REVIEW 3 major objections 7 minor 24 references
Second symmetric powers of matroids settle a 50-year-old conjecture
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 · glm-5.2
2026-07-08 12:37 UTC pith:3JXAPV4B
load-bearing objection Resolves Mason's 1981 conjecture: positive for k=2, negative for k≥3. The duality with rigidity theory is the engine that makes it work. the 3 major comments →
Symmetric Powers of Matroids
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
Mason's Rank Conjecture is true for second symmetric powers of all matroids and for third symmetric powers of non-free uniform matroids, but false for k>=3 in general and for k>=4 for uniform matroids. The positive results are proved by translating the problem to a dual setting involving abstract rigidity matroids, where a Canonical Base Property (CBP) — stating that certain recursively-constructed subsets of Sym^k(V) are bases of N — serves as the bridge between rank conditions and multilinearity conditions. The counterexamples are constructed by building matroids on Sym^k(V) that satisfy the required flat and multilinearity properties but have rank strictly exceeding the conjectured bound,
What carries the argument
Duality between symmetric powers of matroids and abstract rigidity matroids; canonical subsets of Sym^k(V) and construction trees; the Canonical Base Property (CBP); 0-extension operations from rigidity theory
Load-bearing premise
The positive result for k=2 depends on a chain of equivalences culminating in Lemma 5.11, whose proof uses a contradiction argument (Claim 5.12) showing that canonical subsets of Sym^k(X) must be bases when X is M-spanning, via the 0-Extension Property. If this contradiction argument does not preserve independence as claimed through the inductive steps, the entire equivalence chain for k=2 would collapse.
What would settle it
Theorem 4.1 provides an explicit counterexample for k>=3 (using a rank-one matroid on words with support size >=2, with single-support words as loops), and Theorem 4.3 provides one for k=4 with uniform matroids (using a modified incidence matrix with an appended column).
If this is right
- The k=2 case confirms that Anderson's definition of symmetric power (based on multilinearity plus maximum rank) coincides with Mason's definition (based on the flat property plus quasi-power structure) for all matroids, unifying two foundational definitions.
- The failure for k>=3 means the flat property alone cannot characterize maximum-rank symmetric quasi-powers in higher degree, blocking the natural generalization of vector-space symmetric power theory to arbitrary matroids.
- The duality with abstract rigidity matroids opens a transfer of techniques: results about rigidity matroids (generic and non-generic) can now inform questions about symmetric powers and vice versa.
- The counterexample construction using incidence matroids and appended columns provides a template for building further counterexamples in related matroid product settings.
Where Pith is reading between the lines
- The boundary between true and false for Mason's conjecture appears to track the boundary between k=2 (where rigidity theory is well-understood, e.g., 2D rigidity has the 0-extension and 1-extension characterizations) and k>=3 (where rigidity theory is less complete). This suggests the conjecture's validity for specific matroid families at k=3 may correlate with the maturity of the corresponding ri
- The fact that the conjecture holds for non-free uniform matroids at k=3 but fails for the free matroid suggests that the presence of dependencies in M provides enough structure to force the rank condition, while the free matroid (having no dependencies) allows too much freedom for the rank to be controlled by the flat property alone.
- Mason's open maximality question (whether a unique maximal symmetric power exists in the weak order) is independent of the rank question settled here, and the paper's counterexamples do not directly address it — but the rigidity duality machinery developed could potentially be applied to it.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper studies symmetric powers of matroids, resolving Mason's Rank Conjecture (Conjecture 1.1) in a nuanced way: the conjecture is confirmed for k=2 (all matroids) and k=3 (non-free uniform matroids), but refuted for k≥3 (general matroids) and k=4 (uniform matroids). The approach proceeds through a duality between symmetric powers and abstract rigidity matroids (Section 3), translating the problem into properties such as RRP, ExtP, 0-ExtP, CycP, and the Canonical Base Property (CBP). The positive results are proved by establishing equivalence chains among these properties (Theorems 2.8–2.12), while the counterexamples are constructed explicitly using a rank-zero matroid (Theorem 4.1) and a modified incidence matrix I'_{n,4} (Theorem 4.3).
Significance. The paper resolves a long-standing open problem of Mason from 1981, clarifying the relationship between two definitions of symmetric powers of matroids. The results are sharp: a complete positive answer for k=2, a positive answer for k=3 uniform matroids, and clean counterexamples for k≥3 and k=4 uniform. The duality framework connecting symmetric powers to abstract rigidity is a significant conceptual contribution that should enable further work. The counterexamples are concrete and verifiable: Theorem 4.1 uses a rank-one matroid with loops, and Theorem 4.3 uses a block-diagonal argument on I'_{n,4} with an explicit rank computation. The positive results rely on an intricate but carefully structured proof involving canonical subsets, construction trees, and the 0-extension operation (Section 5). The parameter-free nature of all results is a strength: no fitted parameters or ad-hoc constructions are used, and the logical flow is define-properties → prove equivalences via duality → construct counterexamples.
major comments (3)
- Lemma 5.14 (ExtP + RRP → DM): The proof constructs a canonical subset S = B' ∪ (τ·Z) where Z is a base of M, and uses the fact that S is a base of N (via CBP from Lemma 5.11) to derive a contradiction. The contradiction argument requires |B ⊔ (τ·Y)| > |S|, where Y is M-spanning with |Y| ≥ r_M(V)+1. However, the step where B' is shown to be N-independent (and hence |B| ≥ |B'| because B is a base of N|_E) implicitly assumes that B' ⊆ E and that B is a base of N|_E. The set B' is constructed as S ∖ (τ·Z), and S is a base of N by CBP, so B' is N-independent. But the claim that |B| ≥ |B'| requires that B' ⊆ E, which holds by construction. This step appears correct but is condensed; a sentence clarifying that B' ⊆ E follows from the construction (since τ_{λ*} = τ and the rightmost leaf contributes τ·Z to S, all other contributions lie in E) would strengthen the argument. This is not a gap buta
- Theorem 4.3 (counterexample for k=4, uniform matroids): The proof proceeds by induction on n, using the block structure of I'_{n,4}. The key step is Claim 4.5, which shows that the rank of N'|_{Sym^4(u,v,w)} exceeds that of N|_{Sym^4(u,v,w)} by one, via the block matrix (11). The argument that the contracted matroids are U^1_3 and U^2_3 respectively (equations (12)–(13)) relies on the linear independence of the diagonal blocks I_{2,2}, I_{2,3}, I_{2,4}. This is stated to follow from Theorem 2.13 and Proposition 4.2. However, the block matrix (11) also has nonzero entries in the upper-right blocks (denoted *), and the claim that these do not affect the rank of the contracted matroid should be verified more explicitly. In particular, the column ♠ has entries y_α for α ∈ {uuvw, uvvw, uvww} and zeros elsewhere; the argument that the contracted matroid is U^2_3 (rather than U^1_3 or U^3_3) on
- Conjecture 1.1 as stated says: 'the rank of N is binom(r(M)+1, k) if and only if F·Sym^{k-1}(V) is a flat of N for every flat F of M.' Note the rank formula uses binom(r(M)+1, k), but the SP-rank Property in Section 2.2 uses binom(r(M)+k-1, k). These are the same formula, but the discrepancy in notation (r(M)+1 vs. r(M)+k-1 in the binomial) could confuse readers. More importantly, the 'if and only if' in Conjecture 1.1 should be clarified: the 'only if' direction (SP-rank + Flat) is known from Anderson [2], and the paper proves the 'if' direction for k=2. The conjecture statement should make explicit which direction is being verified/refuted in each theorem.
minor comments (7)
- The abstract states the paper 'solves Mason's conjecture' but the results are mixed (positive for k=2, negative for k≥3). Consider rephrasing to 'resolves Mason's conjecture' or 'determines the status of Mason's conjecture' to convey the mixed outcome.
- The term 'Symmetric Q-power' is used in Section 2.1 while the introduction uses 'symmetric quasi-power'. Consistency would improve readability.
- In the proof of Theorem 3.3, the base B of N listed in equation (9) should be cross-referenced with the circuit family C to help the reader verify it is indeed a base.
- The notation ⊔ (disjoint union) is used throughout Section 5 without explicit definition on first use. A brief note when first introduced would help.
- In Section 5.2, the construction tree formalism is introduced with a detailed example (Figure 1). The figure is helpful but the labels α_1 through α_11 are small in the figure; enlarging or using a table format for the node labels would reduce cognitive load.
- Reference [6] is cited as a 2025 arXiv preprint (arXiv:2508.11636). This should be confirmed as available at the time of publication.
- In the example following the definition of canonical subsets (Section 5.1), the set lk(4, S) is stated to be a canonical subset of Sym^2(X) with pivot vertex 2, but the verification is left to the reader. Adding one sentence of explanation would help.
Simulated Author's Rebuttal
We thank the referee for a careful reading and for identifying several points where the exposition can be improved. All three comments are well-taken and will be addressed in revision. Two of the three require only clarifying additions; the third identifies a genuine typo in the statement of Conjecture 1.1.
read point-by-point responses
-
Referee: Lemma 5.14 (ExtP + RRP → DM): The proof constructs a canonical subset S = B' ∪ (τ·Z) where Z is a base of M, and uses the fact that S is a base of N (via CBP from Lemma 5.11) to derive a contradiction. The contradiction argument requires |B ⊔ (τ·Y)| > |S|, where Y is M-spanning with |Y| ≥ r_M(V)+1. However, the step where B' is shown to be N-independent (and hence |B| ≥ |B'| because B is a base of N|_E) implicitly assumes that B' ⊆ E and that B is a base of N|_E. The set B' is constructed as S ∖ (τ·Z), and S is a base of N by CBP, so B' is N-independent. But the claim that |B| ≥ |B'| requires that B' ⊆ E, which holds by construction. This step appears correct but is condensed; a sentence clarifying that B' ⊆ E follows from the construction would strengthen the argument.
Authors: We agree that this step, while correct, is overly condensed. The key point is that B' = S ∖ (τ·Z) is constructed so that every element of B' arises from leaf nodes of the construction tree other than the rightmost leaf λ*, and the rightmost leaf is the only one whose associated set S_{λ*} = τ·Z lies in τ·V. Since E = Sym^k(V) ∖ (τ·V), all elements of B' lie in E. We will add a sentence after the definition of B' making this explicit: 'Since τ_{λ*} = τ and S_{λ*} = τ·Z is the only contribution to S from the rightmost leaf, all other contributions B' = S ∖ (τ·Z) lie in E = Sym^k(V) ∖ (τ·V).' revision: yes
-
Referee: Theorem 4.3 (counterexample for k=4, uniform matroids): The proof proceeds by induction on n, using the block structure of I'_{n,4}. The key step is Claim 4.5, which shows that the rank of N'|_{Sym^4(u,v,w)} exceeds that of N|_{Sym^4(u,v,w)} by one, via the block matrix (11). The argument that the contracted matroids are U^1_3 and U^2_3 respectively (equations (12)–(13)) relies on the linear independence of the diagonal blocks I_{2,2}, I_{2,3}, I_{2,4}. This is stated to follow from Theorem 2.13 and Proposition 4.2. However, the block matrix (11) also has nonzero entries in the upper-right blocks (denoted *), and the claim that these do not affect the rank of the contracted matroid should be verified more explicitly. In particular, the column ♠ has entries y_α for α ∈ {uuvw, uvvw, uvww} and zeros elsewhere; the argument that the contracted matroid is U^2_3 (rather than U^1_3 or U^3_3) on
Authors: The referee is right that the role of the off-diagonal * entries deserves explicit justification. The argument proceeds by block Gaussian elimination. Since the diagonal blocks I_{2,2}, I_{2,3}, I_{2,4} each have linearly independent columns (by Theorem 2.13 and Proposition 4.2), we can use row operations within the last three row blocks to eliminate all * entries in those rows. After this elimination, the * entries in the first three rows (the uvw rows) in the columns corresponding to the last three blocks can also be eliminated using the now-purified identity blocks below them. The resulting matrix has nonzero entries only in the first column (indexed by uvw) and the ♠ column, restricted to the first three rows. For N (without ♠), this gives a single nonzero column of rank 1, i.e., U^1_3. For N' (with ♠), the ♠ column has entries y_{uuvw}, y_{uvvw}, y_{uvww} which are distinct indeterminates, so the resulting 3×2 matrix has rank 2, giving U^2_3. We will add a paragraph after the display of matrix (11) explaining this elimination step explicitly. revision: yes
-
Referee: Conjecture 1.1 as stated says: 'the rank of N is binom(r(M)+1, k) if and only if F·Sym^{k-1}(V) is a flat of N for every flat F of M.' Note the rank formula uses binom(r(M)+1, k), but the SP-rank Property in Section 2.2 uses binom(r(M)+k-1, k). These are the same formula, but the discrepancy in notation (r(M)+1 vs. r(M)+k-1 in the binomial) could confuse readers. More importantly, the 'if and only if' in Conjecture 1.1 should be clarified: the 'only if' direction (SP-rank + Flat) is known from Anderson [2], and the paper proves the 'if' direction for k=2. The conjecture statement should make explicit which direction is being verified/refuted in each theorem.
Authors: We are grateful to the referee for catching this. In fact, the two formulas are not the same for k ≥ 3: binom(r(M)+1, k) ≠ binom(r(M)+k-1, k) in general. The statement of Conjecture 1.1 contains a typo: it should read binom(rankM+k-1, k) to match the SP-rank Property in Section 2.2 and the surrounding text (which states that the maximum possible rank of a k-th symmetric quasi-power is binom(rankM+k-1, k)). This is purely a typographical error and does not affect any of the results or proofs. We will correct it. We will also add explicit remarks after Theorems 2.1, 2.3, and 2.5 stating which direction of the conjecture is being verified or refuted. Specifically: Theorem 2.1 verifies both directions for k=2 (necessity from [2], sufficiency from our equivalence chain); Theorem 2.3(b) refutes the 'if' direction for k≥3; Theorem 2.5(a) verifies the 'if' direction for k=3 uniform matroids; and Theorem 2.5(b) refutes the 'if' direction for k=4 uniform matroids. revision: yes
Circularity Check
No significant circularity identified.
full rationale
This is a pure mathematics paper proving theorems about symmetric powers of matroids. There are no fitted parameters, no empirical predictions, and no ansatz smuggled through citations. The main logical flow proceeds by: (1) defining properties (Multilinearity, Flat Property, SP-rank, etc.), (2) establishing dual formulations via Lemma 2.7 using standard matroid duality, (3) proving equivalence chains through canonical subsets and 0-extension operations (Sections 5-6), and (4) constructing explicit counterexamples (Section 4). Self-citations [6, 12, 13] are used for context and prior results on uniform matroids, but the central theorems (2.1, 2.3, 2.5) are proved from first principles within this paper. The proof of Lemma 5.11 (including Claim 5.12) proceeds by contradiction using 0-ExtP and induction, and each step is justified by previously established lemmas (5.5, 5.6) or definitions. The counterexamples in Section 4 are constructed explicitly (e.g., Theorem 4.1 uses a rank-one uniform matroid with loops, Theorem 4.3 uses a modified incidence matrix). No step reduces to its inputs by construction. The derivation is self-contained against external mathematical benchmarks.
Axiom & Free-Parameter Ledger
axioms (4)
- standard math Standard matroid duality: for a matroid M on V and N on Sym^k(V), rank and closure relations between M and N transfer to dual relations between M* and N* via standard rank formulas.
- standard math Edmonds' theorem: the transversal matroid M(B_{n,k}) is the row matroid of the generic adjacency matrix I_{n,k}.
- standard math The circuit elimination axiom for matroids.
- standard math Pascal's formula for binomial coefficients: binom(n-1,k,r) + binom(n,k-1,r) = binom(n,k,r).
invented entities (4)
-
Canonical subsets of Sym^k(X) with respect to M
independent evidence
-
Construction tree for a canonical subset
independent evidence
-
Abstract M-rigidity matroid
independent evidence
-
Modified incidence matrix I'_{n,4}
independent evidence
read the original abstract
The study of matroid products has become an active area of research, owing to their connections with tropical ideals and linear representability. In this paper, we study matroidal abstractions of the multilinearity of symmetric powers of vector spaces, using a duality between symmetric powers of matroids and abstract rigidity. These observations allow us to solve Mason's conjecture concerning the equivalence of two definitions of a symmetric power of a matroid. We show that Mason's conjecture holds for second symmetric powers of matroids whereas it fails for third symmetric powers.
Figures
Reference graph
Works this paper leans on
-
[1]
thesis, Queen Mary University of London, 2024
Nicholas Anderson,Matroid powers in tropical geometry, Ph.D. thesis, Queen Mary University of London, 2024. 35
work page 2024
-
[2]
,Matroid products in tropical geometry, Research in the Mathematical Sciences 11(2024), no. 2
work page 2024
-
[3]
Krist´ of B´ erczi, Bogl´ arka Geh´ er, Andr´ as Imolay, L´ aszl´ o Lov´ asz, Bal´ azs Maga, and Tam´ as Schwarcz,Matroid products via submodular coupling, Proceedings of the 57th Annual ACM Symposium on Theory of Computing, STOC ’25, Association for Com- puting Machinery, 2025, pp. 2074–2085
work page 2025
-
[4]
Krist´ of B´ erczi, Bogl´ arka Geh´ er, Andr´ as Imolay, L´ aszl´ o Lov´ asz, Carles Padr´ o, and Tam´ as Schwarcz,Interaction between skew-representability, tensor products, exten- sion properties, and rank inequalities, Proceedings of the 2026 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), SIAM, 2026, pp. 328–354
work page 2026
-
[5]
Joshua Brakensiek, Manik Dhar, Jiyang Gao, Sivakanth Gopi, and Matt Larson, Rigidity matroids and linear algebraic matroids with applications to matrix completion and tensor codes, Combinatorica (2026), To appear
work page 2026
-
[6]
James Cruickshank, Bill Jackson, Tibor Jord´ an, and Shin-ichi Tanigawa,Rigidity of graphs and frameworks: A matroid theoretic approach, 2025, arXiv:2508.11636
work page internal anchor Pith review Pith/arXiv arXiv 2025
-
[7]
Jan Draisma and Felipe Rinc´ on,Tropical ideals do not realise all bergman fans, Re- search in the Mathematical Sciences8(2021), no. 3, 44
work page 2021
-
[8]
Jack Edmonds,Systems of discrete representatives and linear algebra, Journal of Research of the National Bureau of Standards, Section B71B(1967), no. 4, 241– 245
work page 1967
-
[9]
Graver,Rigidity matroids, SIAM Journal on Discrete Mathematics4(1991), 355–368
Jack E. Graver,Rigidity matroids, SIAM Journal on Discrete Mathematics4(1991), 355–368
work page 1991
-
[10]
Jack E. Graver, Brigitte Servatius, and Herman Servatius,Combinatorial rigidity, American Mathematical Society, 1993
work page 1993
-
[11]
Bill Jackson, Tibor Jord´ an, and Shin-ichi Tanigawa,Combinatorial conditions for the unique completability of low-rank matrices, SIAM Journal on Discrete Mathematics 28(2014), no. 4, 1797–1819
work page 2014
-
[12]
Bill Jackson and Shin-ichi Tanigawa,Maximal matroids in weak order posets, Journal of Combinatorial Theory, Series B165(2024), 20–46
work page 2024
-
[13]
,Symmetric tensor matroids, dual rigidity matroids, and the maximality con- jecture, 2025, To appear in Fields Institute Communications, arXiv:2503.14780
work page internal anchor Pith review Pith/arXiv arXiv 2025
-
[14]
Gil Kalai,Hyperconnectivity of graphs, Graphs and Combinatorics1(1985), 65–79
work page 1985
-
[15]
Joseph P. S. Kung,Strong maps, Theory of Matroids (Neil White, ed.), Encyclopedia of Mathematics and its Applications, vol. 26, Cambridge University Press, Cambridge, 1986, pp. 224–253
work page 1986
-
[16]
Joseph P. S. Kung and Hien Q. Nguyen,Weak maps, Theory of Matroids (Neil White, ed.), Encyclopedia of Mathematics and its Applications, vol. 26, Cambridge Univer- sity Press, Cambridge, 1986, pp. 254–271. 36
work page 1986
- [17]
-
[18]
L´ aszl´ o Lov´ asz,Flats in matroids and geometric graphs, Combinatorial Surveys (Pe- ter J. Cameron, ed.), Academic Press, 1977, Proceedings of the Sixth British Com- binatorial Conference, Royal Holloway College, Egham, pp. 45–86
work page 1977
-
[19]
Diane Maclagan and Felipe Rinc´ on,Tropical ideals, Compositio Mathematica154 (2018), no. 3, 640–670
work page 2018
-
[20]
John H. Mason,Glueing matroids together: a study of dilworth truncations and ma- troid analogues of exterior and symmetric powers, Algebraic Methods in Graph The- ory (L´ aszl´ o Lov´ asz and Vera T. S´ os, eds.), Colloquia Mathematica Societatis J´ anos Bolyai, vol. 25, North-Holland, 1981, pp. 519–561
work page 1981
-
[21]
Viet-Hang Nguyen,On abstract rigidity matroids, SIAM Journal on Discrete Mathe- matics24(2010), 363–369
work page 2010
-
[22]
Oxley,Matroid theory, 2 ed., Oxford University Press, Oxford, 2011
James G. Oxley,Matroid theory, 2 ed., Oxford University Press, Oxford, 2011
work page 2011
-
[23]
Amit Singer and Mihai Cucuringu,Uniqueness of low-rank matrix completion by rigidity theory, SIAM Journal on Matrix Analysis and Applications31(2010), no. 4, 1621–1641
work page 2010
-
[24]
Walter Whiteley,Some matroids from discrete applied geometry, Matroid Theory (Joseph E. Bonin, James G. Oxley, and Brigitte Servatius, eds.), Contemporary Math- ematics, vol. 197, American Mathematical Society, Providence, RI, 1996, Seattle, WA, 1995, pp. 171–311. 37
work page 1996
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.