A lower bound on the vertices of Specht modules for symmetric groups. (Q403027): Difference between revisions

This paper is a contribution to the modular representation theory of the symmetric groups, and in particular the study of the Specht modules, which continue to attract a great deal of attention. The problem studied here is the determination of the \textit{vertices} of Specht modules: recall that a vertex of a module \(M\) for a finite group \(G\) is a minimal subgroup \(P\) of \(G\) with the property that there is a module \(V\) for \(P\) such that the induced module \(V{\uparrow}^G\) has \(M\) as a direct summand. Finding the vertices of modules is in general very difficult and provides interesting information about modules. The author considers a Specht module \(S^\lambda\) for \(\mathfrak S_n\), and constructs a (typically quite large) subgroup of \(\mathfrak S_n\) which is necessarily contained in some vertex of \(S^\lambda\) (vertices are unique up to conjugacy, so the word `some' here imposes an insignificant restriction). Specifically, given a \(\lambda\)-tableau \(t\), the author constructs a group \(H(t)\times H(t')\), which may be simply described as the set of permutations under which any two entries in the same row of \(t\) are mapped to entries in the same row and any two entries in the same column of \(t\) are mapped to entries in the same column. The main theorem states that there is a vertex of \(S^\lambda\) containing a Sylow \(p\)-subgroup of \(H(t)\times H(t')\). The paper concludes with an observation that in a certain special case (for hook partitions of size not divisible by \(p\)) this bound is achieved. The paper is very well written, giving a very clear and concise summary of the necessary background before proving the main result efficiently. I hope to see more papers on this subject from this author.