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

Let X be a contractible metric space and G a topological group acting transitively on X (i.e. given any two points x,y \in X, there exists g \in G such that gx=y).

My question is the following: is it true that given any x \in X its stabilizer Stab(x)={ g \in G : gx=x } and the whole group G have the same homotopy type?

If the answer is "no", I'd like to know some "mild" hypothesis that could be add to have an affirmative response.

For instance, I know that whenever G is a Lie group and H < G is a closed subgroup such that G/H is contractible, then G and H are homotopically equivalent (in this case H can be seen as the stabilizer of the coset H under the natural G-action on G/H). However, to assume that G is a Lie group seems to be too restrictive. In fact, I'd like to apply this "result" to some groups which are not locally compact.

share|cite|improve this question

The short: no. E.g. let X be a topological group and G be the underlying discrete group of X, acting on X by left translation.

One standard hypothesis is the existence of slices. For some (hence any) point x in X, if there is an open neighborhood U of x and a section s:U -> G such that s(u)*x = u for all u in U, then the map p:G -> X given by p(g) = g*x is a fiber bundle with fiber Stab(x).

This hypothesis automatically implies that Stab(x) is weakly equivalent to G by the long exact sequence of the fibration. If G and Stab(x) have the homotopy type of CW-complexes then they are homotopy equivalent by the Whitehead theorem.

share|cite|improve this answer
Tyler, thanks for the ideas. About the counterexample, I didn't say it explicitly (sorry), but I'm interested in the case where G is connected. About the second and third paragraphs, you're completely right. But, is there any chance of getting an affirmative answer imposing extra hypothesis on G and/or on X rather than on the action itself? Thanks again. – Alejandro Oct 22 '09 at 12:18
If G is abelian, you can apply the standard simplicial "classifying space" functor, which takes topological abelian groups to topological abelian groups, and takes an example of the above type to a connected example. So far as the further question, did you have specific types of less-tractable groups in mind? – Tyler Lawson Oct 23 '09 at 4:45
I'm particularly interested in the case G < Diff(M) and X is some manifold, in general, different of M. So, the CW-complex-hypotheses seems reasonable, but the existence of slices seems to be too strong. – Alejandro Oct 23 '09 at 13:43

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.