Twisted Frobenius-Schur indicators of finite symplectic groups. (Q2575693): Difference between revisions

The reviewer investigated the complex irreducible characters of the finite classical groups with regard to their reality properties in two papers published in 1983 and 1985. It is well known that methods of linear algebra suffice to identify and enumerate the real classes in these groups, but information about the Frobenius-Schur indicators of the irreducible characters is harder to obtain, especially if a complete description of the characters is unavailable. The reviewer showed that the Witt-Berman induction theorem provides a tool to overcome this problem, since it implies that the indicators are controlled by the characters of certain special subgroups whose structure is not too complicated, especially if viewed from an inductive standpoint. As a consequence, he was able to obtain fairly complete details about the indicators and then used the Frobenius-Schur involution formula to deduce some unusual formulae relating to the sum of the degrees of the irreducible characters. Certain of these formulae, especially in the context of the finite general linear group, had interpretations in terms of the concept of a model for the representations of the group, something which A. A. Klyachko had shown to exist in a paper also published in 1983. The paper under review here continues the work begun over twenty years ago and the author fills in some gaps that had persisted since then. He concentrates his efforts on the finite symplectic group \(G=\text{Sp}(2n,q)\), where \(q\) is a power of an odd prime. When \(q\equiv 1\bmod 4\), all the characters of \(G\) are real-valued and the reviewer used the Frobenius-Schur theory to show that the sum of the degrees of the irreducible characters of \(G\) is \[ q^{n(n+1)/2}\prod_{i=1}^n(q^i+1). \] The author shows that an identical formula holds when \(q\equiv 3\bmod 4\), but he must employ a different method to obtain it, since not all the characters of \(G\) are real-valued in this case. His key tool is an involutory automorphism \(\iota\) of \(G\) which, while defined for all odd \(q\), is inner when \(q\equiv 1\bmod 4\), but transforms a non-real conjugacy class of \(G\) into its corresponding inverse class, while fixing the real classes, when \(q\equiv 3\bmod 4\). The appropriate group to examine is then an extension \(G^+\) of \(G\) by an element of order 2 which induces \(\iota\) on \(G\) by conjugation. The author employs the Witt-Berman theorem in \(G^+\) to obtain his main result, rather as the reviewer had done previously. The corresponding problem for the finite unitary group is more complicated and in subsequent work the author and N. Thiem have investigated it using the more powerful machinery of Lusztig's theory of representations of finite groups of Lie type and symmetric functions.