Block (permutation group theory)

Last updated

In mathematics and group theory, a block for the action of a group on a set is a subset of whose images under either coincide with or are disjoint from . These images form a block system, a partition of that is -invariant. In terms of the associated equivalence relation on , -invariance means that

Contents

implies

for all and all . The action of on induces a natural action of on any block system for . [1]

The set of orbits of the -set is an example of a block system. The corresponding equivalence relation is the smallest -invariant equivalence on such that the induced action on the block system is trivial.

The partition into singleton sets is a block system and if is non-empty then the partition into one set itself is a block system as well (if is a singleton set then these two partitions are identical). A transitive (and thus non-empty) -set is said to be primitive if it has no other block systems. [2] For a non-empty -set the transitivity requirement in the previous definition is only necessary in the case when and the group action is trivial.

Stabilizers of blocks

If B is a block, the stabilizer of B is the subgroup

GB = { gG | gB = B }.

The stabilizer of a block contains the stabilizer Gx of each of its elements. Conversely, if xX and H is a subgroup of G containing Gx, then the orbit H.x of x under H is a block contained in the orbit G.x and containing x.

For any xX, block B containing x and subgroup HG containing Gx it's GB.x = BG.x and GH.x = H.

It follows that the blocks containing x and contained in G.x are in one-to-one correspondence with the subgroups of G containing Gx. In particular, if the G-set X is transitive then the blocks containing x are in one-to-one correspondence with the subgroups of G containing Gx. In this case the G-set X is primitive if and only if either the group action is trivial (then X = {x}) or the stabilizer Gx is a maximal subgroup of G (then the stabilizers of all elements of X are the maximal subgroups of G conjugate to Gx because Ggx = gGxg1).

References

  1. Seress, Ákos (2003), Permutation Group Algorithms, Cambridge Tracts in Mathematics, vol. 152, Cambridge University Press, Cambridge, p.  9, doi:10.1017/CBO9780511546549, ISBN   0-521-66103-X, MR   1970241
  2. Dixon, John D.; Mortimer, Brian (1996), Permutation Groups, Graduate Texts in Mathematics, vol. 163, Springer-Verlag, New York, p.  12, doi:10.1007/978-1-4612-0731-3, ISBN   0-387-94599-7, MR   1409812

See also