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

Dummit and Foote leave this as an "easy exercise for verification" and I would like to verify that I am indeed reasoning correctly.

Proof Take an arbitrary element $kH \in K/H$, and let $G/H$ act on $K/H$ by conjugation. $$(gH)(kH)(gH)^{-1} = (gH)(kH)(H^{-1}g^{-1}) = gh_1 kh_2 h_3^{-1} g^{-1} = gh_1kh_4g^{-1}$$ But $H\le K$, and so $h_1 \in H$ is also in $K$, then $h_1 kh_4 \in K$, and $K$ is normal in $G$, so $gKg^{-1} = K$, and thus $(gH)(kH)(gH)^{-1} \in K$.

My Question

How do I recover my coset $kH \in K/H$? I have made an error along the way, and can't see where my error lies.

share|cite|improve this question
The product of elements in $G/H$ is given by $(g_1H)(g_2H) = g_1g_2H$. I'm not sure what you're doing exactly. – Cocopuffs Nov 3 '12 at 15:44
@Cocopuffs jmi4's proof is not efficient, but it is nonetheless correct. – Phira Nov 3 '12 at 15:46
@Phira I see, I was confused by the switch between left cosets and right cosets or elements.. – Cocopuffs Nov 3 '12 at 15:55
up vote 4 down vote accepted

The second equality sign in your proof is not an equality, as you go from a set notation to a notation that gives a generic member of the set. At the end, you did the reverse notational misstep, by writing a set being an element instead of a subset of another set.

So, you start out with a set that you want to be a coset of type $k'H$ with $k' \in K$, (not necessarily the same $k$!) so it is enough to show that your generic element can be written as $k'h'$ with $k' \in K$ and $h' \in H$, but since $H$ is a subgroup of $K$ this is exactly the same thing as showing that the generic element is in $K$ which is precisely what you have done (up to notational problems).

For completeness, let me detail Cocopuffs' comment:

Since $H$ is normal, you can calculate with the cosets "the easy way":

$(gH)(kH)(gH)^{-1} = (gkg^{-1})H= k'H$ where the first equality uses the normality of $H$ and the second equality uses the normality of $K$-

share|cite|improve this answer
Understood. Thanks for the precise explanation – Moderat Nov 3 '12 at 18:52

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.