4] [4]

T. H. Gulliksen, G. Levin, Homology of local rings, Queen ’s papers in pure and applied math- ematics no. 20, 1969

work page 1969

[5] [5]

W. H. Gustafson, Burnside rings which are Gorenstein, Co mm. Algebra 5 (1977) 1–15

work page 1977

[6] [6]

tom Dieck, Transformation Groups and Representation Theory, Lecture Notes in Math., vol

T. tom Dieck, Transformation Groups and Representation Theory, Lecture Notes in Math., vol. 766, Springer, Berlin, 1979

work page 1979

[7] [7]

Weibel, An Introduction to Homological Algebra, Camb ridge University Press, 1994

C. Weibel, An Introduction to Homological Algebra, Camb ridge University Press, 1994

work page 1994