1
$\begingroup$

Suppose that $G$ is a finitely generated group, $X$ is a $G$-set, and $x \in X$ is a point. Are there any sorts of conditions on $X$ and $G$ that would let me conclude that $\operatorname{Stab}(x)$ is finitely generated as well?

$\endgroup$
1
  • $\begingroup$ Note that for any subgroup $H$ you can take $X=G/H$, so what you are asking is for a characterization of all finitely generated subgroups of a given group. $\endgroup$
    – user1688
    Jul 14, 2016 at 8:08

2 Answers 2

6
$\begingroup$

"Stabilizer of $x$" is not any more specific than "subgroup of $G$" (consider the regular action of $G$ on $G/H$). In other words, you are asking for conditions on $G$ such that its subgroups are finitely generated.

This is false, in general. It's true in nilpotent (and virtually nilpotent) groups. It's false in solvable groups (consider the lamplighter).

$\endgroup$
2
$\begingroup$

Lior Silberman is absolutely correct, but I wanted to add one other natural and simple condition in which you can conclude that the stabilizer of $x$ is finitely generated, namely, when the orbit of $x$ is finite. Equivalently, subgroups of finite index in finitely generated groups are finitely generated. This follows from Schreier generators (see "Schreier's lemma" or "Schreier's subgroup lemma").

$\endgroup$

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