Fusion rules from the Norton inequality

Alonso Castillo-Ramirez

Norton inequality forces the 0- and 1-eigenspaces of any idempotent to be subalgebras when the Frobenius form is non-degenerate.

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 →

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

T0 review · grok-4.5

2026-07-12 03:54 UTC pith:UX6QWJIL

load-bearing objection Clean elementary answer to an open axial-algebra question: Norton alone forces A0 and A1 to be subalgebras when the form is non-degenerate, with an explicit radical counter-example.

arxiv 2607.03237 v1 pith:UX6QWJIL submitted 2026-07-03 math.RA

Fusion rules from the Norton inequality

Alonso Castillo-Ramirez This is my paper

classification math.RA MSC 17A9917C2720D08

keywords Norton inequalityFrobenius formaxial algebrasMajorana algebrasidempotentsfusion ruleseigenspaces

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

In commutative real algebras that carry a Frobenius form obeying the Norton inequality, the 0- and 1-eigenspaces of every idempotent behave like subalgebras, and their product lands in the half-eigenspace. When the form is non-degenerate the inclusions are exact; when it is degenerate they hold only modulo the radical. The paper answers a concrete question by showing that the non-degenerate case (which includes all Majorana algebras) yields genuine subalgebras, while an explicit four-dimensional axial algebra with degenerate form shows that the radical can spoil the subalgebra property. The argument never uses a prescribed fusion law; it works for arbitrary idempotents and rests only on the positive-semidefiniteness of an auxiliary bilinear form built from the Norton inequality.

Core claim

If a commutative real algebra admits a non-degenerate Frobenius form satisfying the Norton inequality, then for every idempotent e both A0(e) and A1(e) are subalgebras and A0(e)A1(e) sits inside A1/2(e). In the degenerate case the same statements hold modulo the radical of the form. An explicit axial algebra demonstrates that the radical can prevent A0(e) from being a subalgebra.

What carries the argument

The Norton form Bu(x,y)=(u^{2},xy)-(ux,uy). Positive-semidefiniteness of Bu, together with the elementary fact that a vector of zero Bu-norm is orthogonal to everything, forces (Le-λ id) of products of λ-eigenvectors into the radical for λ=0,1 and forces (2Le-id) of mixed 0-1 products into the radical.

Load-bearing premise

The Norton inequality must make the auxiliary form Bu positive-semidefinite for every element u; if that fails, none of the eigenspace inclusions follow.

What would settle it

Produce a commutative real algebra with a non-degenerate Frobenius form that satisfies the Norton inequality yet contains an idempotent whose 0-eigenspace is not closed under multiplication.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

Clean elementary answer to an open axial-algebra question: Norton alone forces A0 and A1 to be subalgebras when the form is non-degenerate, with an explicit radical counter-example.

read the letter

This short note settles Problem 5.5 of Gorshkov–Shpectorov: under a Frobenius form that obeys the Norton inequality, every idempotent e has A0(e) and A1(e) as subalgebras (and the mixed product lands in A1/2) once the form is non-degenerate. The same inclusions hold modulo the radical in the degenerate case. That is the whole result, and it is new.

What is actually new is the removal of the equality-case hypothesis that appeared in the 2015 Majorana paper, the extension from axes to arbitrary idempotents, and the precise tracking of the radical. The argument is elementary linear algebra: the Norton form Bu is positive-semidefinite by definition of the inequality, vanishes on (e,e) for eigenvectors of eigenvalue 0 or 1, and therefore annihilates everything; polarization then gives the products. The mixed case uses only orthogonality of distinct eigenspaces. No fusion law is assumed, so the conclusion is forced by Norton alone. The four-dimensional counter-example with radical spanned by a2-a3 and r is concrete and shows that non-degeneracy is essential.

Soft spots are minor. The paper is short and specialized; impact stays inside axial/Majorana work. The “Norton form” is just a convenient name for the bilinearization of the given inequality, not a new object. Citations are light and appropriate; the earlier stronger result is properly distinguished. Nothing is circular and nothing is smuggled.

Anyone working on Majorana algebras, Matsuo algebras, or associative subalgebras of the Griess algebra will want the statement. The math is checkable by hand in an afternoon. I would send it to a referee without hesitation; it is a clean, useful note that does exactly what it claims.

Referee Report

0 major / 4 minor

Summary. The paper studies commutative real algebras equipped with a Frobenius form that satisfies the Norton inequality (x^{2},y^{2})≥(xy,xy). It answers a question of Mudziiri Shumba and Shpectorov by showing that, for an arbitrary idempotent e, the eigenspaces A_{0}(e) and A_{1}(e) are subalgebras and A_{0}(e)A_{1}(e)⊆A_{1/2}(e) whenever the form is non-degenerate (Corollary 3.4). In the general (possibly degenerate) case the same inclusions hold modulo the radical A⊥ of the form (Theorem 3.3). The argument introduces the Norton form B_u(x,y):=(u^{2},xy)-(ux,uy), proves it is positive-semidefinite from the Norton inequality, and then uses positivity together with the polarization identity and eigenspace orthogonality to control products of eigenvectors. A short explicit axial algebra with degenerate form is given in which A_{0}(a_{1}) fails to be a subalgebra, showing that the radical is essential. The results apply to every idempotent and do not presuppose any axial fusion law.

Significance. The result cleanly settles an open question that has appeared in the axial-algebra literature and isolates the precise contribution of the Norton inequality alone. Because the argument never assumes a prescribed fusion law and works for arbitrary idempotents, it strengthens the structural foundations of Majorana algebras, Griess algebras and non-degenerate Matsuo algebras. The explicit counter-example with a degenerate form is a useful calibration of the hypotheses. The proofs are elementary, fully written out and free of hidden analytic assumptions; the only auxiliary object introduced is the Norton form, which is a direct rewriting of the given inequality. These features make the paper a solid, self-contained contribution that will be cited by workers in axial algebras and related non-associative structures.

minor comments (4)

In the mixed-case calculation of Theorem 3.3 the identity (u^{2},e)=(u,eu) is used without an explicit parenthetical reminder that the form is Frobenius; a one-line citation of Definition 2.1 would improve readability for non-specialists.
Example 3.7 asserts that the quotient A/A⊥ is associative and that the induced form is the usual dot product; a brief verification of the multiplication table of the two-dimensional quotient would make the claim fully self-contained.
The fusion table displayed in Example 3.7 uses the symbol “0,1” for the product of two 0-eigenvectors; a short sentence clarifying that this means the product may land in either eigenspace would avoid possible ambiguity.
References [2] and [3] are cited as arXiv preprints; if journal versions have appeared by the time of publication they should be updated.

Circularity Check

0 steps flagged

No circularity: Norton inequality alone yields the eigenspace inclusions via an auxiliary positive-semidefinite form; target fusion rules never re-enter the hypotheses.

full rationale

The derivation is self-contained and non-circular. The Norton form Bu is defined directly from the given Frobenius form (Definition 3.1) and is positive-semidefinite solely by rewriting the Norton inequality (Lemma 3.2). Theorem 3.3 then applies the elementary vanishing lemma for positive-semidefinite forms (Lemma 2.3) to Bu(e,·) for idempotent e, together with the polarization identity for products of eigenvectors and the orthogonality of distinct eigenspaces (Lemma 2.4). The resulting radical membership statements specialize, when the form is non-degenerate, to the claimed subalgebra and fusion inclusions (Corollary 3.4). No parameter is fitted, no uniqueness theorem is imported from the author’s prior work, and the target fusion rules do not appear among the hypotheses. The single self-citation to the author’s earlier paper [1] is used only to note that a stronger equality-case version of Norton already implied the same conclusion for Majorana algebras; the present argument is explicitly independent of that stronger hypothesis and of any prescribed axial fusion law. The counter-example (Example 3.7) further confirms that the radical is essential and is not assumed away. Consequently the central claim is a genuine first-principles consequence of the stated axioms.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 1 invented entities

The paper works entirely inside the standard axiomatic setting of commutative real algebras with associative bilinear forms. No free parameters are fitted; the only non-standard object introduced is the auxiliary Norton form, which is defined directly from the given form and is not postulated as a new physical or algebraic entity. Background facts (positive-semidefinite forms vanish on orthogonal complements, eigenspaces of self-adjoint operators are orthogonal) are classical linear algebra.

axioms (4)

domain assumption The bilinear form is associative: (xy,z)=(x,yz) for all x,y,z (Definition 2.1).
Standard definition of a Frobenius form in the axial-algebra literature; used throughout to make every adjoint self-adjoint.
domain assumption Norton inequality (x^{2},y^{2})≥(xy,xy) for all x,y (Definition 2.2).
The sole non-trivial hypothesis; taken as given for the class of algebras under study.
standard math A positive-semidefinite symmetric bilinear form B with B(v,v)=0 satisfies B(v,w)=0 for all w (Lemma 2.3).
Elementary calculus/quadratic-form fact used to extract the radical membership of products.
standard math Distinct eigenspaces of a self-adjoint operator are orthogonal (Lemma 2.4).
Classical linear-algebra fact for the Frobenius form.

invented entities (1)

Norton form Bu(x,y):=(u^{2},xy)-(ux,uy) no independent evidence
purpose: Auxiliary positive-semidefinite form that converts the Norton inequality into a tool for controlling products of eigenvectors.
Defined directly from the given Frobenius form; not an independent algebraic structure but a convenient rewriting. No external evidence is claimed or needed.

pith-pipeline@v1.1.0-grok45 · 10867 in / 2509 out tokens · 19978 ms · 2026-07-12T03:54:00.785579+00:00 · methodology

0 comments

Cite this review

Pith. "Pith review of Fusion rules from the Norton inequality." pith.science (2026). https://pith.science/paper/UX6QWJIL

@misc{pith2026260703237,
  author       = {Pith},
  title        = {Pith review of: Fusion rules from the Norton inequality},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UX6QWJIL}},
  note         = {Machine review of arXiv:2607.03237}
}

Share X Bluesky LinkedIn Reddit HN

read the original abstract

We study fusion rules forced by the Norton inequality in commutative non-associative real algebras equipped with a Frobenius form. We answer a question of T. M. Mudziiri Shumba and S. Shpectorov concerning whether the eigenspace $A_0(e)$ associated with an arbitrary idempotent $e\in A$ must be a subalgebra of $A$. If the Frobenius form is non-degenerate, as in Majorana algebras, then for every idempotent $e\in A$, both $A_0(e)$ and $A_1(e)$ are subalgebras of $A$, and \[ A_0(e)A_1(e)\subseteq A_{1/2}(e). \] In the degenerate case, the corresponding inclusions hold modulo the radical of the Frobenius form. We also give an explicit axial algebra with a degenerate Frobenius form satisfying the Norton inequality for which $A_0(e)$ is not a subalgebra for one of its axes.

discussion (0)

Reference graph

Works this paper leans on

9 extracted references · 1 linked inside Pith

[1]

Castillo-Ramirez,Associative subalgebras of low-dimensional Majorana algebras, J

A. Castillo-Ramirez,Associative subalgebras of low-dimensional Majorana algebras, J. Algebra 421 (2015), 159–188

2015
[2]

I. B. Gorshkov, J. McInroy, T. M. Mudziiri Shumba, and S. Shpectorov,Automorphism groups of axial algebras, J. Algebra 661 (2025), 657–712

2025
[3]

Gorshkov and S

I. Gorshkov and S. Shpectorov,Axial Algebras: Questions and Conjectures, arXiv:2606.30048

Pith/arXiv arXiv
[4]

J. I. Hall, F. Rehren, and S. Shpectorov,Universal axial algebras and a theorem of Sakuma, J. Algebra 421 (2015), 394–424

2015
[5]

A. A. Ivanov,The Monster Group and Majorana Involutions, Cambridge Tracts in Math- ematics 176, Cambridge University Press, 2009

2009
[6]

Khasraw, J

S. Khasraw, J. McInroy, and S. Shpectorov,On the structure of axial algebras, Trans. Amer. Math. Soc. 373 (2020), no. 3, 2135–2156

2020
[7]

McInroy and S

J. McInroy and S. Shpectorov,Axial algebras of Jordan and Monster type, London Math. Soc. Lecture Note Ser. 496, Cambridge University Press, 2025, 246–294

2025
[8]

Meyer and W

W. Meyer and W. Neutsch,Associative subalgebras of the Griess algebra, J. Algebra 158 (1993), no. 1, 1–17

1993
[9]

Norton,The Monster algebra: some new formulae, inMoonshine, the Monster, and Related Topics, Contemp

S. Norton,The Monster algebra: some new formulae, inMoonshine, the Monster, and Related Topics, Contemp. Math. 193, Amer. Math. Soc., Providence, RI, 1996, 297–306. 7

1996

[1] [1]

Castillo-Ramirez,Associative subalgebras of low-dimensional Majorana algebras, J

A. Castillo-Ramirez,Associative subalgebras of low-dimensional Majorana algebras, J. Algebra 421 (2015), 159–188

2015

[2] [2]

I. B. Gorshkov, J. McInroy, T. M. Mudziiri Shumba, and S. Shpectorov,Automorphism groups of axial algebras, J. Algebra 661 (2025), 657–712

2025

[3] [3]

Gorshkov and S

I. Gorshkov and S. Shpectorov,Axial Algebras: Questions and Conjectures, arXiv:2606.30048

Pith/arXiv arXiv

[4] [4]

J. I. Hall, F. Rehren, and S. Shpectorov,Universal axial algebras and a theorem of Sakuma, J. Algebra 421 (2015), 394–424

2015

[5] [5]

A. A. Ivanov,The Monster Group and Majorana Involutions, Cambridge Tracts in Math- ematics 176, Cambridge University Press, 2009

2009

[6] [6]

Khasraw, J

S. Khasraw, J. McInroy, and S. Shpectorov,On the structure of axial algebras, Trans. Amer. Math. Soc. 373 (2020), no. 3, 2135–2156

2020

[7] [7]

McInroy and S

J. McInroy and S. Shpectorov,Axial algebras of Jordan and Monster type, London Math. Soc. Lecture Note Ser. 496, Cambridge University Press, 2025, 246–294

2025

[8] [8]

Meyer and W

W. Meyer and W. Neutsch,Associative subalgebras of the Griess algebra, J. Algebra 158 (1993), no. 1, 1–17

1993

[9] [9]

Norton,The Monster algebra: some new formulae, inMoonshine, the Monster, and Related Topics, Contemp

S. Norton,The Monster algebra: some new formulae, inMoonshine, the Monster, and Related Topics, Contemp. Math. 193, Amer. Math. Soc., Providence, RI, 1996, 297–306. 7

1996

Fusion rules from the Norton inequality

Core claim §

What carries the argument

Load-bearing premise §

What would settle it §

discussion (0)

Core claim

Load-bearing premise

What would settle it