Weakly Noetherian Lie algebras and the Sierra-Walton Conjecture
Olivier Mathieu · 2026
Read Weakly Noetherian Lie algebras and the Sierra-Walton Conjecture by Olivier Mathieu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-18116.txt.
Introduction
Olivier Mathieu's 2026 paper defines a weak form of the Noetherian property for Lie algebras, spurred by the guess that enveloping algebras are Noetherian only in finite dimensions. His main theorem shows such algebras have a highly constrained structure, proved at length with results of Tits, Formanek, Razmyslov, Grabowski, and the author. For graded algebras he gives a full list of the perfect strictly weak cases. That list settles the guess for all perfect graded Lie algebras, bettering a celebrated result of Sierra and Walton.
Catalog record
| CREATOR | Olivier Mathieu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 136240 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.18116
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Weakly Noetherian Lie algebras and the Sierra-Walton Conjecture Thanks: 2020 MSC. 16S30, 17B35. N. A. was partially supported by the Secyt (UNC), Research partially supported by UMR 5028 du CNRS and by the Shenzhen International Center of Mathematics, SUSTech. Olivier Mathieu Abstract . Let K K be a field of characteristic zero. Motivated by the conjecture that an enveloping algebra U ( 𝔤 ) U(\mathfrak{g}) is Noetherian only if 𝔤 \mathfrak{g} is finite dimensional, we define the notion of weakly Noetherian Lie algebras. The main result, Theorem A, states that weakly Noetherian Lie algebras have a very constrained structure. In the specific case of ℤ n {\mathbb{Z}}^{n} graded Lie algebras, it implies an explicit classification of the perfect strictly weakly Noetherian Lie algebras, stated in Theorem B. The proofs of both theorems are quite long, and uses concrete results due to Tits, Formanek, Razmyslov, Grabowski and the author. The first theorem provides some insight on the desired conjecture. The second one implies the conjecture for all perfect ℤ n {\mathbb{Z}}^{n} -graded Lie algebras, improving a celebrated theorem of Sierra and Walton. 1991 Mathematics Subject Classification 17B35, 17B68, 16P99, 16R40 Introduction 0.1. General introduction Over a field K of characteristic 0, any finite dimensional Lie algebra \mathfrak{g} admits a Levi decomposition \mathfrak{g}\simeq\mathfrak{s}\ltimes\mathfrak{r} , where \mathfrak{s} is semisimple and \mathfrak{r} is a solvable ideal. Motivated by the conjecture mentionned in the abstract, we pose the very naive question: Do infinite dimensional Lie algebras admit a structure theory similar to the finite dimensional case? In general the answer is obviously negative, but we can expect a positive answer for some classes of infinite dimensional Lie algebras. For example, Aldosray, Amayo and Stewart [ 1 , 2 , 3 , 32 ] has proved some general results for Lie algebras satisfying some Artinian-Noetherian conditions. Here we try a different approach: we say that a Lie algera \mathfrak{G} is weakly Noetherian if (a) for any subalgebra \mathfrak{m}\subset\mathfrak{G} , the space \mathfrak{m}/[\mathfrak{m},\mathfrak{m}] is finite dimensional. If, in addition (b) any chain of characteristic ideals in \mathfrak{G} stabilizes, \mathfrak{G} is called Noetherian . Moreover \mathfrak{G} is called strictly weakly Noetherian if \mathfrak{G}\otimes L is weakly Noetherian, for any finite extension L of K . Typical examples of Noetherian Lie algebras are the Krichever-Novikov Lie algebras \mathrm{Vect}_{X} of vector fields on a smooth algebraic curve X , as it is proved in Section 4 . At first glance, the notion of weakly Noetherian Lie algebras looks like an abstract nonsense notion. In fact, we will see that it is connected with interesting concrete algebraic theories, as explained in Subsection 0.3 . First we consider the case of {\mathbb{Z}}^{n} -graded Lie algebras, \mathcal{L}=\oplus_{{\bf m}\in{\mathbb{Z}}^{n}}\,\mathcal{L}_{\bf m} , where is is assumed that all homogenous components have finite dimensions. Given a finite extension L of K , we define \displaystyle\mathrm{Witt}(L) \displaystyle=\operatorname{Der}\,L[t] \displaystyle\mathrm{Vir}(L) \displaystyle=\operatorname{Der}\,L[t,t^{-1}], and we denote as \widehat{\mathrm{Vir}}(L) the Virasoro algebra, that is the universal central extension of \mathrm{Vir}(L) . For perfect {\mathbb{Z}}^{n} -graded Lie algebra, there is an explicit classification, up to a finite dimensional factor. Theorem B . Let \mathcal{L} be a perfect strictly weakly Noetherian {\mathbb{Z}}^{n} -graded Lie algebra and let \hat{\mathcal{L}} be its its universal central extension. Then we have \displaystyle\widehat{\mathcal{L}}\simeq\mathfrak{g}\,\oplus\,\big[\oplus_{i=1}^{n}\mathrm{Witt}(E_{i})\big]\,\oplus\,\big[\oplus_{j=1}^{m}\widehat{\mathrm{Vir}}(F_{j})\big] where \mathfrak{g}\simeq\mathfrak{s}\ltimes\mathfrak{r} is a (perfect) finite dimensional Lie algebra, n , m are integers and E_{1},\ldots,E_{n},F_{1},\ldots,F_{m} are finite extensions of K . In order to state the structure result in general we first need some additional definitions. Recall that the radical \operatorname{rad}(\mathfrak{G}) of a Lie algebra \mathfrak{G} is the intersection of all maximal ideals \mathfrak{m} of codimension >1 . Assume now that \mathfrak{G} be a weakly Noetherian Lie algebra satisfying the following simplifying hypothesis (h1) \mathfrak{G}=[\mathfrak{G},\mathfrak{G}] , and (h2) \mathfrak{G} does not admit finite dimensional simple quotients. Then we define, by transfinite induction, a decreasing sequence, indexed by ordinals \alpha , of Lie subalgebras \mathfrak{G}=\mathfrak{G}_{(0)}\supset\mathfrak{G}_{(1)}\supset\ldots\mathfrak{G}_{(\alpha)}\supset\ldots as follows: (a) if the ordinal \alpha is a successor, that is \alpha=\beta+1 for some \beta , then \mathfrak{G}_{(\alpha)}=[\operatorname{rad}(\mathfrak{G}_{(\beta)}),\operatorname{rad}(\mathfrak{G}_{(\beta)})] (b) If \alpha>0 is a limit ordinal, then \mathfrak{G}_{(\alpha)}=[\cap_{\beta<\alpha}\,\operatorname{rad}(\mathfrak{G}_{(\beta)}),\cap_{\beta<\alpha}\,\operatorname{rad}(\mathfrak{G}_{(\beta)})]. In its simplified form, the main result is as follows: Theorem A . (simplified version) Let \mathfrak{G} be a weakly Noetherian Lie algebra satisfying (h1) and (h2). Then, for any ordinal \alpha : (a) the Lie algebra \mathfrak{G}_{(\alpha)} is a perfect ideal, (b) \mathfrak{G}_{(\alpha)}/\mathfrak{G}_{(\alpha+1)} is a central extension of \mathfrak{G}_{(\alpha)}/\operatorname{rad}\mathfrak{G}_{(\alpha)} by a finite dimensional center, (c) any simple quotient of \mathfrak{G}_{(\alpha)} is infinite dimensional, and (d) for \alpha\geq 1 , no Krichever-Novikov algebras occur as a quotient of \mathfrak{G}_{(\alpha)} . Moreover if \mathfrak{G} is Noetherian, then \mathfrak{G}_{(\alpha)}=0 for \alpha big enough. Therefore, Theorem A shows that Noetherian Lie algebras \mathfrak{G} have a very constrained structure. Roughly speaking, \mathfrak{G} is a "tower" of central extensions of infinite dimensional simple Lie algebras. 0.2. Relation with the Sierra-Walton conjecture Our investigation is motivated by the following: Sierra-Walton Conjecture 1 . [ 31 ] If \mathfrak{g} is infinite dimensional, then U(\mathfrak{g}) is not Noetherian. This question has been stated by several authors since the 70’s, among them R. K. Amayo, I. Stewart, J. Dixmier, V. Latishev, K. A. Brown [ 3 , 13 ] A breakthrough result was obtained in 2013 by S. Sierra and Ch. Walton: Theorem 2 . [ 31 ] The enveloping algebra of the Witt algebra W(K) is not Noetherian. Consequently, S. Sierra and Ch. Walton stated this question as a conjecture. From Sierra-Walton Theorem, one can derived the same result for all Krichever-Novikov algebras [ 14 ] . The conjecture was confirmed recently for some other classes of Lie algebras [ 7 , 8 ] and [ 5 ] . The old work [ 35 ] is also connected with these questions. As a consequence of Theorem B, we obtain the following generalization of Sierra-Walton Theorem. Corollary 65 . Let \mathcal{L} be a perfect {\mathbb{Z}}^{n} -graded Lie algebra. The algebra U(\mathcal{L}) is left Noetherian only if \mathcal{L} has finite dimension. A refined version of Theorem A is proved in section 8 . As a consequence we deduce: Corollary 45 . Any infinite dimensional Noetherian Lie algebra \mathfrak{G} contains two characteristic ideals \mathfrak{p}\supset\mathfrak{q} with \displaystyle\mathrm{codim} \displaystyle\,\mathfrak{p}<\infty \displaystyle\mathrm{codim}\,\mathfrak{q}=\infty such that the Lie algebra \mathfrak{p}/\mathfrak{q} is just-infinite and either (type A) \mathfrak{p}/\mathfrak{q} is simple, or (type B) \mathfrak{p}/\mathfrak{q} is residually nilpotent, or (type C) \mathfrak{p}/\mathfrak{q} is perfect, residually semi-simple and does not satisfies any polynomial identity. By [ 18 ] , any Krichever-Novikov algebra \mathrm{Vect}_{X} is simple, that is of type A. Given any point P on the curve X , the Lie subalgebra \mathrm{Vect}_{X}^{(2)} of vector fields vanishing twice at P is residually nilpotent, that is of type B. However, we do not know examples of Noetherian Lie algebras of type C . It follows from Corollary 45 that the Sierra-Walton conjecture can be reduced to these three types of Lie algebras. Hovewer Theorem A and some intuitive arguments, briefly outlined in Subsection 9.2 , suggest that the conjecture could be undecidable for some simple Lie algebras \mathfrak{S} . 0.3. The ideas of the proofs The proof of Theorem A is quite long. Although its statement looks formal, its proof is based on concrete results. The core of the proof consists of the following result: Theorem 41 . Let \mathfrak{G} be a weakly