1
$\begingroup$

Given the Dihedral group $ D_4 $ (that is where $ D_4 = $ { $ id, R, R^{2}, R^{3}, F, RF, R^{2}F, R^{3}F $} ); Let $B =$ {$id, RF$}

I now wish to prove that $B$ is a subgroup of $D_4$:

  • Note that $B =$ {$id, RF$}, by definition of B we see it is 'non-empty'

  • $RF RF = id \in B$
    $ \therefore B$ is 'closed under operation'

  • $RF$ and $id$ are 'flips' and are their own inverses, thus:

    $\to$ $RF RF^{-1} = id \in B$

    $\to$ $id id^{-1} = id \in B$

    $ \therefore B$ is 'closed under inverses'

Now although i believe this holds true and accurately proves that $B$ is a subgroup of $D_4$ (correct me if I'm wrong though), i wish to show that $B$ is a subgroup using the definition of cyclic groups and how its properties can prove it's a subgroup. My definition of cyclic subgorup is:

  • "Cyclic Subgroup: if $a_1, a_2, .... ,a_n$ are any finite number of elements in $G$ (an arbitrary group), we define the subgroup generated '$a$' to by a cyclic subgroup (that's to say everything in this subgroup is a multiple or power of the generator) and is denoted $\langle a \rangle$"

any and all help is appreciated as i try to work through this !

$\endgroup$
4
  • $\begingroup$ Use $\{{\rm id}\}$ for $\{{\rm id}\}$ and $\langle a\rangle$ for $\langle a\rangle$. $\endgroup$
    – Shaun
    May 13, 2020 at 23:14
  • 1
    $\begingroup$ @Shaun will do! $\endgroup$ May 13, 2020 at 23:15
  • $\begingroup$ This might come in handy. Also, are you familiar/comfortable with group presentations? I might be able to cook up an answer if so. $\endgroup$
    – Shaun
    May 13, 2020 at 23:19
  • $\begingroup$ @Shaun I am vaguely familiar with them since i'm only at an introductory level of abstract algebra/group theory at the moment, but i understand the syntax and premise of them! $\endgroup$ May 13, 2020 at 23:30

2 Answers 2

1
$\begingroup$

Consider the subgroup $\langle RF\rangle$. You already said that you know $(RF)^2=id$. So now you have that the only elements in $\langle RF\rangle$ are $id$ and $RF$ so it is the same as your “set” $B$. So $B$ is a subgroup and a cyclic one at that.

$\endgroup$
1
$\begingroup$

A presentation for $D_4$ in your notation would be

$$\langle F,R\mid R^4=F^2={\rm id}, FRF=R^{-1}\rangle.\tag{1}$$

Using Tietze transformations, we can introduce a generator $x$ with relation $x=RF$, i.e., $F=R^{-1}x$ to get $(1)$ equivalent to

$$\langle R,x\mid R^4=(R^{-1}x)^2={\rm id}, R^{-1}x^2=R^{-1}\rangle,\tag{2}$$

whose second relation gives $R=xR^{-1}x$ and whose third relation is equivalent to $x^2={\rm id}$, so that we get

$$\langle R,x\mid R^4={\rm id}, R^{-1}=xRx, x^2={\rm id}\rangle\tag{3}$$

from $(2)$. But note that $(3)$ is just like $(1)$; geometrically, $x$ is just another flip of the square other than $F$.

It follows that $\langle x\mid x^2\rangle$ (equivalent to your definition of a cyclic group) is a subgroup of $D_4$, which is what you're after.

$\endgroup$
4
  • $\begingroup$ Does that make sense to you, @AJtheKiddd? I feel like I've skipped a few details. I work with presentations a lot, so I take short cuts. $\endgroup$
    – Shaun
    May 13, 2020 at 23:55
  • 1
    $\begingroup$ wow thank you @Shaun! this took me a few reads through to grasp it all, but it definitely shows the inner workings of this idea very well and explicitly! i have an additional question upon reading this; to enumerate $D_{4/B}$ could it be written as $={[id],[R],[R^{2}],[R^{3}]}$ where each [ ] is a coset? (eg. $[R] = BR = {id, RF}R = {R, R^{2}F} $ ) $\endgroup$ May 14, 2020 at 0:20
  • $\begingroup$ You're welcome, @AJtheKiddd! I'm happy that it helped. I'm not sure how to answer your follow-up question. I'll have to think about it tomorrow. (It's 01:28 here . . . now.) $\endgroup$
    – Shaun
    May 14, 2020 at 0:28
  • $\begingroup$ no worries! i appreciate the help ! $\endgroup$ May 14, 2020 at 0:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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