Nonarithmetic superrigid groups: Counterexamples to Platonov's conjecture

From MaRDI portal



Abstract: Margulis showed that "most" arithmetic groups are superrigid. Platonov conjectured, conversely, that finitely generated linear groups which are superrigid must be of "arithmetic type." We construct counterexamples to Platonov's Conjecture.


A finitely generated group \(\Gamma\) is called representation superrigid if for all its finite dimensional complex representations \(\rho\) the dimension of the Zariski closure of \(\rho(\Gamma)\) remains bounded. This property implies that \(\Gamma\) is representation rigid, i.e., that it admits only finitely many non-isomorphic irreducible representations of any given dimension. The main result of the paper is as follows. Let \(\Gamma\) be a cocompact lattice in the real rank 1 group \(G\) of type \(F_4\). Then there exist a finite index normal subgroup \(\Gamma_1\) of \(\Gamma\) and an infinite index subgroup \(\Lambda\) of \(\Gamma_1\times\Gamma_1\), containing the diagonal, such that any representation of \(\Lambda\) extends uniquely to a representation of \(\Gamma_1\times\Gamma_1\); the groups \(\Lambda\) and \(\Gamma_1\times\Gamma_1\) are representation superrigid, and all their representations are semi-simple; \(\Lambda\) is not isomorphic to a lattice in any product of groups \(H(k)\), where \(H\) is a linear algebraic group over a local field \(k\). The proof relies on two opposite properties of \(\Gamma\): it is superrigid in \(G\) and hyperbolic in the sense of Gromov. The result contradicts a conjecture of Platonov claiming that a rigid linear group must be of arithmetic type.











This page was built for publication: Nonarithmetic superrigid groups: Counterexamples to Platonov's conjecture

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1585692)