REVIEW 2 major objections 4 minor 30 references
Rook sums in the symmetric group algebra
T0 review · 2 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read This paper proves that row-to-row rook sums generate ideals whose annihilators and orthogonal complements are exactly complementary ideals, free over any commutative ring, with ranks equal to pattern-avoiding counts, and that rectangular…
desk verdict Solid, mostly self-contained paper on rook sums in the symmetric group algebra; the central theorem is proved, but two auxiliary results are incomplete. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is a lexicographic triangularity result in the spirit of Erdős-Szekeres. Lemma 2.5.5 decomposes any permutation avoiding $12\cdots(k+1)$ into $k$ blocks on which it is decreasing, and Lemma 2.5.6 shows that every other permutation appearing in the corresponding row-to-row sum is lexicographically smaller; this makes the non-avoiding residue classes a triangular basis of $\mathcal{A}/I_k$, and the mirror image makes the avoiding classes a basis of $\mathcal{A}/J_k$. The rectangular product rule is carried by the elementary counting lemma that the number of factor pairs $(u,v)$ with $u(C)=D$, $v(A)=B$, $uv=w$ is either $0$ or the fixed integer $\omega_{B,C}$, depending only on $|w(A)\cap D|$; a binomial inclusion-exclusion identity then converts the product into a $\mathbb{Z}$-linear combination of rook sums and feeds a descending filtration that forces the linear-factor polynomial identity.
What would settle it
Take $n=5$, $k=2$ over $\mathbb{F}_2$. Write the coefficient matrix whose rows are indexed by the row-to-row sums $\nabla_{\mathbf{B},\mathbf{A}}$ for set compositions with at most two blocks and whose columns are indexed by the 120 permutations, and compute its rank by Gaussian elimination. Theorem 2.4.1 predicts the rank is the Catalan number $C_5=42$; any other value, or any linear dependence among the residue classes of the 78 non-123-avoiding permutations in $\mathcal{A}/I_2$, would falsify the central rank claim.
Extended reading notes
Core claim
The central claim is that the rook sums organize the symmetric group algebra into two exactly complementary ideals. Over any commutative ring $k$, $I_k$ equals $J_k^\perp = \operatorname{LAnn} J_k = \operatorname{RAnn} J_k$ and $J_k$ equals $I_k^\perp = \operatorname{LAnn} I_k = \operatorname{RAnn} I_k$; both $I_k$ and $J_k$ are free $k$-modules, of ranks $|\operatorname{Av}_n(k+1)|$ and $n!-|\operatorname{Av}_n(k+1)|$, and the quotient $\mathcal{A}/I_k$ has basis the non-avoiding permutations while $\mathcal{A}/J_k$ has basis the avoiding ones. In the rectangular case, the paper proves the product rule $\nabla_{D,C}\nabla_{B,A} = \omega_{B,C} \sum_{w: |w(A)\cap D|=|B\cap C|} w$ with $\omega_{B,C}=|B\cap C|!\,|B\setminus C|!\,|C\setminus B|!\,|[n]\setminus(B\cup C)|!$, and derives from a filtration argument that every $\nabla_{B,D}$ and tilde $\tilde{\nabla}_{B,D}$ satisfies a product of $|D|+2$ linear factors with integer constants, so its minimal polynomial splits over $\mathbb{Z}$. The paper also identifies $I_k$ and $J_k$ as annihilators of tensor power modules and, when $n!$ is invertible, as the pieces of the Artin-Wedderburn decomposition corresponding to Specht modules of length at most $k$ and greater than $k$.
Load-bearing premise
The argument's load-bearing premise is a purely combinatorial ordering fact: every permutation that avoids the increasing pattern $12\cdots(k+1)$ can be cut into $k$ blocks on which it is decreasing, and every other permutation that sends the same blocks to the same images is lexicographically smaller; if this triangularity statement ever failed, the basis of $\mathcal{A}/I_k$ and the rank formula for $I_k$ would fail with it.
Editorial extensions
If this is right
- Every rectangular rook sum $\nabla_{B,A}$ and every tilde sum $\tilde{\nabla}_{B,D}$ has a minimal polynomial with integer coefficients splitting into linear factors, with at most $|D|+2$ distinct roots, so their spectra are accessible directly from subset sizes.
- Products of rectangular rook sums have the closed form $\omega_{B,C}$ times the sum of all permutations whose images intersect $D$ in exactly $|B\cap C|$ points, and also expand as integer linear combinations of rook sums, giving structure constants for the ideal they span.
- The ideal $I_k$ has rank equal to the number of $12\cdots(k+1)$-avoiding permutations; for $k=2$ this recovers the Catalan number as the dimension of the span of all rectangular rook sums.
- Over any ring $k$, the mutual-annihilation identities $I_kJ_k=J_kI_k=0$ and the annihilator equalities hold, so the two ideals form a pair of complementary pieces inside the group algebra.
- If $n!$ is invertible, $\mathcal{A}\cong I_k\times J_k$ as $k$-algebras, so every Specht module with length at most $k$ is annihilated by $J_k$ and every Specht module with length greater than $k$ by $I_k$.
Reading between the lines
- Going beyond the paper, the characteristic-free rank formulas suggest a concrete test: the same dimensions should hold after reducing modulo a prime, meaning the triangular basis algorithm for the span of row-to-row sums can serve as a modular-basis computation independent of any semisimplicity arguments.
- The abstract nabla algebra $\mathcal{D}$ of dimension $\binom{2n}{n}$, built from formal symbols with the same multiplication rule, is conjectured to be unital exactly when $n!$ is invertible; if that conjecture holds, $\mathcal{D}$ would be a natural noncommutative host whose center dimension and Cartan data could connect to planar rook algebras.
- The paper notes that mixed quotients such as $I_k \cap T_{\operatorname{sign}}(I_\ell)$ have dimensions that depend on the characteristic of $k$; this suggests that any combinatorial model for simultaneous increasing and decreasing pattern avoidance must be genuinely modular, a sharper question than the characteristic-free results proved here.
- One could test whether the same filtration and minimal-polynomial method extends to boards that are unions of finitely many rectangles, where the product rule would presumably acquire additional intersection parameters and the linear-factor splitting would fail at a controlled boundary.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies two families of elements in the symmetric group algebra A = k[S_n]. For subsets A,B ⊆ [n] it defines rectangular rook sums ∇_{B,A} and ~∇_{B,A}, proves an explicit product rule (Theorem 1.2.2), and uses a length filtration to show that these elements satisfy polynomial equations with all roots in the base ring, hence that their minimal polynomials factor into linear factors (Theorem 1.4.1, Corollary 1.4.3). Section 2 generalizes the construction to set decompositions of [n], defining ideals I_k and J_k. The main theorem (Theorem 2.4.1) states that I_k = J_k^⊥ = LAnn J_k = RAnn J_k and J_k = I_k^⊥, that I_k and J_k are free k-modules of ranks |Av_n(k+1)| and |S_n \ Av_n(k+1)|, respectively, that the quotients A/I_k and A/J_k have permutation bases indexed by the complementary sets, and that when n! is invertible the algebra splits as A ≅ I_k × J_k. The paper also identifies J_k and a sign-twist of I_{n-k-1} as annihilators of tensor modules, describes both ideals under the Artin–Wedderburn correspondence in terms of Specht modules, and recovers the classical enumeration of pattern-avoiding permutations.
Significance. If the results hold, the paper gives a substantial structural description of a natural family of elements in the symmetric group algebra. The main strength is that the central theorem of Section 2 is proved by elementary linear algebra and combinatorics, with the lexicographic triangularity argument (Lemmas 2.5.5 and 2.5.6) correctly supporting the spanning and independence claims; I traced this part of the proof and found it sound. The paper also honestly acknowledges overlap with Murphy cellular bases and prior annihilator results, while stressing that the main proof is independent of that machinery. The explicit product rule and the rank/basis statements for the ideals are valuable and give concrete, falsifiable statements. The principal weaknesses are two incomplete theorem-level statements, discussed below, neither of which undermines the proof of Theorem 2.4.1 itself.
major comments (2)
- [Section 1.6, Theorem 1.6.1] This theorem is not proved: the text states "Proof omitted due to excessive ugliness." Since associativity is exactly what makes the vector space D into a nonunital k-algebra, the central claim of the subsection is unsupported. The gap does not affect Theorem 2.4.1, but the statement should be repaired before publication: either provide the proof, weaken the statement to a conjecture, or explicitly label it as a SageMath-verified computational observation with the verification data included.
- [Section 2.9.3, Theorem 2.9.6] The theorem is presented as a formal result but is only supported by an outline. In particular, step (b), which asserts that A/(I_k + T_sign(J_ℓ)) is free of rank |Av'_n(ℓ+1) \ Av_n(k+1)|, depends on Theorem 2.9.4 and on the Murphy-basis identifications from Remark 2.4.3, and the final appeal to [Grinbe25, Lemma 5.21.9] is not fully demonstrated. If Theorem 2.9.6 is intended as a theorem, the proof should be completed; otherwise it should be demoted to a conjecture or a remark with a clear statement of what is checked.
minor comments (4)
- [Section 1.1 and Theorem 1.2.2] The letter A is used both for the group algebra A = k[S_n] and for an arbitrary subset A in expressions such as ∇_{B,A}; in Theorem 1.2.2 the same symbol names both the algebra and one of the four subsets. Renaming the algebra as κ or α (or using lowercase letters for the subsets) would avoid genuine confusion.
- [Section 1.5] The table of minimal polynomials is described as produced by SageMath, but no code, no complete input data, and no independent verification method are supplied. Since the table is a non-trivial computational claim, the data should be made reproducible or at least one non-trivial row should be checked against Corollary 1.4.3.
- [Sections 2.8.2 and 2.9.3] The convention that J_m = A for negative m is introduced when needed (e.g., in Lemma 2.8.5 and Theorem 2.8.3) rather than at the definition of J_k. State this convention near Definition 2.2.1 so that formulas such as T_sign(J_{n-k-1}) are unambiguous from the outset.
- [Remark 2.4.3 and throughout] Several standard facts and the Murphy-basis identifications are cited to the author's own lecture notes [Grinbe25]. Because that reference is a preprint with potentially unstable numbering, it would be helpful to give either a published reference or a short statement of the specific result being used at each citation.
Circularity Check
No significant circularity: central theorem proved from first principles; self-citations are minor and non-load-bearing.
full rationale
The derivation of the central Theorem 2.4.1 is self-contained. The product rule Theorem 1.2.2 follows by counting factorizations (Lemma 1.2.3) and binomial inversion; the minimal-polynomial factorization Corollary 1.4.3 follows from the D-filtration, with the coefficients being explicit integers rather than fitted parameters. The Section 2 main theorem is proved from elementary lemmas: mutual annihilation IkJk=JkIk=0 (Lemma 2.5.4) via a pigeonhole argument and antisymmetrizer identities; lexicographic triangularity (Lemmas 2.5.5-2.5.6) is proved in the text; the spanning/independence arguments (Lemmas 2.5.7-2.5.15) use only this triangularity, the annihilation relation, and the bilinear-form/annihilator correspondence (Lemma 2.5.2). The rank and basis claims follow from the module lemma Lemma 2.5.3. Maschke averaging (Lemmas 2.5.16-2.5.18) yields the split when n! is invertible. No fitted parameter is renamed as a prediction; the counts of Av_n(k+1) emerge as ranks, not as inputs. The paper cites the author's lecture notes [Grinbe25] for standard facts (antipode properties, antisymmetrizer identities, Murphy-basis background, dot-product identities), but these are elementary, parameter-free, and their assumptions do not include the target result; Remark 2.4.3 explicitly states that the main proofs are independent of the Murphy-basis identification. The only flagged gaps are non-circular: Theorem 1.6.1 omits an associativity proof 'due to excessive ugliness', and Theorem 2.9.6 is an outline relying on Theorem 2.9.4; neither feeds into Theorem 2.4.1. No circular step can be exhibited, so the score is in the 0-2 range; 2 is assigned solely because the non-load-bearing self-citations to [Grinbe25] occur in the proof of supporting lemmas.
Assumptions & free parameters
assumptions (5)
- standard math k[S_n] is a free k-module with basis Sn, and the antipode S is an involutive k-algebra anti-automorphism.
- standard math Standard identities for antisymmetrizers ∇^-_U, including ∇^-_U = (1-τ)q and ∇^-_U τ = -∇^-_U, as quoted from [Grinbe25, Props 3.7.4, 5.5.8].
- standard math Murphy cellular bases are bases of k[S_n] over any commutative ring, and the spans of their subfamilies define the ideals F^Row and F^{-Col}.
- standard math When n! is invertible in k, the Artin-Wedderburn map A → ∏_{λ⊢n} End(S^λ) is an isomorphism, with dim S^λ = f^λ and (S^λ)^sign ≅ S^{λ^t}.
- standard math Schensted's theorem and Stanley's refinement counting permutations by longest increasing and decreasing subsequence lengths.
invented entities (1)
-
Abstract Nabla algebra D
Cite this review
Pith. "Pith review of Rook sums in the symmetric group algebra." pith.science (2026). https://pith.science/paper/VXMDCA66
@misc{pith2026250722386,
author = {Pith},
title = {Pith review of: Rook sums in the symmetric group algebra},
year = {2026},
howpublished = {\url{https://pith.science/paper/VXMDCA66}},
note = {Machine review of arXiv:2507.22386}
}
abstract
Let $\mathcal{A}$ be the group algebra $\mathbf{k}[S_n]$ of the $n$-th symmetric group $S_n$ over a commutative ring $\mathbf{k}$. For any two subsets $A$ and $B$ of $[n]$, we define the elements \[ \nabla_{B,A}:=\sum_{\substack{w\in S_n;\\w\left( A\right) =B}} w \qquad \text{and} \qquad \widetilde{\nabla}_{B,A}:=\sum_{\substack{w\in S_n;\\w\left( A\right) \subseteq B}}w \] of $\mathcal{A}$. We study these elements, showing in particular that their minimal polynomials factor into linear factors (with integer coefficients). We express the product $\nabla_{D,C}\nabla_{B,A}$ as a $\mathbb{Z}$-linear combination of $\nabla_{U,V}$'s. More generally, for any two set compositions (i.e., ordered set partitions) $\mathbf{A}$ and $\mathbf{B}$ of $\left\{ 1,2,\ldots,n\right\} $, we define $\nabla_{\mathbf{B},\mathbf{A}}\in\mathcal{A}$ to be the sum of all permutations $w\in S_n$ that send each block of $\mathbf{A}$ to the corresponding block of $\mathbf{B}$. This generalizes $\nabla_{B,A}$. The factorization property of minimal polynomials does not extend to the $\nabla_{\mathbf{B},\mathbf{A}}$, but we describe the ideal spanned by the $\nabla_{\mathbf{B},\mathbf{A}}$ and a further ideal complementary to it. These two ideals have a "mutually annihilative" relationship, are free as $\mathbf{k}$-modules, and appear as annihilators of tensor product $S_n$-representations; they are also closely related to Murphy's cellular bases, Specht modules, pattern-avoiding permutations and even some algebras appearing in quantum information theory.
Reference graph
Works this paper leans on
-
[1]
Rains, Algebraic aspects of increasing subsequences, Duke Math
Jinho Baik, Eric M. Rains, Algebraic aspects of increasing subsequences, Duke Math. J. 109 (2001), no. 1. https://doi.org/10.1215/S0012-7094-01-10911-3
-
[2]
Remmel, Rook Theory Notes, 2011
Fred Butler, Mahir Can, Jim Haglund, Jeffrey B. Remmel, Rook Theory Notes, 2011. ://mathweb.ucsd.edu/ remmel/files/Book.pdf
work page 2011
-
[3]
An integral second fundamental theorem of invariant theory for partition algebras
Chris Bowman, Stephen Doty, Stuart Martin, An integral second fundamental theorem of invariant theory for partition algebras, arXiv:1804.00916v5, published in: Represent. Theory 26 (2022), pp. 437--454. https://doi.org/10.1090/ert/593
work page Pith review arXiv 2022
-
[4]
Canonical bases and new applications of increasing and decreasing subsequences to invariant theory
Chris Bowman, Stephen Doty, Stuart Martin, Canonical bases and new applications of increasing and decreasing subsequences to invariant theory, arXiv:2206.12001v4. https://arxiv.org/abs/2206.12001v4
-
[5]
Mikl\' o s B\' o na, Combinatorics of Permutations, 3rd edition, CRC Press 2022
work page 2022
-
[6]
Adam Bene Watts, Anirban Chowdhury, Aidan Epperly, J. William Helton, Igor Klep, Relaxations and Exact Solutions to Quantum Max Cut via the Algebraic Structure of Swap Operators, arXiv:2307.15661v3, published in: Quantum 8 (2024), 1352. https://arxiv.org/abs/2307.15661v3
arXiv 2024
-
[7]
E. Rodney Canfield, S. G. Williamson, A combinatorial study of straightening bases, Linear and Multilinear Algebra 24 (1989), no. 4, pp. 241--264. https://doi.org/10.1080/03081088908817919
-
[8]
C. De Concini, C. Procesi, A Characteristic Free Approach to Invariant Theory, Advances in Mathematics 21 (1976), Issue 3, pp. 330--354. https://doi.org/10.1016/S0001-8708(76)80003-5
Show all 30 references
-
[9]
https://arxiv.org/abs/2412.01627v1
Stephen Donkin, Greene's Theorem and ideals of the group algebra of a symmetric group, arXiv:2412.01627v1. https://arxiv.org/abs/2412.01627v1
-
[10]
87--102 https://doi.org/10.1093/qmath/hap020
Stephen Doty, Kathryn Nyman, Annihilators of permutation modules, arXiv:0711.3219v5 https://arxiv.org/abs/0711.3219v5, published in: The Quarterly Journal of Mathematics 62, Issue 1, March 2011, pp. 87--102 https://doi.org/10.1093/qmath/hap020
2011 arXiv
-
[11]
https://math.mit.edu/ etingof/repb.pdf
Pavel Etingof, Oleg Golberg, Sebastian Hensel, Tiankai Liu, Alex Schwendner, Dmitry Vaintrob, Elena Yudovina, Introduction to Representation Theory, with historical interludes by Slava Gerovitch, Student Mathematical Library 59, AMS 2011. https://math.mit.edu/ etingof/repb.pdf
2011
-
[12]
David Ellis, Ehud Friedgut, Haran Pilpel, Intersecting families of permutations, 1011.3342v2, updated version of a paper published in: J. Amer. Math. Soc. 24 (2011), pp. 649--682. https://arxiv.org/abs/1011.3342v2
2011 arXiv
-
[13]
775--804
Chaim Even-Zohar, Patterns in random permutations, Combinatorica 40 (2020), pp. 775--804. https://doi.org/10.1007/s00493-020-4212-z See arXiv:1811.07883v4 https://arxiv.org/abs/1811.07883v4 for a preprint
2020 arXiv
-
[14]
Flath, T
D. Flath, T. Halverson, and K. Herbig, The planar rook algebra and Pascal's triangle, Enseign. Math. (2) 55 (2009), no. 1-2, pp. 77--92. https://ems.press/content/serial-article-files/44173
2009
-
[15]
Goldman, J
Jay R. Goldman, J. T. Joichi and Dennis E. White, Rook Theory. I. Rook Equivalence of Ferrers Boards, Proceedings of the American Mathematical Society 52 (1975), pp. 485--492. https://www.ams.org/journals/proc/1975-052-01/S0002-9939-1975-0429578-4/S0002-9939-1975-0429578-4.pdf
1975
-
[16]
://www.cip.ifi.lmu.de/ grinberg/t/19fco/n/n.pdf
Darij Grinberg, Enumerative Combinatorics (Drexel Fall 2019 Math 222 notes), 18 September 2020. ://www.cip.ifi.lmu.de/ grinberg/t/19fco/n/n.pdf
2019
-
[17]
://www.cip.ifi.lmu.de/ grinberg/t/24s/sga.pdf
Darij Grinberg, An introduction to the symmetric group algebra [Math 701, Spring 2024 lecture notes], 27 July 2025, arXiv:2507.20706v1 https://arxiv.org/abs/2507.20706v1. ://www.cip.ifi.lmu.de/ grinberg/t/24s/sga.pdf
2024 arXiv
-
[18]
https://arxiv.org/abs/2503.17552v1
Zachary Hamaker, Brendon Rhoades, Partial permutations and character evaluations, arXiv:2503.17552v1. https://arxiv.org/abs/2503.17552v1
-
[19]
https://doi.org/10.1007/s000130050341 Martin H\" a rterich, Murphy bases of generalized Temperley--Lieb algebras, Arch. Math. 72 (1999), pp. 337--345
1999 doi
-
[20]
https://arxiv.org/abs/2503.20942v2 Igor Klep, Tea S trekelj, Jurij Vol c i c , Quantum Max d-Cut via qudit swap operators, arXiv:2503.20942v2
-
[21]
Loehr, Jeffrey B
Nicholas A. Loehr, Jeffrey B. Remmel, Rook-by-rook rook theory: Bijective proofs of rook and hit equivalences, Advances in Applied Mathematics 42 (2009), Issue 4, pp. 483--503. https://doi.org/10.1016/j.aam.2008.09.003
2009 doi
-
[22]
G. E. Murphy, On the Representation Theory of the Symmetric Groups and Associated Hecke Algebras, Journal of Algebra 152 (1992), pp. 492--513. https://doi.org/10.1016/0021-8693(92)90045-N
1992 doi
-
[23]
https://doi.org/10.1006/jabr.1995.1079 G. E. Murphy, The Representations of Hecke Algebras of Type A_ n , Journal of Algebra 173 (1995), pp. 97--121
1995
-
[24]
https://arxiv.org/abs/2008.11129v6
Claudio Procesi, A note on the Weingarten function, arXiv:2008.11129v6. https://arxiv.org/abs/2008.11129v6
2008 arXiv
-
[25]
K. N. Raghavan, Preena Samuel, K. V. Subrahmanyam, RSK bases and Kazhdan-Lusztig cells, arXiv:0902.2842v3 https://arxiv.org/abs/0902.2842v3, published in: Annales de l'institut Fourier 62 (2012), issue 2, pp. 525--569 https://eudml.org/doc/251104
2012 arXiv
-
[26]
Daniel Edwin Rutherford, Substitutional Analysis, Edinburgh University Press 1948
1948
-
[27]
Schensted, Longest increasing and decreasing subsequences, Canadian Journal of Mathematics 13 (1961), pp
C. Schensted, Longest increasing and decreasing subsequences, Canadian Journal of Mathematics 13 (1961), pp. 179--191. https://doi.org/10.4153/CJM-1961-015-3
1961 doi
-
[28]
Travis Scrimshaw, Cellular subalgebras of the partition algebra, J. Comb. Algebra 8 (2024), no. 1/2, pp. 147--207. https://doi.org/10.4171/jca/84
2024 doi
-
[29]
Stanley, Enumerative Combinatorics, volume 2, Second edition, Cambridge University Press 2023
Richard P. Stanley, Enumerative Combinatorics, volume 2, Second edition, Cambridge University Press 2023. See http://math.mit.edu/ rstan/ec/ for errata
2023
-
[30]
Stanley, Theory and Application of Plane Partitions
Richard P. Stanley, Theory and Application of Plane Partitions. Part 2, Studies in Applied Mathematics L (1971), no. 3, pp. 259--279. https://math.mit.edu/ rstan/pubs/pubfiles/12-2.pdf
1971
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.