REVIEW 3 major objections 4 minor 13 references
On the local representation theory of symmetric groups
T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read The action of the normalizer of a Sylow p-subgroup on all its irreducible characters is fully described by relabeling rules on T-functions, and the Galois action by a parallel relabeling rule.
desk verdict Solid, likely-correct formulas for the normalizer and Galois actions on Irr(P_n), but the load-bearing character parametrization is asserted, not proved. 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 central object is the $T$-function: an equivalence class of labeling functions $t\colon s_{p^k} \to [0,p]$ on the $p^k$-skeleton (the set of finite sequences in $[1,p]$ of length at most $k-1$, ordered by concatenation), where two labelings are equivalent when they differ by cyclically permuting subtrees below vertices as allowed by elements of $P_p$, and admissibility conditions select precisely the classes corresponding to irreducible characters. The action of the normalizer is carried by the permutation $\tau$ on $[0,p]$ (fixing $0$ and $p$, and raising the $p$-cycle $(1,\ldots,p)$ to a primitive power $b$) together with generators $\sigma_j^{(k)}$ that act on trees level by level: conjugation by $\sigma_k^{(k)}$ replaces each entry by its $\tau$-image and permutes the $p$ subfunctions by $\tau$, while $\sigma_j^{(k)}$ for $j<k$ applies the same rule recursively inside every subfunction. This machinery reduces character-action questions to elementary relabeling computations on finite sequences.
What would settle it
For $p=5$ and $n=125$, take the two labeling functions $t_1,t_2$ displayed in Remark 4.8 and exhaustively apply the 64 conjugating elements of $N_{125}/P_{125} \cong (C_4)^3$ using the relabeling rules of Theorem 4.4; if any of these maps $t_1$ to $t_2$, then the paper's claimed counterexample collapses and the stated action formulas are wrong.
Extended reading notes
Core claim
The paper's central claim is that the conjugation action of the Sylow normalizer on $\mathrm{Irr}(P_n)$ is fully described once characters are viewed as $T$-functions. Theorem 4.4 states that for every generator $\sigma_{(i,j,\ell)}$ and $\rho_y$ of the normalizer $N_n$, a $T$-function transforms by the prime-power relabeling rule of Theorem 4.2 on the corresponding component (for $\sigma$) and by permuting the components according to $\rho_y$ (for $\rho$), so every normalizer orbit is computable by these rules. Theorem 4.7 states the Galois action in the same language: $T^\sigma(s) = ((T(\theta_1)(s_1))^\tau, \ldots, (T(\theta_{q_t})(s_{q_t}))^\tau)$, i.e., the fixed generator $\sigma$ of the Galois group acts by the single permutation $\tau$ on each label. As a direct consequence, the paper exhibits two irreducible characters of the Sylow $5$-subgroup of $S_{125}$ that are Galois conjugate but lie in distinct normalizer orbits, showing that a rigidity property true for linear characters does not extend to all characters.
Load-bearing premise
The entire description rests on the bijection asserted in Lemma 3.16 between the $T$-function equivalence classes defined in this paper and the irreducible characters of $P_n$; if those classes do not match the tree parametrization exactly, then the relabeling formulas act on the wrong set of characters.
Editorial extensions
If this is right
- All normalizer orbits and Galois orbits in $\mathrm{Irr}(P_n)$ can be enumerated by repeated application of the relabeling rules in Theorems 4.4 and 4.7, for any prime $p$ and any $n$.
- The known description of the normalizer action on linear characters is recovered as the special case where every label on a given level of the $T$-function is equal.
- Because the $T$-function parametrization is equivalent to the tree model of [GL25], the action formulas transfer directly to that model and to the study of Sylow branching coefficients.
- The $S_{125}$ example shows that the induced character $\theta^{\uparrow S_n}$ does not determine the normalizer orbit of a non-linear character, so the linear-character rigidity result of [L19] has no full analogue.
- The Galois action formula allows the field of values of every irreducible character of $P_n$ to be read off from the $T$-function labels.
Reading between the lines
- The level-by-level character of the relabeling rules suggests that the size of a normalizer orbit is governed by multiplicities of equal subfunctions at each node, which could lead to closed formulas for orbit counts that the paper does not derive.
- The same $T$-function formalism may apply to Sylow subgroups of other groups built from iterated wreath products of $C_p$, where analogous normalizer actions are of interest.
- A natural testable extension is to compare, for small $p$ and $n$, the number of Galois orbits with the number of normalizer orbits in $\mathrm{Irr}(P_n)$; if the discrepancy grows systematically, the $S_{125}$ phenomenon is generic rather than exceptional.
- Because the formulas depend on a single permutation $\tau$ of labels, they may also yield an explicit description of how far the Galois action deviates from being realizable inside the normalizer, quantifying the failure of linear-character rigidity for higher-degree characters.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops a function-based parametrization of the irreducible characters of a Sylow p-subgroup P_n of a symmetric group. The main objects are admissible equivalence classes of labeling functions on the p-ary tree skeleton, called T-functions; Lemma 3.16 asserts a bijection between Irr(P_{p^k}) and the set of admissible k-tree functions, and the general case for arbitrary n is assembled from direct products in Definition 3.20. The paper then proves formulas for the action of the normalizer generators on these T-functions: Theorem 4.2 treats the prime-power case, Corollary 4.3 gives a pointwise formula, and Theorem 4.4 describes the action for arbitrary n using Proposition 2.10. Theorems 4.6 and 4.7 give the Galois action pointwise as T^σ(s) = (T(s))^τ. Remark 4.8 presents an explicit pair of Galois-conjugate but not N-conjugate characters in S_125.
Significance. If the stated bijection is fully established, the paper would provide a transparent and algorithmically usable description of both the normalizer action and the Galois action on Irr(P_n), going substantially beyond the linear-character case treated in [Gia21] and [L19]. The explicit recursive formulas in Theorem 4.2 and the pointwise formula in Corollary 4.3 are attractive, and the counterexample in Remark 4.8 is a valuable and checkable contribution. The main conceptual ingredient, the T-function reformulation of the tree parametrization from [GL25], is plausible but currently rests on an unproved bijection, so the significance of the paper is conditional on completing that proof.
major comments (3)
- [Section 3, Lemma 3.16] Lemma 3.16 is the load-bearing result of the paper but is stated without proof. The sentence preceding it says the construction 'closely mirrors' [GL25] and 'allows us to recover [GL25, Lemma 3.6]', but the lemma itself is not proved. What is missing is: (i) a proof that the equivalence relation ∼_{p^k} of Definition 3.7 has exactly the same classes as the tree equivalence in [GL25, Definition 3.5(a)]; (ii) a proof that admissibility (Definition 3.12) is invariant under this equivalence, so that F_{p^k} is well defined as a subset of the quotient; and (iii) a proof that the recursively defined Ψ_{p^k} is well defined on equivalence classes and is the two-sided inverse of Φ_{p^k}. Every formula in Section 4, including Theorems 4.2, 4.4, 4.6, and 4.7, and the counterexample in Remark 4.8, uses this bijection; without a proof, the paper does not establish that the T-functions describe Irr(P_n).
- [Notation 3.8 and Definition 3.7] The paper identifies an equivalence class with an arbitrary chosen representative and then evaluates pointwise, for instance writing T^σ(s) = (T(s))^τ in Theorems 4.6 and 4.7. This is legitimate only if the proposed action is well defined on equivalence classes. The manuscript does not show that the action of σ_j^{(k)} on representatives descends to the quotient, nor does it prove that the pointwise formulas in Corollary 4.3 are independent of the chosen representative. The author should either choose a canonical transversal for the equivalence relation or prove invariance of the action and of the evaluation formulas under ∼_{p^k}.
- [Section 2.3.2, Proposition 2.10] Proposition 2.10 is stated with the proof omitted, described only as 'a straightforward consequence of Proposition 2.5'. This proposition is the bridge from the prime-power normalizer action to the general case in Theorem 4.4, which is one of the paper's central claims. The proof should either be supplied in full or the statement should be accompanied by a precise reference to a proved result that covers exactly this normalizer action.
minor comments (4)
- [Section 3, after Definition 3.21] The displayed 'Theorem' following Definition 3.21 is unnumbered and unproved. If it is intended as an immediate consequence of Lemma 3.16 and Definition 3.20, it should be stated as such, and if not, it needs its own proof.
- [Remark 4.8] The counterexample claims that the T-functions T_1 and T_2 satisfy T_2 = T_1^σ and are not N_{125}-conjugate, but the non-conjugacy assertion is only described as 'readily verified'. Since this is a central advertised consequence, a short verification of the non-conjugacy, for example by exhibiting an invariant that distinguishes the two N-orbits, should be included.
- [Remark 4.8, item (ii)] The preimage set of 5 in the second labeling function is written as (t_1)^{-1}(5), but it should presumably be (t_2)^{-1}(5). Please correct the subscript.
- [Corollary 4.3] In the formula for j < k and ℓ(s) > k−j, the symbol i is used in the notation (i)_τ but is not defined in the statement; presumably i = s_{k−j}. The notation should be clarified.
Circularity Check
No circular dependence found: the normalizer and Galois action formulas are derived from direct character computations, while the unproved parametrization lemma is an external prior result rather than a self-fed input.
full rationale
The target claims are Theorem 4.4 (normalizer action on Irr(P_n)) and Theorem 4.7 (Galois action). Theorem 4.2 is proved by direct character evaluation: for instance, in the subcase j = k of Case (a), the proof computes θ^{σ_j^(k)}(g) step by step from the definition of X(θ1;ε) and invokes [JK81, 4.4.10]; it does not assume the orbit structure that the theorem is meant to establish. Theorem 4.6 is likewise proved by induction on k, showing (X(ψ;ε))^σ = X(ψ^σ;(ε)^τ) and that (θ1×⋯×θp)↑ transforms under σ to (θ1^σ×⋯×θp^σ)↑; the argument is based on values of characters, not on the desired conclusion. The main load-bearing input is the T-function parametrization of Irr(P_{p^k}). Lemma 3.16 states the required bijection, but its proof is omitted: the preceding sentence says the construction 'closely mirrors' [GL25] and 'allows us to recover [GL25, Lemma 3.6]'. This is a genuine completeness gap, and the skeptic's concern about well-definedness of the quotient and of pointwise evaluation is legitimate. However, this is not circularity: [GL25] is an external prior paper by Giannelli and Law, with no author overlap with the present work, and the equivalence classes of Definition 3.7 and admissibility conditions of Definition 3.12 are explicit constructions rather than definitions of the normalizer or Galois orbits. Proposition 2.10 is also stated without proof ('The proof is omitted'), but it is presented as a straightforward transport of Proposition 2.5 and is not used to define the characters. No parameter is fitted and no predicted quantity is fed back into the formulas; Remark 4.8's counterexample is verified using Theorems 4.2 and 4.6, not by building orbit equivalence into the T-functions. In summary, the derivation chain is not circular; the identified weaknesses are omitted proofs and dependence on prior literature, not self-referential reductions.
Assumptions & free parameters
assumptions (5)
- standard math Sylow p-subgroup structure: P_{p^k} is isomorphic to P_{p^{k-1}} wreath C_p, and P_n decomposes into a product of such groups; the normalizer N_n has the wreath-product form given in [Ol76, Lemma 4.1].
- standard math For a wreath product G wreath H, the irreducible characters lying over phi^{times n} are exactly X(phi; psi), and the product and induction rules of [JK81, 4.4.10] hold.
- domain assumption The tree parametrization of Irr(P_n) in [GL25] is correct, and the quotient equivalence in Definition 3.7 reproduces it.
- standard math The cyclotomic field Q(omega) is a splitting field for P_n, and the Galois action on linear characters acts as phi_epsilon maps to phi_{(epsilon)tau}, from [JK81, 4.4.8].
- standard math The base group B of P_{p^k} is generated by elements with fixed points, and B is normal in the normalizer, from [Ol76, Lemma 4.2].
Cite this review
Pith. "Pith review of On the local representation theory of symmetric groups." pith.science (2026). https://pith.science/paper/UOB55MW7
@misc{pith2026250604773,
author = {Pith},
title = {Pith review of: On the local representation theory of symmetric groups},
year = {2026},
howpublished = {\url{https://pith.science/paper/UOB55MW7}},
note = {Machine review of arXiv:2506.04773}
}
abstract
Given a Sylow $p$-subgroup $P$ of a symmetric group, we describe the action of its normalizer on $\mathrm{Irr}(P)$. To this end, we establish a one-to-one correspondence between the irreducible characters of $P$ and certain equivalence classes of explicitly defined functions, which are also naturally suited to describing the Galois action.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
Giannelli, Characters of odd degree of symmetric groups, J
E. Giannelli, Characters of odd degree of symmetric groups, J. London Math. Soc. (1), 96 (2017), 1--14
work page 2017
-
[2]
E. Giannelli, McKay bijections for symmetric and alternating groups, Algebra Number Theory 15(7) (2021), 1809--1835
work page 2021
-
[3]
E. Giannelli and S. Law, Sylow branching coefficients for symmetric groups, J. London Math. Soc. (2) 103 (2021), 697--728
work page 2021
-
[4]
E. Giannelli and S. Law, Sylow branching trees for symmetric groups, Trans. Amer. Math. Soc. in press
-
[5]
I. M. Isaacs, Character theory of finite groups, Dover, New York, 1976
work page 1976
- [6]
-
[7]
Law, On problems in the representation theory of symmetric groups, Ph.D
S. Law, On problems in the representation theory of symmetric groups, Ph.D. thesis, University of Cambridge, 2019
work page 2019
-
[8]
I. G. Macdonald , On the degrees of the irreducible representations of symmetric groups. Bull. London Math. Soc. 3 (1971), 189--192
work page 1971
Show all 13 references
-
[9]
Navarro , Characters and Blocks of Finite Groups
G. Navarro , Characters and Blocks of Finite Groups. London Mathematical Society Lecture Note Series, 250. Cambridge University Press, Cambridge, 1998
1998
-
[10]
Navarro and P.H.Tiep The fields of values of characters of degree not divisible by p , Forum of Math
G. Navarro and P.H.Tiep The fields of values of characters of degree not divisible by p , Forum of Math. Pi 9 (2021), e2, 1 - 28
2021
-
[11]
Olsson, McKay numbers and heights of characters, Math
J. Olsson, McKay numbers and heights of characters, Math. Scand. 38 (1976), no. 1, 25--42
1976
-
[12]
Olsson , Combinatorics and Representations of Finite Groups
J.B. Olsson , Combinatorics and Representations of Finite Groups. Vorlesungen aus dem Fachbereich Mathematik der Universit\"at Essen, Heft 20, 1994
1994
-
[13]
Ruhstorfer , The Alperin--McKay and Brauer’s Height Zero Conjecture for the prime 2
L. Ruhstorfer , The Alperin--McKay and Brauer’s Height Zero Conjecture for the prime 2. Ann. of Math (2), to appear; https://arxiv.org/pdf/2204.06373.pdf
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.