On the congruence kernel for simple algebraic groups (Q2631151)

Over the past two decades, the authors have discovered techniques which, on the one hand, simplify and explain ideas involved in the congruence subgroup problem and, on the other, have much wider applicability. Some of these developments have been explained in their survey [the authors, ``Developments on the congruence subgroup problem after the work of Bass, Milnor and Serre,'' in: Collected papers of John Milnor. V: Algebra. Providence, RI: American Mathematical Society (AMS). 307--325 (2010)]. Due to the fertility of the ideas expounded in the above-mentioned survey, this paper does have some intersection with the earlier survey. However, in this paper, a number of new results (some of which were announced more than a decade earlier) have been proved here for the first time. We just single out two of the striking results. One is a very simple proof of Lubotzky's conjecture on the finiteness of the congruence kernel for arithmetic groups in characteristic \(0\) which are adelic profinite groups. As this was proved earlier by \textit{V. P. Platonov} and the reviewer [J. Algebra 193, No. 2, 757--763, Art. No. JA967011 (1997; Zbl 0884.20018)], the 75th birthday conference volume is an apt place for this proof to appear. The other striking result of the paper we would like to draw attention to, addresses the structure of the \(S\)-congruence kernel of an \(S\)-arithmetic group over a number field when it is infinite. The authors prove that the congruence kernel has a closed normal subgroup generated by a single element such that the quotient is a finite cyclic group. This is proved for \(\mathrm{SL}_2\) first and then Jacobson-Morozov is used. One of the valuable ideas employed is the strong approximation property for tori with respect to arithmetic progressions as developed by the authors in the last 15 years. The paper is also interspersed with several valuable remarks.