Discrete subgroups acting transitively on vertices of a Bruhat-Tits building (Q408112): Difference between revisions

Let \(\mathfrak{B}\) be an affine Bruhat-Tits building and let \(\Gamma\) be a discrete subgroup of its group of automorphisms which acts transitively on the facets of \(\mathfrak{B}\) of a given type. These groups were intensively studied by a number of authors. In a 1987 paper [Bull. Am. Math. Soc., New Ser. 16, 129--133 (1987; Zbl 0631.20033)], \textit{W. M. Kantor, R. A. Liebler} and \textit{J. Tits} showed that (if the relative rank of \(\mathfrak{B}\) is at least \(2\)) there are only very few such \(\Gamma\)'s that act transitively on the \textit{chambers} of \(\mathfrak{B}\). This paper was a motivation for an important work of \textit{A. Borel} and \textit{G. Prasad} [Publ. Math., Inst. Hautes Étud. Sci. 69, 119--171 (1989; Zbl 0707.11032)] in which among the other results the authors proved a finiteness theorem for the transitive actions on facets whose type has at least two elements. The only remaining case of the \textit{vertex-transitive actions} is the subject of the present paper. In contrast with the result of Borel-Prasad, the authors show that there are infinitely many vertex-transitive actions. The method of the proof is based on the work of Borel-Prasad [loc. cit.] but the emphasis is different: instead of looking for a finiteness result the authors of the present paper are interested in a classification of all the possible actions. The main result of the paper is an impressive theorem which gives a \textit{complete classification} of the vertex-transitive actions assuming that \(\mathfrak{B}\) is the Bruhat-Tits building of an absolutely almost simple group over a local field of characteristic \(0\) and the dimension of \(\mathfrak{B}\) is at least \(4\). As a corollary of the main theorem we see that there are no such actions if the dimension of \(\mathfrak{B}\) is larger than \(7\). This is in contrast with a result of \textit{D. Cartwright} and \textit{T. Steger}, who showed that in the positive characteristic case there are simply transitive actions on the vertices of the buildings of type \(\widetilde A_n\) for any \(n \geq 2\) [Isr. J. Math. 103, 125--140 (1998; Zbl 0923.51010)]. In an appendix the methods of the paper are applied to give an alternative proof of the Siegel-Klingen theorem on the rationality of certain Dedekind zeta and \(L\)-functions.