Conjecture: For transitive imprimitive permutation groups, the intersection of the stabilisers of any pair of blocks is not trivial

48 Views Asked by At

Let $G < S_\Omega$ be a group of permutations of a finite set $\Omega$ such that it is transitive, and has at least one, non trivial, system of imprimitivity $\mathcal{P} = \{B_1, \dots, B_m\}$.

I have the following conjecture: for any pair of blocks $B_i$ and $B_j$ the intersection of their stabilizers is not trivial, i.e., $$\{id\} < Stab_G(B_i) \cap Stab_G(B_j)$$ Where $id$ is the identity permutation.

By the moment my approach to this is the following. Consider $h \in Stab_G(B_i)$. For $x \in \Omega$ denote the minimal whole number $k\geq 0$ such that $x\cdot h^k = x$ as $o_h(x)$. As well for $B_j \in \mathcal{P}$, $o_h(B_j)$ the minimal whole number such that $B_j \cdot h^k = B_j$. For example $o_h(B_i) = 0$. We can see that if $x \in B_j$ then $$o_h(x)\geq o_h(B_j)$$ the problem would be if for every $h \in Stab_G(B_i)$ we have that $o_h(x) = o_h(B_j)$. By the moment I have not found neither a counter example to the conjecture nor why the inequality above is nor strict for every permutation in the stabilizer.

Any hint or reference for this would be appreciated.

1

There are 1 best solutions below

1
On BEST ANSWER

This does not hold. Small counterexample:

$$ g:=\langle (1,3,5)(2,4,6), (1,4)(2,3)(5,6) \rangle\cong S_3 $$ Then $\{\{1,4\},\{2,5\},\{3,6\}\}$ is a block system, but the stabilizers of $\{1,4\}$ and $\{2,5\}$ intersect trivially.

This is not dependent on the representation being regular. For example take $$ g=\langle (1,3,5,8,10,12,14,15)(2,4,6,7,9,11,13,16), (1,5,16)(2,6,15)(7,10,14)(8,9,13)\rangle\cong GL(2,3)$$ and the blocks $\{1,2\}$, $\{3,4\}$ (in the same block system).

But what it of course requires is that the permutation representation is not ``minimal'' -- if the action on the blocks has a nontrivial kernel, this kernel will lie in every block stabilizer.