Pith. sign in

REVIEW 4 major objections 4 minor 4 cited by

An elementary classification of the quaternionic reflection groups of rank two

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Every imprimitive rank-two quaternionic reflection group has a canonical form, with two known duplicates; the paper gives the complete corrected list.

desk verdict A real correction to Cohen's classification, with the dicyclic part solidly argued but the polyhedral reflection-system lists asserted rather than demonstrated; refereeing should press on the enumeration. read the letter →

arxiv 2509.01849 v1 pith:5NYV5SSB submitted 2025-09-02 math.GR math.RT

Shayne Waldron This is my paper · ORCID
classification math.GRmath.RT MSC 05B3015B3320C2520G2051M0551M2015B5751E99
keywords imprimitivequaternionicreflectiongroupsranktwosystemsbinarypolyhedraldicyclicfinitesubgroupsofunitquaternionsclassificationmonomialmatrices
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 paper aims to give a complete, elementary classification of the finite imprimitive irreducible quaternionic reflection groups of rank two, meaning finite groups generated by reflections, i.e. linear maps that fix a one-dimensional subspace of a two-dimensional quaternionic space. It shows that each such group is determined by a 'reflection system': a subset L of a finite subgroup K of the unit quaternions that contains 1, generates K, and is closed under the binary operation a∘b=ab^{-1}a. The main theorem lists all these groups in canonical form in two tables and proves the form is unique except for two explicit isomorphisms. The classification corrects and extends Cohen's earlier list: for example, there are four non-isomorphic imprimitive groups of order 192 with 22 reflections, one of which had not been previously identified. The point of the elementary approach is that questions about reflection groups become finite algebraic computations over the classical finite quaternion groups: cyclic, dicyclic, and binary tetrahedral, octahedral, and icosahedral.

What carries the argument

The central object is the reflection system: for a finite subgroup K of the unit quaternions, a reflection system is a subset L with 1∈L, K=⟨L⟩, and closure under a∘b=ab^{-1}a. Closure encodes the algebra of products and conjugates of the off-diagonal reflection matrices; products of such reflections naturally produce diagonal entries from L, and the operation a∘b keeps L closed under conjugation-like moves. Each reflection system L, together with a normal subgroup H, yields the canonical group G_K(L,H), so enumerating reflection systems replaces the classification of reflection groups by finite combinatorics. For the dicyclic groups the enumeration reduces to a parameter set Ω_n of coprime

What would settle it

Run an exhaustive computer search over all subsets of the binary tetrahedral, octahedral, and icosahedral groups that contain 1 and are closed under ab^{-1}a; if any subset appears that is not one of the systems listed in Examples 4.3 and 4.5, the classification is incomplete. Similarly, finding one unlisted isomorphism between groups from Tables 1 and 2 would refute the uniqueness claim.

Watch

Extended reading notes

Core claim

The central claim is Theorem 6.1: every finite imprimitive irreducible rank-two quaternionic reflection group can be written uniquely in the canonical form G_K(L,H), except for two cases that have two canonical forms: G_O(L_O^14,1) is isomorphic to G_T(L_T^12,C_2), and G(n,1,n,2) is isomorphic to G(2n,2,n,1) for odd n. Here K is one of the finite subgroups of the unit quaternions, L is a reflection system for K, and H is a normal subgroup satisfying H⊂L and LH=L; the canonical form consists of all matrices with a diagonal part from K, a coset factor determined by L, and the optional swap matrix. The proof enumerates reflection systems for each possible K, computes the base reflection group f

Load-bearing premise

The classification is complete only if the hand listings of reflection systems for the binary tetrahedral, octahedral, and icosahedral groups are complete; the paper states these listings come from 'elementary calculations' without giving an exhaustive search procedure.

Editorial extensions

If this is right

  • If the classification is correct, Tables 1 and 2 form the complete roster: every imprimitive irreducible rank-two quaternionic reflection group appears exactly once, except the two flagged duplicate labels.
  • The canonical form gives immediate structural data, including the order |G|=2|H||K| and the number of reflections 2|H|+|L|-2, without constructing the group.
  • Several previously unlisted groups enter the classification, such as the order-192 group with index [6,1,3,4], and the corrected list reconciles earlier counts of imprimitive versus primitive-conjugate groups.
  • For rank n≥3, the same setup yields the canonical form G_n(K,H), so the higher-rank imprimitive classification follows once the rank-two case is settled.

Reading between the lines

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

  • The paper's reliance on hand-enumerated reflection systems for the binary tetrahedral, octahedral, and icosahedral groups suggests a natural testable extension: an exhaustive machine search over all subsets of these groups closed under a∘b would independently certify completeness.
  • Because the operation a∘b=ab^{-1}a depends only on the underlying group multiplication, the reflection-system method should transfer to imprimitive reflection groups over other finite subgroups of the unit quaternions, or to analogous noncommutative coefficient rings, not just the quaternionic case treated here.
  • The infinite families of index pairs described in Corollary 5.1 point to many non-isomorphic groups that agree in order and reflection count; the reflection-orbit invariant is likely the practical distinguisher, and it may reveal further unknown groups at higher orders.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper gives an elementary classification of finite imprimitive irreducible rank-two quaternionic reflection groups. The main new tool is a "reflection system" L, a subset of a finite subgroup K of the unit quaternions containing 1 and closed under a∘b=ab^{-1}a. The author shows that any imprimitive group can be put in a canonical form G_K(L,H), with H a normal subgroup of K, and reduces the classification to enumerating reflection systems and then determining the admissible H. The cyclic, dicyclic, and binary polyhedral cases are treated; Tables 1 and 2 list the resulting groups. The classification theorem (Theorem 6.1) states that every such group has a unique canonical form except for two explicit isomorphism exceptions: G_O(L_O^{14},1) ≅ G_T(L_T^{12},C_2) and G(n,1,n,2) ≅ G(2n,2,n,1) for odd n. The paper also claims to correct and complete Cohen's 1980 table, including a previously missing group of order 192 with 22 reflections, and sketches a rank-n analogue.

Significance. If the completeness claims are correct, this is a substantial and useful contribution: it gives a transparent, structural classification of an important family, identifies gaps and double-counting in the existing literature, and provides explicit generators and reflection-orbit data for every group. The Section 5 dicyclic-group analysis is a genuine derivation: Lemma 5.2, Theorem 5.1, and Theorem 5.3 give real proofs, and the canonical-form lemma (Lemma 2.2) is clean and load-bearing. The paper is also careful to state exactly where it relies on calculations. However, the central classification claim is broader than the proved lemmas: the completeness of the reflection-system inventories for T, O, and I, and several uniqueness/no-isomorphism assertions, are presented as finite computations without a certified enumeration or reproducible check. The Magma code in Section 6 constructs groups from indices but does not certify that the lists of reflection systems are exhaustive. Thus the paper's main theorem is defensible but not yet fully supported.

major comments (4)
  1. [Examples 4.3 and 4.5; after (2.14)] The classification theorem depends critically on the assertions that the binary octahedral and icosahedral groups have exactly five and four reflection systems, respectively, and that each polyhedral K has at most one reflection system of a given size. These are introduced with "Elementary calculations show" (Examples 4.3 and 4.5) and "It turns out (from our calculations)" (after (2.14)). No derivation, pseudocode, or certified computation is supplied. A missing reflection system would remove a group from Tables 1 and 2; an invalid listed system would add a spurious group. This is a computational premise, not covered by the lemmas in Sections 2 and 5. Please provide a verifiable certificate: for example, an exhaustive closure computation over conjugacy classes of subsets, or a proof using the automorphism description (2.10) together with the explicit finite subgroup data in the appendix.
  2. [Section 2, case (i) after (2.14)] Uniqueness of the canonical form for a fixed K is justified by the sentence "It turns out (from our calculations) that this is always the case." This is load-bearing for the "uniquely in canonical form" part of Theorem 6.1. If, for some K, there were two inequivalent reflection systems of the same size or two distinct normal subgroups of the same order, then the order/reflection-count conditions in (2.14) would not distinguish the corresponding groups, and the uniqueness claim would fail. The normal-subgroup part for the listed K is easy to check, but the reflection-system uniqueness is exactly the unshown enumeration. Please supply the calculation or a complete argument.
  3. [Theorem 5.2 proof] The proof that the higher-order dicyclic groups G_{D_n}(L^{(n)}_{(a,b)}, C_{2n/ab}) are in canonical form for odd ab is deferred: the text says "in can be shown" (presumably "it can be shown") and gives no argument. This is load-bearing because it determines which H can be added to a reflection system without introducing new nondiagonal reflections; if the claim failed for some (a,b), Table 2 would contain invalid lines. The promised "considering all the diagonal matrices" case check should be written out or replaced by a complete, checkable computation. Also, the displayed "C_{2n/ab}=⟨ω^{2b}⟩" appears inconsistent with Table 2's "C_{2n/ab}=⟨ω^{ab}⟩" and with the convention C_r=⟨ω^{2n/r}⟩; please correct the theorem statement.
  4. [Proposition 5.1 and Example 5.4] The exclusion of isomorphisms between polyhedral-group reflection groups and dicyclic-group reflection groups is argued by reducing to finitely many orders (48, 96, 192, 384, 480, 768) and then "simply examin[ing] the reflection structure of each group, or their isomorphism class." Example 5.4 lists SmallGroup identifiers for the relevant orders, but it does not state the isomorphism tests performed or provide a script that certifies absence of further coincidences. This is also load-bearing for Theorem 6.1's uniqueness claim. Please make the finite check explicit and reproducible, and clarify the meaning of the "*" entries in Example 5.4 (which pairs are asserted to be isomorphic).
minor comments (4)
  1. [Various] Typos and wording: Table 2 heading "imprimtive" should be "imprimitive"; Corollary 5.1 has "descriminant" for "discriminant"; Theorem 5.2 has "in can be shown" for "it can be shown"; Example 4.1 has "occurences" for "occurrences."
  2. [Reference [DZ24]] The title "N =▽ SCFTs" appears to contain a placeholder or rendering error; please check that the symbol is correct.
  3. [Figure 3] The edge label "G(n,a,b,2n/ab)" in Figure 3 applies only when ab is odd; the caption or figure should state this explicitly, as the surrounding text does.
  4. [Example 5.4] The asterisk notation in the order-48 table is introduced only as "with * indicating an isomorphism," but it is not immediately clear which starred entries are being identified with which. Please define the pairing explicitly.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the central classification derivation is self-contained; self-citations are peripheral and the hand-enumerated reflection systems are computational premises, not circular inputs.

full rationale

The core derivation chain is not circular. Starting from the monomial form (1.1), the paper derives the reflection-system closure condition ab^{-1}a in L from closure under conjugation of reflections (Section 2, before Definition 2.1), and Lemma 2.2 derives the canonical form and group order from the group structure. The classification in Theorem 6.1 then follows by enumerating reflection systems for the finite subgroups K of U(H), which are taken from Stringham's external classification. No equation is fitted to data, and no fitted parameter is renamed as a prediction. The self-citations [Wal24], [BW25], and [Wal25] are not load-bearing: [Wal24]/[BW25] only contextualize how primitive rank-two groups arise from imprimitive ones, and [Wal25] is used in Example 4.1 for reflection-orbit terminology and normality observations, not to prove the completeness or uniqueness assertions of Theorem 6.1. The paper does contain unshown computational assertions that are relevant to completeness: Examples 4.3 and 4.5 introduce the reflection systems for O and I with 'Elementary calculations show', the discussion after (2.14) uses 'It turns out (from our calculations)' to justify uniqueness of canonical labels for a fixed K, Theorem 5.2 says 'it can be shown' for the canonical form of one higher-order family, and Proposition 5.1 relies on a finite case check (Example 5.4). These are gaps in the written verification and computational premises, not circular steps: the theorem does not assume those lists; it asserts them as the output of the enumerated calculation. The dicyclic case, by contrast, is proved in detail in Lemma 5.2 and Theorems 5.1-5.3. Thus the central claim retains independent mathematical content, and the only circularity concern is the minor presence of self-citations that are not load-bearing.

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

The classification rests on standard finite-subgroup structure plus several unstated computational completeness assertions. There are no numerical free parameters fitted to data. The reflection-system concept is a mathematical definition rather than an empirical postulate.

assumptions (5)
  • standard math Finite subgroups of the unit quaternions are exactly the Stringham list: cyclic, dicyclic, binary tetrahedral, binary octahedral, binary icosahedral.
    Invoked at the start of Section 3 as Theorem 5.12 of [LT09]; the dicyclic and polyhedral case analysis depends on this list being exhaustive.
  • domain assumption Every irreducible imprimitive rank-two quaternionic reflection group is conjugate to a unitary monomial group whose reflections have the three forms in (1.1).
    Stated at the beginning of Section 1 without proof; it is the imprimitivity structure on which the canonical-form reduction and the definition of L and H rest.
  • ad hoc to paper The reflection systems for the binary tetrahedral, octahedral, and icosahedral groups are exactly the lists given in Examples 4.1, 4.3, and 4.5.
    These exhaustive lists are asserted via "Elementary calculations show" with no derivation or code; Theorem 6.1 inherits completeness from them.
  • ad hoc to paper The canonical form of the higher-order dicyclic groups for odd ab in Theorem 5.2 holds as asserted.
    The proof states "it can be shown" without giving the diagonal-product argument; this supports the infinite family of dicyclic groups.
  • ad hoc to paper The case check in Proposition 5.1 showing no isomorphisms between polyhedral and dicyclic groups is exhaustive.
    Proposition 5.1 refers to a calculation said to be possible by examining reflection structures, but the details and the full case analysis are not included.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An elementary classification of the quaternionic reflection groups of rank two." pith.science (2026). https://pith.science/paper/5NYV5SSB

@misc{pith2026250901849,
  author       = {Pith},
  title        = {Pith review of: An elementary classification of the quaternionic reflection groups of rank two},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5NYV5SSB}},
  note         = {Machine review of arXiv:2509.01849}
}
read the original abstract

We give an elementary classification and presentation of the finite quaternionic reflection groups of rank two, based on the notion of a``reflection system''. This simplifies the existing classification, which is shown to be incomplete, e.g., there exist four imprimitive quaternionic reflection groups of order 192 with 22 reflections which are not isomorphic (one of which was previously unknown).

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Systems of imprimitivity for rank two quaternionic reflection groups

    math.GR 2025-10 accept novelty 7.0 of 10

    Revises Cohen's classification of imprimitive rank-two quaternionic reflection groups, adds missing groups, and shows some primitive complex reflection groups admit infinitely many quaternionic systems of imprimitivity.

  2. Namikawa--Weyl groups of symplectic quotient singularities

    math.SG 2026-07 accept novelty 6.5 of 10

    Every irreducible Weyl group arises as a factor of the Namikawa–Weyl group of some symplectic quotient singularity V/G with G a symplectic reflection group.

  3. The lattice of normal reflection subgroups of an irreducible reflection group

    math.GR 2026-07 conditional novelty 6.5 of 10

    Normal reflection subgroups of an irreducible reflection group form a lattice combinatorially indexed by conjugacy orbits of root-line reflection subgroups, and every complex reflection group is normal in a unique max...

  4. Invariants in the cohomology of the complement of quaternionic reflection arrangements

    math.RT 2025-10 conditional novelty 6.0 of 10

    Invariant cohomology Poincaré polynomials of quaternionic reflection arrangements coincide with the complex cases except for imprimitive groups with non-cyclic K/H, where P(t^{1/3}) = 1+2t+...+2t^{n-2}+5t^{n-1}+4t^n.

Reference graph

Works this paper leans on

17 extracted references · 15 canonical work pages · cited by 4 Pith papers

  1. [1]

    Finite collineation groups : with an introduction to the theory of operators and substitution groups

    Hans Frederik Blichfeldt. Finite collineation groups : with an introduction to the theory of operators and substitution groups. University of Chicago science series. University of Chicago Press, Chicago, 1917

    work page 1917

  2. [2]

    On parabolic subgroups of symplectic reflection groups

    Gwyn Bellamy, Johannes Schmitt, and Ulrich Thiel. On parabolic subgroups of symplectic reflection groups. Glasg. Math. J. , 65(2):401--413, 2023

    work page 2023

  3. [3]

    Quaternionic MUBs in ^2 and their reflection symmetries

    Zachary Buckley and Shayne Waldron. Quaternionic MUBs in ^2 and their reflection symmetries. preprint, 9 2025

    work page 2025

  4. [4]

    Arjeh M. Cohen. Finite quaternionic reflection groups. J. Algebra , 64(2):293--324, 1980

    work page 1980

  5. [5]

    H. S. M. Coxeter. Discrete groups generated by reflections. Ann. of Math. (2) , 35(3):588--621, 1934

    work page 1934

  6. [6]

    Conway and Derek A

    John H. Conway and Derek A. Smith. On quaternions and octonions: their geometry, arithmetic, and symmetry . A K Peters, Ltd., Natick, MA, 2003

    work page 2003

  7. [7]

    N = 5 SCFT s and quaternionic reflection groups

    Anirudh Deb and Gabi Zafrir. N = 5 SCFT s and quaternionic reflection groups. J. High Energy Phys. , (8):Paper No. 17, 58, 2024

    work page 2024

  8. [8]

    William M. Kantor. Generation of linear groups. In The geometric vein , pages 497--509. Springer, New York-Berlin, 1981

    work page 1981

Show all 17 references
  1. [9]

    Lehrer and Donald E

    Gustav I. Lehrer and Donald E. Taylor. Unitary reflection groups , volume 20 of Australian Mathematical Society Lecture Series . Cambridge University Press, Cambridge, 2009

    2009

  2. [10]

    a lzische Technische Universit \

    Johannes Schmitt. On Q-factorial terminalizations of symplectic linear quotient singularities . doctoralthesis, Rheinland-Pf \"a lzische Technische Universit \"a t Kaiserslautern-Landau, 2023

    2023

  3. [11]

    G. C. Shephard and J. A. Todd. Finite unitary reflection groups. Canad. J. Math. , 6:274--304, 1954

    1954

  4. [12]

    W. I. Stringham. Determination of the F inite Q uaternion G roups. Amer. J. Math. , 4(1-4):345--357, 1881

  5. [13]

    Finite quaternionic reflection groups in magma

    Don Taylor. Finite quaternionic reflection groups in magma. talk, 4 2025

    2025

  6. [14]

    Quaternion algebras , volume 288 of Graduate Texts in Mathematics

    John Voight. Quaternion algebras , volume 288 of Graduate Texts in Mathematics . Springer, Cham, [2021] 2021

    2021

  7. [15]

    The geometry of the six quaternionic equiangular lines in H ^2

    Shayne Waldron. The geometry of the six quaternionic equiangular lines in H ^2 . arXiv , 2411.16766, 2024

    2024 arXiv

  8. [16]

    The lattice of normal reflection subgroups of an irreducible reflection group

    Shayne Waldron. The lattice of normal reflection subgroups of an irreducible reflection group. preprint, 6 2025

    2025

  9. [17]

    Quaternions and matrices of quaternions

    Fuzhen Zhang. Quaternions and matrices of quaternions. Linear Algebra Appl. , 251:21--57, 1997

    1997

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.