Zimmer’s conjecture for actions of SL(m,Z)

Zimmer’s conjecture aims to extend Margulis’ superrigidity theorem to a nonlinear setting. Except for actions on the circle there has not been much progress in the last 35 years. A new preprint of Brown-Fisher-Hurtado and its predecessor are now settling the conjecture for finite index subgroups of \(SL(n,{\mathbb Z})\) and cocompact lattices in \(SL(n,{\mathbb R})\).

Brown, Fisher, Hurtado: Zimmer’s conjecture for actions of SL(m,ℤ), https://arxiv.org/abs/1710.02735

Margulis’ superrigidity theorem from 1975 says that linear representations of a lattice in a higher rank Lie group are either restrictions of a Lie group representation or have finite image. Its nonlinear version is Zimmer’s conjecture from 1983 that such lattices can not act on low-dimensional closed manifolds except for actions factoring over a finite group action. There is a precise statement of it, which in the case of lattices \(\Gamma\in SL(n,{\mathbb R})\) says that any \(\Gamma\)-action on an (n-2)-dimensional closed manifold should factor over a finite group action. (Note that \(SL(n,{\mathbb Z})\) acts on the \((n-1)\)-dimensional projective space, so the conjecture is sharp.)

A survey of the state of the conjecture around 2008 can be found in Fisher: Groups Acting on Manifolds: Around the Zimmer Program, in Geometry, Rigidity, and Groups Actions, Chicago Lectures in Mathematics.

For \(n=3\), i.e. lattice actions on the circle, it is at least known by Witte Morris, Burger-Monod and Ghys that any such action has a global fixed point. In higher dimensions very little is known. So the new preprints by Brown-Fisher-Hurtado seem to be a major breakthrough. The first preprint from December 2016 settles the case of cocompact lattices in \(SL(n,{\mathbb R})\) (Brown, Fisher, Hurtado: Zimmer’s conjecture: Subexponential growth, measure rigidity, and strong property (T)) and the new preprint from this month now settles the technical problems with extending the proof to non-cocompact lattices.

The proofs in both cases work for \(C^2\)-actions and the actual result is that in both cases actions are shown to leave some Riemannian metric invariant, which implies that the action factors over some compact and hence (in the presence of unipotent elements) over some finite group.

The idea behind the proofs is that an action preserving a Riemannian metric needs to have zero Lyapunov exponents

\(L(\gamma,x)=\liminf_{n\to\infty}\frac{1}{n}\int_X \log \parallel D_x\gamma^n\parallel d\mu(x)\)

What the papers do is to provide a kind of converse. (Assuming that one already has an invariant measure, which is known from the early work of Zimmer.) First they prove that the actions in question must have subexponential growth of derivatives, that is, for any \(\epsilon>0 \) exists \(C_\epsilon \) such that

\(\parallel D\gamma\parallel \le C_\epsilon e^{\epsilon l(\gamma)} \)

with \(l(\gamma) \) the word length of \(\gamma \). (Among other things this uses Ratner’s measure rigidity theorem.)
And second they prove that subexponential growth of derivatives together with Lafforgue’s strong property (T) implies (for \(C^2\)-actions of finitely generated groups) the preservation of a Riemannian metric. Lafforgue’s strong property (T) is known for lattices in \(SL(n,{\mathbb R})\) by a recent preprint (de la Salle: Strong (T) for higher rank lattices, Preprint 2017), and so Zimmer’s conjecture follows.
So far that was the argument of the first preprint, which proved the conjecture for cocompact lattices. There have been several problems with extending this approach to the non-compact case – see Section 1.3 of the new preprint for a discussion of those – and they are overcome in this new paper.