Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

Given a group $G$ which acts on a set $X$ we define that a group $G$ acts doubly transitive on $X$ as $G$ is transitive on $X$ and $G_x$ is transtive on $X \backslash \{x\}$ where $G_x$ the stabilizer is and $x \in X$. I want to prove that for $x\neq x'$ and $y \neq y'$ an element $g \in G$ exists with $(g(x),g(x')) = (y,y')$. How can I do that ?

share|cite|improve this question
up vote 2 down vote accepted

Let's just think about our definitions. We know that there is a $g_1 \in G$ with $$ g_1 \cdot x = y. $$ We also know that for any point $z \ne x$, there is a $g_2 \in G_x$ with $$ g_2 \cdot x' = z. $$ Let's check that we can take $$ z = g_1^{-1}\cdot y'; $$ in other words, we need to check that $g_1^{-1} \cdot y' \ne x$. If this were the case, then we would have $$ g_1\cdot x = y', $$ but we already know that $g_1 \cdot x = y$ and that $y \ne y'$. Now I claim that $g_1g_2$ is the required element. We have $$ (g_1g_2)\cdot x = g_1\cdot (g_2 \cdot x) = g_1 \cdot x = y, $$ since $g_2 \in G_x$, and $$ (g_1g_2)\cdot x' = g_1\cdot (g_2 \cdot x') = g_1 \cdot (g_1^{-1} \cdot y') = y'. $$ So intuitively, what we're doing here is first acting by an element of the stabilizer so that $x'$ has moved to the right position for it to then be moved to $y'$ when we act by the element moving $x$ to $y$.

share|cite|improve this answer
Greath thanks to you. Now i can sleep :) – Epsilon Sep 30 '12 at 22:56
Is the converse of this theorem true ? – Epsilon Oct 1 '12 at 15:17
Yes, it is. Perhaps you would like to try showing it yourself. – NKS Oct 1 '12 at 21:50

The idea will be to pick $g_1 \in G$ with $g_1 \cdot x = y$ and then use $g_2 \in G_x$ to move $x'$ to $y'$. We'll have to use $g = g_1 g_2$ to insure that $g \cdot x = y$. And in order to have $g \cdot x' = y'$, we have to pick $g_2 \cdot x' = g_1^{-1} \cdot y'$.

share|cite|improve this answer
A small comment, but it wasn't clear to me at first why such a $g_2$ had to exist (as in, why couldn't $g_1^{-1}\cdot y' = x$?) It turns out to be easy, but I do think it is worth checking, or at least noting. – NKS Sep 30 '12 at 22:46
You're absolutely right, +1 to your thorough response. – Michael Joyce Sep 30 '12 at 22:48
You seem to be assuming that the $x$ in the second sentence of the problem is the same as the $x$ in the first sentence, but it need not be. You only know that $G_x$ is transitive on $\Omega \setminus \{x\}$ for some specific $x \in X$, but you need to prove the condition for all $x,x',y,y'$. – Derek Holt Oct 1 '12 at 15:19
@DerekHolt: Although it's not clearly stated in the problem, my assumption was that you know $G_x$ is transitive on $X \setminus \{x\}$ for all $x \in X$. Can one really prove the result only assuming it holds for a single $x \in X$? – Michael Joyce Oct 1 '12 at 16:12
@Michael Joyce. Yes one can, but it needs an extra step. Suppose $G_x$ is transitive on $X \setminus \{x\}$ for some fixed $x$ and $y \in X$. Let $y_1,y_2 \in Y \setminus \{y\}$. Since $G$ is transitive on $X$, there exists $g_1 \in G$ with $g_1.y=x$. So $g_1.y_1,g_1.y_2 \in X \setminus \{x\}$ and there exists $g_2 \in G_x$ with $g_2.g_1.y_1=g_1.y_2$. Now $g = g_1^{-1}g_2g_1$ satisfies $g \in G_y$ with $g.y_1=y_2$, and we have the condition for all elements of $X$. – Derek Holt Oct 1 '12 at 17:49

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.