MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

In the literature there are several definitions of "rigidity" (or "super-rigidity") of representations, adapted to the circumastances. I wonder what are the relations between them; I excuse in advance if the answers are known, after some research I could not find them. I have singled out a few definitions for concreteness.

Notation first: let $\Gamma = \pi_1(X, x_0)$ where $X$ is Kähler, and $G$ a (semi-)simple connected algebraic group.

  1. (classic rigidity): Every $\phi$ close to $\rho$ (in the analytic topology of $\text{Hom}(\Gamma, G)$) is conjugated to $\rho$ by an element of $G$.
  2. (super-rigidity, e.g. Margulis): The harmonic $\rho$-equivariant map $f \colon \tilde{X} \to G/K$ is totally geodesic.
  3. (Hermitian super-rigidity, e.g. Siu): Here we suppose $G/K$ is Hermitian symmetric. Then $f$ as above is holomorphic.
  4. (from Kim-Pansu, Duke Math. J., to appear): (Here we may need to suppose $X$ to be a surface of high enough genus) $\rho$ is rigid if every $\phi$ close enough to $\rho$ is not smooth in $\text{Hom}(\Gamma, G)$ (equivalently, under the assumption on $X$: not Zariski-dense).

The last one is a bit "ad hoc" to their situation, since it allows them to prove that $\text{Hom}(\Gamma, G)$ splits in connected components, either entirely rigid or where Zariski dense points form a dense subset.

I think that 3. has been introduced as an intermediate step to 2. (when $\tilde{X}$ is a Hermitian symmetric space), but I do not know how general this is. To me, the only clear implication is 1. $\implies$ 4. (if $\phi_n \to \rho$ are smooth, then $\dim T_{\phi_n}\text{Hom}(\Gamma, G) = vdim = (1-\chi(X)) \dim(G)$ but if a neighborhood of $\phi_n$ is conjugated to $\rho$ then $\dim T_{\phi_n}\text{Hom}(\Gamma, G) \leq \dim G$).

I also think that despite the "super-" name there is a good chance that 1. $\implies$ 3. (at least if $X$ is a surface); so, main question:

Is there any implication between 1. and 3.? Or counter-example?

More with general-knowledge purposes:

Is there any other general implication / counterexample of implications between the points above? For example, how generally 3. $\implies$ 2.? Is 4. much weaker then any other one?

Of course, I would also be happy with answers for $\Gamma$ a cocompact lattice in $H$ and $\tilde{X} = H/K'$, or even just $X$ a Riemann surface.

share|cite|improve this question

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.