Specht modules with Abelian vertices. (Q658540): Difference between revisions

The vertex of a module for a finite group was defined by \textit{J.~A.~Green} [in Math. Z. 70, 430--445 (1959; Zbl 0086.02403)]. It has since become a critical invariant in many conjectures in modular representation theory, including Alperin's Weight Conjecture [\textit{J. L. Alperin} and \textit{P.~Fong}, J. Algebra 131, No. 1, 2--22 (1990; Zbl 0714.20007)]. Despite this, comparatively little is known about the vertices of `naturally occurring' modules such as Specht modules for symmetric groups. Let \(F\) be an algebraically closed field of prime characteristic \(p\) and let \(S^\mu\) be an indecomposable Specht module for \(FS_n\). Theorems 3.1 and 3.2 in the paper under review state that if \(p\geq 3\) and \(S^\mu\) has an Abelian vertex of \(p\)-rank \(m\), then \(m\) is equal to the \(p\)-weight of \(\mu\), and the vertex is the product of \(m\) cyclic subgroups generated by mutually disjoint \(p\)-cycles. Moreover, \(\mu\) is a \(p^2\)-core, and the complexity of \(S^\mu\) is \(m\). When \(p=2\), Theorem~3.3 gives an analogous result for those Specht modules known to be indecomposable. As a corollary, the author shows that if \(1\leq m\leq p-1\) and \(p\geq 3\) then the vertex of \(S^\mu\) is Abelian of \(p\)-rank \(m\) if and only if the \(p\)-weight of \(\mu\) is \(m\). This is a considerable generalization of an earlier result of the reviewer [Arch. Math. 81, No. 5, 505--511 (2003; Zbl 1046.20011)], namely that \(S^\mu\) has a cyclic vertex if and only if the \(p\)-weight of \(\mu\) is \(1\). The author also shows that if \(\mu\) is one of the \(p\times p\) partitions considered by \textit{D. J. Hemmer} [in J. Algebr. Comb. 30, No. 4, 421--427 (2009; Zbl 1191.20010)], that is to say, the Young diagram of \(\lambda\) is made up of \(p\times p\) blocks, then the vertices of \(S^\mu\) are non-Abelian. The proofs make use of the rank variety of a module, as defined by \textit{J. F. Carlson} [J. Algebra 85, 104--143 (1983; Zbl 0526.20040)]. The paper ends with some natural further questions, including Question 6.4: `For \(p=2\), is there an indecomposable Specht module \(S^\mu\) whose vertices are Abelian and contain \(\mathbb Z_4\) as a direct factor?'