The projective cover of the trivial module in characteristic 11 11 for the sporadic simple Janko group J 4 J_{4} revisited

2025

Read The projective cover of the trivial module in characteristic 11 11 for the sporadic simple Janko group J 4 J_{4} revisited (2025) 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-2511-00939.txt.

Introduction

Jurgen Muller revisits the projective cover of the trivial module for the sporadic simple Janko group J4 in characteristic eleven. The paper is a sequel to his earlier preprint, where the modular projective indecomposable summands of a permutation character for J4 were determined up to one remaining parameter. Here that parameter is fixed, using a new condensation method for induced modules that enumerates long orbits to make the computation feasible. The result completes the identification of this basic building block in the modular representation theory of the group.

Catalog record

YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-00939

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The projective cover of the trivial module in characteristic 11 11 for the sporadic simple Janko group J 4 J_{4} revisited Abstract This is a sequel to [ 9 ] , where we have determined the 11 11 -modular projective indecomposable summands of the permutation character of J 4 J_{4} on the cosets of an 11 ′ 11^{\prime} -subgroup of maximal order, amongst them the projective cover of the trivial module, up to a certain parameter. Here, we fix this parameter, by applying a new condensation method for induced modules which uses enumeration techniques for long orbits. Mathematics Subject Classification: 20C20, 20C34. Keywords: Sporadic simple Janko group, ordinary characters, Brauer characters, decomposition numbers, projective modules, permutation modules, endomorphism algebras, orbit enumeration, condensation. Jürgen Müller 1 Introduction The present article is a sequel to [ 9 ] , which is devoted to answering (to the negative) the question (posed in [ 8 ] ) whether the projective cover of the trivial module in characteristic 11 for the largest sporadic simple Janko group G:=J_{4} is a permutation module. Actually, this question boils down to determine whether or not the projective permutation character 1_{H}^{G} , where H<G is a maximal ( 11^{\prime} -)subgroup of shape H\cong 2^{10}\colon L_{5}(2) , is projective indecomposable. According to the ModularAtlasHomepage , virtually nothing is known about the decomposition numbers of G is characteristic 11 . Thus the strategy in [ 9 ] was to find the decomposition of 1_{H}^{G} into projective indecomposable characters. It turned out that there are four distinct indecomposable summands, which are reproduced in Table 1 (on page 1 ). This can be seen as the first step towards the ambitious goal of determining the 11 -modular decomposition matrix of G , which is particularly compelling as G has trivial-intersection Sylow 11 -subgroups. Alone, in [ 9 ] , we have not been able to fix the parameter a\in\{0,1\} appearing in Table 1 . But this should be done before proceeding to find more decomposition numbers of G . Hence the purpose of the present article is to close this gap, by showing that we actually have a=0 , obeying to the conventional choices for decomposition maps made in the ModularAtlas . To this end, we invoke a maximal subgroup U<G of shape U\cong U_{3}(11)\colon 2 , whose 11 -modular decomposition matrix is well-known. In order to relate the decomposition matrices of U and G , it turns out that subtle details of the embedding of U into G play a crucial role here. These can be captured by a consideration of the automorphism group of the ordinary character table of G . Having this in place, letting \mathcal{O} be a G -set affording the permutation character 1_{H}^{G} , and letting \mathbb{F} be a field of characteristic 11 , we examine how the restriction of the permutation module \mathbb{F}[\mathcal{O}] to U decomposes into projective indecomposable modules. To do so, we are finally led to consider the action of \text{End}_{\mathbb{F}[G]}(\mathbb{F}[\mathcal{O}]) on \text{Hom}_{\mathbb{F}[U]}(\mathbb{F}[\mathcal{O}]|_{U},V)\cong\text{Hom}_{\mathbb{F}[G]}(\mathbb{F}[\mathcal{O}],V^{G}) , for certain simple \mathbb{F}[U] -modules V . The latter step essentially amounts to computing the ‘condensed module’ afforded by the module V^{G} induced from the subgroup U , with respect to the ‘condensation subgroup’ H . (As a general reference for ‘condensation’, see for example [ 10 ] .) A technique to compute condensed modules of induced modules, for subgroups U of smallish index in G , has been developed in [ 13 ] . The present approach, combining these ideas with the orbit enumeration techniques available in the GAP package ORB , now allows both subgroups U and H to have large index in G . We expect this to be of independent interest. The present article is organized as follows: In the rest of Section 1 we indicate the computational tools we are using, and we sketch the ideas behind the orbit enumeration techniques available in ORB . In Section 2 we present the piece of theory underlying the new condensation technique advertised above. In Section 3 we collect some character theoretic facts on G and various of its subgroups, and we consider table automorphisms, to clarify where and where not choices can be made. In Section 4 we enumerate \mathcal{O} by U -orbits, by applying ORB . In Section 5 , using an idea inspired by [ 6 ] , we examine certain condensed induced modules, in order to finally determine the missing parameter. Computational tools. To facilitate group theoretic and character theoretic computations we use the computer algebra system GAP [ 3 ] , and its comprehensive database CTblLib [ 1 ] of ordinary and Brauer character tables. In particular, [ 1 ] encompasses the data given in the Atlas [ 2 ] and the ModularAtlas [ 4 ] , as well as the additional data collected on the ModularAtlasHomepage [ 17 ] . Data concerning explicit permutation representations, ordinary and modular matrix representations, and the embedding of (maximal) subgroups of sporadic simple groups is available in the AtlasOfGroupRepresentations [ 18 ] , and through the GAP package AtlasRep [ 19 ] . To compute with matrix representations over finite fields we use the MeatAxe [ 14 , 15 ] and its extensions described in [ 6 , 7 ] . Enumerating long orbits. As our computational workhorse to facilitate computations with (large) permutation representations we use the GAP package ORB [ 11 ] , whose orbit enumeration techniques are comprehensively described in [ 10 , 12 ] . For convenience, we give a brief sketch of the approach: Let G be a (large) finite group, and let \mathcal{O} be a (large) transitive G -set, which we assume to be implicitly given, for example as a G -orbit of a vector v_{1} in an F[G] -module V over a finite field F . Letting H\leq G be a (still large) subgroup, we are interested in finding the H -orbits \mathcal{O}_{j}\subseteq\mathcal{O} , their length n_{j} , representatives v_{j}\in\mathcal{O}_{j} , elements g_{j}\in G such that v_{1}\cdot g_{j}=v_{j} , and the point stabilizers H_{j}=\text{Stab}_{H}(v_{j}) . To achieve this, we assume to be able to compute efficiently within H (but not within G ), for example by having a (smallish) faithful permutation representation of H at hand. To find the H -orbits \mathcal{O}_{j} , we choose a (smallish) helper subgroup K\leq H , and enumerate the various \mathcal{O}_{j} by the K -orbits they contain. To do so, we choose a (not too small) helper K -set \mathcal{Q} together with a homomorphism \pi_{K}\colon\mathcal{O}\rightarrow\mathcal{Q} of K -sets, which again we assume to be implicitly given, for example by an F[K] -quotient module of V . We assume that K has sufficiently long orbits in \mathcal{Q} , and that we are able to classify them, by giving representatives, their point stabilizers in K , as well as complete Schreier trees. Thus for the K -action on \mathcal{Q} we are facing a similar problem as for the H -action on \mathcal{O} , apart from the requirement on Schreier trees; so we can just recurse. For any K -orbit in \mathcal{Q} , we choose a representative, called its ‘distinguished point’. Then, for any K -orbit \mathcal{O}^{\prime}\subseteq\mathcal{O} , the \pi_{K} -preimages of the distinguished point of \pi_{K}(\mathcal{O}^{\prime})\subseteq\mathcal{Q} are

Read the complete text