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

Q: How does one construct a non continuous representation $\rho:SL_2(\mathbf{R})\rightarrow G$ for some connected (finite dimensional) Lie group $G$?

share|cite|improve this question
You can embed $SL_2(\mathbf{R})$ into $SL_2(\mathbf{C})$ and twist by a wild automorphism of $\mathbf{C}$. – YCor Mar 23 '12 at 12:37
Right, good point! – Hugo Chapdelaine Mar 23 '12 at 14:39
For the existence of wild automorphisms of $\mathbf{C}$ and for the construction Misha quoted (due to Borel--Tits) you need the axiom of choice, however. – Guntram Mar 24 '12 at 10:00
@Guntram: You are right of course, but if you eliminate the (uncountable) Axiom of Choice, then, at least in Solovay's model, all subsets of ${\mathbb R}$ are measurable and we are in the situation discussed in… (see pm's answer). – Misha Mar 24 '12 at 20:33
up vote 7 down vote accepted

Example can be found, for instance, in Boris Weisfeiler's paper "Abstract homomorphisms of big subgroups of algebraic groups", pages 149-150, see

His example of a discontinuous representation $\rho$ of $SO(n, {\mathbb R})$ to a semidirect product $H$ of $SO(n, {\mathbb R})$ with the abelian group ${\mathbb R}^N$ (the Lie algebra of $SO(n)$), works for $SL(2, {\mathbb R})$ as well. Actually, Weisfeiler's example is even more dramatic: The image of the compact group $SO(n)$ under $\rho$ is dense in the noncompact Lie group $H$. Weisfeiler's paper also lists many positive results on rigidity of abstract homomorphisms of Lie groups.

share|cite|improve this answer
Wow ! I would have bet that no such things could exist ! – BS. Mar 23 '12 at 12:31
Thanks a lot Misha for the reference. – Hugo Chapdelaine Mar 23 '12 at 14:40

There are natural function spaces on Lie groups that are nevertheless not continuous (and, thus, are not representations in any usual, useful sense). For example, already on $G=\mathbb R$, the Frechet space $V$ of all continuous functions, and/or the Frechet space of bounded continuous functions, with the translation action of $G$, are not repn spaces, in the sense that $G\times V\rightarrow V$ is not continuous. The reason is the existence of not-uniformly-continuous continuous functions. For example, $f(x)=\sin(x^2)$.

share|cite|improve this answer

A partial answer:

No measurable constructions are possible. Any measurable group homomorphism between locally compact groups is automatically continuous, in fact $C^\infty$ for Lie groups. You can have a look at the answers in an old question of mine:

Are measurable automorphism of a locally compact group topological automorphisms?

If I would like to find something non continuous, I personally would start with finding some non measurable automorphism of the circle first.

share|cite|improve this answer

Your Answer


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

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