Composition algebras in symmetric tensor categories

Alberto Elduque; Javier R\'andez-Ib\'a\~nez

REVIEW 1 major objections 6 minor 23 references

Composition algebras in symmetric tensor categories

T0 review · 1 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper defines composition algebras in symmetric tensor categories and classifies all unital ones in the characteristic-2 category $\mathrm{Ver}_4^+$, proving a dichotomy between explicit 2- and 4-dimensional families and ordinary…

desk verdict Solid first classification of composition algebras in Ver_4^+; the imported case-split theorem is a normal citation, and the only real issue is the size of the finite check in Theorem 5.13. read the letter →

arxiv 2608.04668 v1 pith:YEGDMJZ7 submitted 2026-08-05 math.RA math.QA

Alberto Elduque , Javier Rández-Ibáñez This is my paper · ORCID

classification math.RAmath.QA MSC 17A7518M05

keywords compositionalgebrassymmetrictensorcategoriescharacteristic2Ver_4^+categoryquadraticformsFrobeniustwistunitalclassification

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

This paper defines composition algebras in symmetric tensor categories, paying special attention to characteristic 2, and then classifies all unital composition algebras in the category $\mathrm{Ver}_4^+$, the counterpart to vector superspaces in characteristic 2. The classification splits on the Frobenius twist $U^{(1)}$: when the twist is trivial, every unital composition algebra is one of the explicit two-dimensional algebras $U(\lambda)$ or four-dimensional algebras $U(\lambda,\nu)$ described in Theorem 5.13; when the twist is nontrivial, the twisting derivation $t$ must vanish, so the algebra is an ordinary unital composition algebra over the field. The result matters because composition algebras sit behind exceptional Lie and Jordan algebras, and $\mathrm{Ver}_4^+$ is the correct category-theoretic home for structures that replace super geometry in characteristic 2. The paper also checks that its new definition reproduces the classical classifications in $\mathrm{Vec}$ and $\mathrm{sVec}$, with the expected characteristic-2 subtlety that quadratic forms, rather than symmetric bilinear forms, carry the norm.

What carries the argument

The load-bearing mechanism is the decomposition of $U \otimes U$ under the braiding $c_{U,U}$. In characteristic 2 the quadratic form $Q$ lives on $\Gamma^2(U)$, the kernel of $\mathrm{id} - c_{U,U}$, and the morphism $\rho_U$ determined by $\iota_U \rho_U = \mathrm{id} + c_{U,U}$ turns $Q$ into the symmetric bilinear form $B$. The Frobenius twist $U^{(1)}$ is defined as the image of the composite $\pi_U \iota_U$, and $\rho_U$ is surjective exactly when $U^{(1)} = 0$. In $\mathrm{Ver}_4^+$ the braiding is $u \otimes v \mapsto v \otimes u + t(v) \otimes t(u)$ for a square-zero derivation $t$, so multiplicativity of the norm unfolds into the master identity (5.5), whose specializations are the linearized identities (5.6)--(5.10). The classification then runs on a nondegenerate associative bilinear form $\beta(xy) = B(v, xy)$ on $X = \operatorname{im} t$ obtained after choosing $v$ with $t(v) = 1$; nondegeneracy of $\beta$ forces $\dim X \le 2$, giving the explicit tables, while in the nontrivial-twist case the identities force $t = 0$ by constructing a 6-dimensional ordinary composition subalgebra, contradicting the classical bound.

What would settle it

A concrete test: over an algebraically closed field of characteristic 2, look for a unital composition algebra in $\mathrm{Ver}_4^+$ with dimension 8 and $\ker t \neq \operatorname{im} t$; the theorem says no such algebra exists. Alternatively, in the trivial-twist case, attempt to construct an example with $\dim \operatorname{im} t = 3$; Theorem 5.12 forbids it. A direct check of the imported equivalence $U^{(1)} = 0 \iff \ker t = \operatorname{im} t$ on the underlying objects of the explicit families would also settle whether the case split is sound.

Watch

Extended reading notes

Core claim

The central discovery is a dichotomy for unital composition algebras in $\mathrm{Ver}_4^+$ over a field of characteristic 2. If the Frobenius twist is trivial, equivalently $\ker t = \operatorname{im} t$ for the square-zero derivation $t$ that enters the braiding, then the underlying object has dimension 2 or 4: the dimension-2 algebras are the $U(\lambda)$ with a basis $1,v$ and $v^2 = \lambda 1$, and the dimension-4 algebras are the $U(\lambda,\nu)$ with basis $1,v,x,vx$ and the multiplication table in Theorem 5.13. The isomorphism classes are controlled by $\lambda$ modulo $F^2$ and $\nu$ modulo $(F^\times)^2$. If the Frobenius twist is nontrivial, then $t = 0$, so the algebra is a usual unital composition algebra in $\mathrm{Vec}$ and therefore appears in the classical list of dimensions 1, 2, 4, and 8. Thus the only new characteristic-2 phenomena are the two explicit families, and over an algebraically closed field only $U(0)$ and $U(0,1)$ survive.

Load-bearing premise

The load-bearing premise is an imported theorem from the companion paper on quadratic forms: for objects of $\mathrm{Ver}_4^+$, the Frobenius twist is trivial exactly when $\ker t = \operatorname{im} t$, and both classification theorems partition on this dichotomy, so if that equivalence failed the listed algebras would not exhaust the possibilities.

Editorial extensions

If this is right

Over an algebraically closed field of characteristic 2, there are exactly two unital composition algebras in $\mathrm{Ver}_4^+$ with trivial Frobenius twist: $U(0)$ and $U(0,1)$.
Any unital composition algebra in $\mathrm{Ver}_4^+$ of dimension greater than 4 must have $t = 0$ and be an ordinary composition algebra in $\mathrm{Vec}$, so the classical dimension bound of 1, 2, 4, or 8 persists.
For every such algebra, $\ker t$ is a conic alternative algebra with $\operatorname{im} t$ as its radical, and in the trivial-twist case the dimension of $\operatorname{im} t$ is at most 2.
The norm does not determine the algebra: nonisomorphic algebras $U(\lambda)$ can have isometric quadratic forms, a behavior not seen for unital composition algebras in $\mathrm{Vec}$.

Reading between the lines

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

If the classification is right, characteristic-2 'super' composition algebras are much poorer than in characteristic 3: the only exotic unital examples are 2- and 4-dimensional, and the 8-dimensional octonion algebra appears only as an ordinary $\mathrm{Vec}$ algebra, not in a genuinely twisted form.
The invariant $\lambda \bmod F^2$ for $U(\lambda)$ suggests reading these families as twisted forms of the dual numbers, so the classification could be reformulated in Galois-cohomological terms as a statement about forms of $U(0)$ and $U(0,1)$; that reformulation is not in the paper.
A testable extension is to apply the same $\beta$-form technique to other characteristic-2 symmetric tensor categories whose braiding is a first-order deformation, predicting that unital composition algebras there will again be either small or classical.
One could try to prove the equivalence $\ker t = \operatorname{im} t \iff U^{(1)} = 0$ directly for the specific objects underlying $U(\lambda)$ and $U(\lambda,\nu)$, which would make the classification independent of the imported result from the companion paper.

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.

Desk Editor's Note

read the letter

Dear colleague,

The main news: this paper gives the first classification of unital composition algebras in Ver_4^+ in characteristic 2, and it introduces a definition of composition algebras in symmetric tensor categories that actually handles characteristic 2. I think the main theorems are correct and the paper deserves peer review.

What is genuinely new: Definition 3.2, where a quadratic form is Q: Γ2(U)→1 and multiplicativity is Q∘μ~ = Q⊗Q, is more careful than the existing characteristic-not-2 treatment in [DES24]. The partial linearizations in Proposition 3.4 are useful, and the graphical proof of Theorem 3.1 is clean. The actual classification in Section 5 is new: Theorem 5.13 gives the trivial-Frobenius-twist cases, the 2-dimensional family U(λ) and the 4-dimensional family U(λ,ν), with precise isomorphism conditions; Theorem 5.16 shows that nontrivial twist forces t=0 and hence the algebra is an ordinary composition algebra in Vec. The explicit multiplication tables and norm values make the results reproducible.

I do not share the reader's strongest worry about the imported equivalence U^(1)=0 iff kert=imt from [CKP26]. That is exactly what citations are for; every classification leans on prior structure theory. The authors state the dependence clearly in Section 5.2. If the cited result is correct, the case split is exhaustive, and a referee can check the companion paper. I would still ask the authors to quote the precise proposition or give a short proof sketch, because the whole classification hinges on that dichotomy.

The genuine soft spots are modest. The existence check in Theorem 5.13 is delegated to "by inspection" of an 8×8 multiplication table. The table is actually given and the check is finite, but it is tedious; a referee will want either a cleaner structural argument or an appendix spelling out the verification. Section 4 is mostly review, which is fine but should stay clearly labeled as context.

The nontrivial-twist proof in Section 5.3 is intricate, but I checked the steps: the contradiction builds a 6-dimensional unital composition algebra in Vec with nondegenerate bilinear form, contradicting Hurwitz's theorem. The reasoning hangs together, and I see no circularity. Self-citations appear mainly in review sections and are not load-bearing.

Bottom line: this is a solid, carefully written paper for people working on algebras in positive-characteristic tensor categories. It deserves a serious referee. I would send it to review and ask the authors to clarify the [CKP26] dependence and expand the finite check.

Referee Report

1 major / 6 minor

Summary. The paper defines composition algebras in symmetric tensor categories, with special attention to characteristic 2, where the notion of quadratic form is more subtle than in ordinary vector spaces. After deriving the relevant partial and full linearizations of the multiplicative norm property, the authors review the classical classifications in Vec and in sVec, and then classify unital composition algebras in the Verlinde category Ver_4^+ over a field of characteristic 2. The main classification is split according to whether the Frobenius twist is trivial: when it is trivial, the algebras are exactly the two-dimensional algebras U(λ) and the four-dimensional algebras U(λ,ν) described in Theorem 5.13; when it is nontrivial, the action of t is trivial and the algebra is an ordinary composition algebra in Vec (Theorem 5.16). The proofs are detailed and largely self-contained, building on the structure of kert and explicit computations.

Significance. If the results are correct, this is a substantial contribution to nonassociative algebra in tensor categories: it gives the first classification of unital composition algebras in Ver_4^+, the characteristic-2 counterpart of vector superspaces, and it provides explicit families with concrete multiplication tables and isomorphism invariants. The categorical framework introduced here is natural and should be useful for further work on compositions algebras and related structures in positive-characteristic tensor categories. The paper is also careful about the characteristic-2 pitfalls and the distinctions between various notions of quadratic form. The authors provide explicit finite verifications, detailed structural lemmas, and a coherent case split, and they correctly cite and use previous work on quadratic forms in Ver_4^+.

major comments (1)

[Section 5.2 (and 5.3)] The exhaustiveness of the classification depends on the equivalence U^(1)=0 iff kert=imt, which is imported from [CKP26, Proposition 3.4 and Corollary 3.5] rather than proved here. This equivalence is load-bearing: Theorem 5.13 is proved only under kert=imt, and Theorem 5.16 is proved only under kert≠imt, so without the imported dichotomy the two cases do not provably cover all unital composition algebras in Ver_4^+. The authors should either state the exact result from [CKP26] with its hypotheses and verify that those hypotheses apply to the objects arising from unital composition algebras, or include a proof of the equivalence in this paper. As written, a referee cannot check the central case-split from the present manuscript alone.

minor comments (6)

[Theorem 5.13] The verification that the displayed multiplication table defines a composition algebra in Ver_4^+ is summarized as 'by inspection.' Although the table is explicit and the row-wise Q-value check is described, the authors should spell out the verification for the rows with Q(A)=1, or provide a supplementary computation, so that the existence of the four-dimensional family is fully machine-checkable.
[Theorem 5.13] The isomorphism criterion is justified only by the phrase 'follows from the invariance of the subalgebra S and of the subspace X_0.' This is rather terse; please expand the argument, in particular explaining why λ is determined modulo F^2 and ν is determined modulo (F^×)^2 under isomorphisms of algebras in Ver_4^+.
[Section 4.2] The text mentions 'B(1,2) and B(2,4)' after Theorem 4.2 has listed 'B(1,2) or B(4,2)'; please correct the inconsistency.
[Corollary 5.15] The phrase 'algebraically field' should be 'algebraically closed field.'
[Introduction and throughout] There are several typographical errors, such as 'compostion algebras' in the introduction and 'restricition' in the proof of Theorem 5.16; a careful proofreading pass is recommended.
[Theorem 5.16, Step 3] The equality imt∩U_12=t(U_12) is used without comment; a short justification (using that T⊂kert and t preserves the Peirce components) would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the new Ver_4^+ classification is derived from the category axioms and the multiplicative norm; the load-bearing imported equivalence from [CKP26] is external and is not the target classification.

full rationale

The paper's central claim is a classification theorem for unital composition algebras in Ver_4^+, obtained by fixing a unital algebra (U, μ, Q, 1) in the category and deriving structural equations (5.2)-(5.5), (5.6), (5.7), (5.10) from Definition 3.2. No parameter is fitted to data and no closely related quantity is 'predicted' from a fitted subset. The only non-self-contained input is the case-split equivalence U^(1)=0 iff kert=imt, quoted from [CKP26, Proposition 3.4 and Corollary 3.5] at the beginning of Section 5.2 and used to partition the classification into Theorems 5.13 and 5.16. That is a genuine external dependence, and if the imported theorem were false the classification would not be exhaustive; however, this is reliance on an independent prior theorem, not circularity: [CKP26] is not by the present authors, is not merely a citation of the present paper's own conclusions, and is not the classification of Ver_4^+ unital composition algebras being proved. The self-citations [DES24] and [EO02] are confined to introductory motivation and to the review of superalgebra classifications in Section 4.2; the final contradiction in Theorem 5.16 uses Theorem 4.1, attributed to [ZSSS82], and the final verification in Theorem 5.13 is an explicit finite multiplication-table check. The derivation is therefore self-contained modulo the stated external equivalence, and no claim reduces by definition to its own input.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper's central claim rests on the framework of quadratic forms and the Frobenius twist from [CKP26], the identification of Ver_4^+ as F[t]-modules from [Ven16], and the classical Hurwitz theorem. None of these are re-derived, but they are standard or from companion papers. No parameters are fitted to make the classification work; λ and ν are honest invariants of the classified algebras.

assumptions (6)

domain assumption Ver_4^+ is the category of finite-dimensional modules over the Hopf algebra F[t] with t^2=0, t primitive, braiding from R=1⊗1+t⊗t.
Used throughout Section 5; this is the definition of the category, following [Ven16] and [EGNO15, §8.3].
domain assumption A quadratic form in a symmetric tensor category is a morphism Q:Γ2(U)→1 with associated bilinear form B=Q∘ρU, equivalently the definition in [CKP26, Def. 2.5].
Definition 2.1 and Remark 2.6; the paper builds on this framework without reproving it.
domain assumption In characteristic 2, trivial Frobenius twist is equivalent to kert=imt for objects of Ver_4^+ ([CKP26, Prop. 3.4, Cor. 3.5]).
Invoked at the start of Section 5.2 to split the classification into two cases.
standard math Hurwitz theorem: unital composition algebras in Vec have dimension 1, 2, 4, or 8 (Theorem 4.1, from [ZSSS82]).
Used in Theorem 5.16 to obtain a contradiction from a 6-dimensional subalgebra, and in Theorem 4.1 to recall the classical classification.
standard math Classification of unital composition superalgebras in [EO02] (Theorem 4.2).
Used in Section 4.2 to compare the new definition with the existing superalgebra classification, including the B(1,2) and B(4,2) cases in characteristic 3.
standard math A finite-dimensional associative algebra over a field of characteristic 2 in which x^2=0 for every element is nilpotent.
Used in Theorem 5.16, Step 1, to show imt is nilpotent.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Composition algebras in symmetric tensor categories." pith.science (2026). https://pith.science/paper/YEGDMJZ7

@misc{pith2026260804668,
  author       = {Pith},
  title        = {Pith review of: Composition algebras in symmetric tensor categories},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YEGDMJZ7}},
  note         = {Machine review of arXiv:2608.04668}
}

abstract

Composition algebras in symmetric tensor categories are defined. The classical results on standard unital composition algebras, as well as the classification of the unital composition superalgebras, are reviewed in light of this definition, and the unital composition algebras in the category $Ver_4^+$ (the counterpart to the category of vector superspaces in characteristic 2) are classified.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 22 canonical work pages

[1]

Angiono, J

I. Angiono, J. Plavnik, and G. Sanmarco, Semisimplification of contragredient Lie algebras, Orbita Math. 2 (2025), no. 1, 1--32

work page 2025
[2]

Chen, A.S

I. Chen, A.S. Kannan, and K. Pothapragada, Classification of non-degenerate symmetric bilinear and quadratic forms in the Verlinde category _4^+ , J. Algebra 686 (2026), 220--262

work page 2026
[3]

Baez, The octonions, Bull

J.C. Baez, The octonions, Bull. Amer. Math. Soc. (N.S.) 39 (2002), no. 2, 145--205

work page 2002
[4]

Benkart and A

G. Benkart and A. Elduque, A new construction of the Kac Jordan superalgebra, Proc. Amer. Math. Soc. 130 (2002), no. 11, 3209--3217

work page 2002
[5]

Benson and J

D. Benson and J. Pevtsova, Group schemes and their Lie algebras over a symmetric tensor category, preprint arXiv:2507.02031v1

arXiv
[6]

Cunha and A

I. Cunha and A. Elduque, An extended Freudenthal magic square in characteristic 3 , J. Algebra 317 (2007), no. 2, 471--509

work page 2007
[7]

Daza-Garc \' a, A

A. Daza-Garc \' a, A. Elduque, and U. Sayin, From octonions to composition superalgebras via tensor categories, Rev. Mat. Iberoam. 40 (2024), no. 1, 129--152

work page 2024
[8]

Deligne, Cat\'egories tannakiennes

P. Deligne, Cat\'egories tannakiennes. pp. 111–195 in The Grothendieck Festschrift, II, edited by P. Cartier et al., Progr. Math. 87, Birkhäuser, Boston, 1990

work page 1990

Show all 23 references

[9]

Deligne, Cat\'egories tensorielles, Mosc

P. Deligne, Cat\'egories tensorielles, Mosc. Math. J. 2:2 (2002), 227--248

2002
[10]

Elduque, Composition algebras, in Algebra and applications 1---Non-associative Algebras and Categories , 27--57, ISTE, London 2020

A. Elduque, Composition algebras, in Algebra and applications 1---Non-associative Algebras and Categories , 27--57, ISTE, London 2020

2020
[11]

Elduque, P.I

A. Elduque, P.I. Etingof, and A.S.Kannan, From the Albert algebra to Kac's ten-dimensional Jordan superalgebra via tensor categories in characteristic 5 , J. Algebra 666 (2025), 387--414

2025
[12]

Elduque and S

A. Elduque and S. Okubo, Composition superalgebras, Commun. Algebra 30 (2002), no. 11, 5447--5471

2002
[13]

Etingof, S

P. Etingof, S. Gelaki, D. Nikshych, and V. Ostrik, Tensor Categories, Mathematical Surveys and Monographs 205, American Mathematical Society, 2015

2015
[14]

Etingof and V

P. Etingof and V. Ostrik, On semisimplification of tensor categories, Representation Theory and Algebraic Geometry: A Conference Celebrating the Birthdays of Sasha Beilinson and Victor Ginzburg; Springer, 2021, pp. 3--35

2021
[15]

Garibaldi, H.P

R.S. Garibaldi, H.P. Petersson, and M.L. Racine, Albert algebras over commutative rings---the last frontier of Jordan systems, New Mathematical Monographs 48, Cambridge Univ. Press, Cambridge, 2024

2024
[16]

Hogben and V.G

L. Hogben and V.G. Kac, Erratum: Classification of simple -graded Lie superalgebras and simple Jordan superalgebras, Comm. Algebra 5 (1977), 1375--1400 , Comm. Algebra 11 (1983), 1155--1156

1977
[17]

Hu, Lie algebras in _4^+ , J

S. Hu, Lie algebras in _4^+ , J. Algebra 689 (2026), 112--158

2026
[18]

Kac, Classification of -graded Lie superalgebras and simple Jordan superalgebras, Comm

V.G. Kac, Classification of -graded Lie superalgebras and simple Jordan superalgebras, Comm. Algebra 5 (1977), 1375--1400

1977
[19]

Kannan, New constructions of exceptional simple Lie superalgebras with integer Cartan matrix in characteristics 3 and 5 via tensor categories, Transform

A.S. Kannan, New constructions of exceptional simple Lie superalgebras with integer Cartan matrix in characteristics 3 and 5 via tensor categories, Transform. Groups 29 (2024), no. 3, 1065--1103

2024
[20]

M.A. Knus, A. Merkurjev, M. Rost, and J.P. Tignol, The book of involutions. American Mathematical Society Colloquium Publications 44, American Mathematical Society, Providence, RI, 1998

1998
[21]

McCrimmon, Nonassociative algebras with scalar involution, Pacific J

K. McCrimmon, Nonassociative algebras with scalar involution, Pacific J. Math. 116 (1985), no. 1, 85--109

1985
[22]

Venkatesh, Hilbert basis theorem and finite generation of invariants in symmetric tensor categories in positive characteristic, Int

S. Venkatesh, Hilbert basis theorem and finite generation of invariants in symmetric tensor categories in positive characteristic, Int. Math. Res. Not. IMRN 2016, no. 16, 5106--5133

2016
[23]

Zhevlakov, A.M

K.A. Zhevlakov, A.M. Slin'ko, I.P. Shestakov, and A.I. Shirshov, Rings that are nearly associative. Academic Press, New York-London, 1982

1982

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Angiono, J

I. Angiono, J. Plavnik, and G. Sanmarco, Semisimplification of contragredient Lie algebras, Orbita Math. 2 (2025), no. 1, 1--32

work page 2025

[2] [2]

Chen, A.S

I. Chen, A.S. Kannan, and K. Pothapragada, Classification of non-degenerate symmetric bilinear and quadratic forms in the Verlinde category _4^+ , J. Algebra 686 (2026), 220--262

work page 2026

[3] [3]

Baez, The octonions, Bull

J.C. Baez, The octonions, Bull. Amer. Math. Soc. (N.S.) 39 (2002), no. 2, 145--205

work page 2002

[4] [4]

Benkart and A

G. Benkart and A. Elduque, A new construction of the Kac Jordan superalgebra, Proc. Amer. Math. Soc. 130 (2002), no. 11, 3209--3217

work page 2002

[5] [5]

Benson and J

D. Benson and J. Pevtsova, Group schemes and their Lie algebras over a symmetric tensor category, preprint arXiv:2507.02031v1

arXiv

[6] [6]

Cunha and A

I. Cunha and A. Elduque, An extended Freudenthal magic square in characteristic 3 , J. Algebra 317 (2007), no. 2, 471--509

work page 2007

[7] [7]

Daza-Garc \' a, A

A. Daza-Garc \' a, A. Elduque, and U. Sayin, From octonions to composition superalgebras via tensor categories, Rev. Mat. Iberoam. 40 (2024), no. 1, 129--152

work page 2024

[8] [8]

Deligne, Cat\'egories tannakiennes

P. Deligne, Cat\'egories tannakiennes. pp. 111–195 in The Grothendieck Festschrift, II, edited by P. Cartier et al., Progr. Math. 87, Birkhäuser, Boston, 1990

work page 1990