Counting conjugacy classes in the unipotent radical of parabolic subgroups of \(\text{GL}_n(q)\). (Q848762)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Counting conjugacy classes in the unipotent radical of parabolic subgroups of \(\text{GL}_n(q)\).
scientific article

    Statements

    Counting conjugacy classes in the unipotent radical of parabolic subgroups of \(\text{GL}_n(q)\). (English)
    0 references
    0 references
    0 references
    23 February 2010
    0 references
    Let \(G=\mathrm{GL}_n(q)\) be the general linear group over the field \(\mathbb{F}_q\) with \(q\) elements, where \(q\) is a power of a prime \(p\). It has been conjectured by \textit{G. Higman} [Proc. Lond. Math. Soc., III. Ser. 10, 24-30 (1960; Zbl 0093.02603)] that the number of conjugacy classes of a Sylow \(p\)-subgroup of \(G\) is given by a polynomial in \(q\) with integer coefficients. In this note the authors consider a natural extension of this question: given a parabolic subgroup \(P\) of \(G\), let \(U=O_p(P)\) denote the largest normal \(p\)-subgroup of \(P\). Then is the number of conjugacy classes in \(U\) given by a polynomial in \(q\) with integer coefficients (Higman's original conjecture is the case where \(P\) is a Borel subgroup of \(G\))? The main result of the paper is an affirmative answer to this question when the parabolic subgroup \(P\) is the stabilizer of a flag in \(\mathbb{F}_q^n\) of length at most \(5\). The proof of this result relies on translating the problem into a representation theoretic setting, where conjugacy classes in \(U\) correspond to a class of modules for a certain algebra. Also essential is a result of \textit{L. Hille} and \textit{G. Röhrle} [Transform. Groups 4, No. 1, 35-52 (1999; Zbl 0924.20035)] about the action by conjugation of a parabolic subgroup \(\mathbf P\) of \(\mathrm{GL}_n(\overline\mathbb{F}_q)\) on its unipotent radical \(\mathbf U\), where \(\overline\mathbb{F}_q\) denotes the algebraic closure of \(\mathbb{F}_q\). This result says that the number of \(\mathbf P\)-classes in \(\mathbf U\) is finite precisely when \(\mathbf P\) is the stabilizer of a flag of length at most \(5\) in \(\overline\mathbb{F}_q^n\); moreover, in these cases, the parameterization of the classes is independent of \(q\). This is the reason for the restriction on the length of flags in the main result.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Higman conjecture
    0 references
    parabolic subgroups
    0 references
    unipotent radicals
    0 references
    general linear groups
    0 references
    numbers of conjugacy classes
    0 references
    0 references
    0 references