Consider $f: \mathbb Z_8 \rightarrow \mathbb Z_4$ given by

$$\begin{pmatrix} 0&1&2&3&4&5&6&7 \\ 0&1&2&3&0&1&2&3 \end{pmatrix}$$

Verify the $f$ is a homomorphism, find its kernal $K$, and list the cosets of $K$.

I have verified that this is a homomorphism by a Cayley Table, and the kernal $K=\{0,4\}$.

Now it says list the cosets of $K$. I am not sure of this.

Is the coset of $K$ the elements that $f$ does not map into the identity, i.e.: $\{1,2,3,5,6,7\}$?

Or, since the kernal of $f$ is a normal subgroup of the domain, in this case $\mathbb Z_8$ (i.e.: $K$ is subgroup of $\mathbb Z_8$), are the cosets of the kernal equal to $Ka$, where $a$ is any element of $\mathbb Z_8$?

  • 3
    $\begingroup$ Note the spelling "Kernel" not "Kernal". $\endgroup$ – Mark Bennet Nov 7 '14 at 20:57

Since $|ker|=2$, so there will be $4$ elements in the coset. Now follow the usual rule for forming cosets of a subgroup, i.e., $aH=\{ah\mid a\in G\}$. Thus for this case the coset corresponding to the element $0,4$ is $K$ itself. For example, for $1$ its $1+K=\{1,5\}$. In a similar way cosets for other elements of $\mathbb{Z}_8$ can be formed, keeping in mind the operation on $\mathbb{Z}_8$. See that the cosets corresponding to $1$ and $5$ and $3$ and $7$ are the same.


Hint: Note that the permutation corresponds to the map

$f:a\to a(\mod 4)$

Clearly, Kernel of the map is elements of $\mathbb{Z}_8$ which map to $0$ which are $0$ and $4$, which you have already written. $K=\{ 0,4\}$

What are cosets? Suppose $N$ is a normal subgroup of $G$

Then $G/N=\{ a\oplus_8N : a\in G\}$. Note $\oplus_8$ is addition modulo 8.

The distinct cosets are

$\mathbb{Z}_8/K=\{ K,1\oplus_8K,2\oplus_8K,3\oplus_8K\}$

The number of distinct cosets is given by $|\mathbb{Z}_8|/|K|$


$K=\{ 0,4\}$

$1\oplus_8K=\{ 1,5\}$

$2\oplus_8K=\{ 2,6\}$

$3\oplus_8K=\{ 3,7\}$

  • $\begingroup$ That doesn't help me find the cosets of the kernal. I am confused as to the what is a coset of the kernal. I have only seen cosets of a subgroup of a group. $\endgroup$ – Al Jebr Nov 7 '14 at 20:23
  • $\begingroup$ I'll edit accordingly. $\endgroup$ – Swapnil Tripathi Nov 7 '14 at 20:24
  • $\begingroup$ @JohannFranklin The Kernel of a homomorphism is a subgroup - so there is no new kind of coset here. $\endgroup$ – Mark Bennet Nov 7 '14 at 20:56

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.