0
$\begingroup$

I am asked to describe the cosets of the subgroup H of $D_n$ (Dihedral group on n vertices) generated by the reflection element s, $H=\langle s \rangle$. So $H=\{id,s\}$, and for $r^i \in D_n ,1 \le i \le n $, $$ r^iH = \{r^i,r^is\}\quad Hr^i =\{r^i,sr^i\}$$ $$sr^iH=\{sr^i,r^{n-i}\} \quad Hsr^i=\{sr^i,r^i\} $$

My question is if this adequately describes the set of cosets of $H \lt D_n$, and if in general when describing cosets, one can separate the cosets into cases as I have done above.

$\endgroup$
2
  • $\begingroup$ Why wouldn't this be an adequate description of the cosets? All of them are written on the page and even their elements are listed out explicitly. $\endgroup$ Oct 6, 2016 at 16:20
  • $\begingroup$ @EricStucky I wasn't sure if I was missing any. Is there a way to show no other cosets exist? Or is there a better way to describe all cosets of a given group and subgroup? $\endgroup$ Oct 6, 2016 at 16:29

1 Answer 1

2
$\begingroup$

In general there's no nice way of describing all of the cosets for a given group and subgroup. For this specific case, this presentation is just fine.

However, you may worry (as you do in the comments) that you might have missed some of the cosets. Fortunately there is an easy way to check this: all left cosets (and all right cosets) must cover the entire group. So you can make sure there are no more just by guaranteeing that any element of the group is in exactly one left coset and one right coset in your list.

For the dihedral group this is fairly straightforward, since you have a nice simple set of generators and relations.

$\endgroup$
1
  • $\begingroup$ Thanks for the insight, so essentially one must verify that the set of cosets partitions the group. Although it seems that finding the description of the set of cosets for a given group and subgroup could be a hard problem, depending on the structure of the group. $\endgroup$ Oct 6, 2016 at 16:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .