Let $G$ be a semisimple algebraic group, and let $\Gamma$ be a finitely generated subgroup. Given the generators of $\Gamma$, is there a good way to determine if $\Gamma$ is Zariski-dense in $G$?

For example, let $\Gamma \subseteq \mathrm{PSL}_2$ where $ \displaystyle \Gamma = \left\langle \gamma_1 , \gamma_2 \right\rangle$ and $\gamma_1 = \begin{pmatrix} 1 & i \newline 0 & 1 \end{pmatrix}$ and $\gamma_2 = \begin{pmatrix} 1 & 0 \newline 1 & 1 \end{pmatrix}$.

  1. What do you mean by "good"? In practice, a theorem of either Lubotzky or Weigel (depending on who you ask) states that if a congruence quotient is surjective for SOME prime (bigger than 3, say), then the subgroup is Zariski-dense, so generally checking for a couple of primes gives a certificate.

  2. In $PSL_2,$ Zariski-dense is equivalent to non-elementary, so your group is Zariski-dense.

EDIT To answer the OP's comment: in this paper: Zariski density and genericity I Rivin - International Mathematics Research Notices, 2010, I show that two "random" matrices generate a Zariski dense subgroup. R. Aoun in this paper: Transience of algebraic varieties in linear groups and application to generic Zariski density R Aoun - Arxiv preprint arXiv:1103.0944, 2011 - arxiv.org

Generalized it (at the cost of losing the effectiveness). Aoun also showed (in a different paper) that a random subgroup is free (which, in higher rank) means that it is not a lattice. In rank one a corresponding result (not yet written up fully) was shown by Elena Fuchs, myself, and Peter Sarnak: there you show that generically the Hausdorff dimension of the limit set is small, so again you don't get a lattice generically (of course lattices are Zariski dense by the Borel Density Theorem).

You might also look at these slides:

  • $\begingroup$ By "good," I meant that I was looking for some criterion for Zariski-density based on the generators of the group. I was looking for a way to create examples of Zariski-dense subgroups that are a little more interesting than $\mathrm{SL}_2(\mathbb{Z})$ in $\mathrm{SL}_2$ and most importantly, be able to prove that these examples are Zariski-dense. Thank you very much for your answer; it sounds like it's what I was looking for. $\endgroup$
    – JS_UNCUVa
    Feb 23 '12 at 20:56
  • $\begingroup$ See the edit for more info... $\endgroup$
    – Igor Rivin
    Feb 24 '12 at 2:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.