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

Theorem: Let $G$ and $G'$ be groups and let $f:G\to G'$ be a group homomorphism. Then $G/\textrm{ker}\, f \cong\textrm{im}\, f$.

My question is how to understand this theorem intuitively.

share|cite|improve this question
The most intuitive way I ever got to is "all short exact sequences really look the same". – Alexei Averchenko Oct 27 '12 at 8:05
1… – JavaMan Oct 27 '12 at 8:07
But I am thinking what does it mean that we are moding out kernel from G and then it is isomorphic to image how to think about that? – Reader Oct 27 '12 at 8:10
up vote 7 down vote accepted

The intuition is that the group looks similar to its image under the homomorphism when it is divided by a certain subgroup, since distinct elements may get mapped to the same element. By taking the quotient, elements which are mapped to the same are grouped together.

share|cite|improve this answer
By why are we just moding out kernel, is it enough to mod out kernel. Because you can also have that $x\mapsto p$ and $y\mapsto p$ what about this, then we are not moding out these elements? Why? – Reader Oct 27 '12 at 8:18
@Reader: In that case $xy^{-1} \in \ker f$. – wj32 Oct 27 '12 at 8:24
And since $Ker f$ is a group so we are also moding out xy^{-1} this element. It means that we are moding out all elements which have same image in $G'$. Therefore it make sense that $G/ker\, f\cong im \, f.$ Thanks – Reader Oct 27 '12 at 8:28
Now, I don not understand one thing, why $xy^{-1}\in ker f$ – Reader Oct 27 '12 at 8:32
@Reader: $f(xy^{-1})=f(x)f(y)^{-1}=pp^{-1}=e$. Another way of looking at it is to notice that $x$ and $y$ are in the same coset in $G/\ker f$, so they are grouped together because as you said, they have the same image. – wj32 Oct 27 '12 at 8:40

Maybe it gets more intuitive if you look at the situation with groups replaced by sets. If $f : M \rightarrow N$ is a map between sets $M$ and $N$, then you get an equivalenve relation $R_f := \{(x,y) \in M \times M \::\: f(x) = f(y)\}$. The equivalence class of $x \in M$ is $f^{-1}(\{f(x)\})$, so we have a well-defined (and injective) map from $M/R_f$ (the set of equivalence classe) to $N$, sending $f^{-1}(\{f(x)\})$ to $f(x)$. Thus, first collecting all elements, which map to the same element via $f$, and then mapping these collections to their respective values gives you a bijection onto the image.

Note that in the case of groups $G$ and $G'$ and group homomorphism $f: G \rightarrow G'$ you have $f^{-1}(\{f(x)\}) = x\ker(f)$ for all $x \in G$ and the induced map is a group homomorphism, yielding an isomorphism to the image.

share|cite|improve this answer
Nice solution. First we collect elements which have same image and then mapping. – Reader Oct 27 '12 at 8:37
Yes. In nice structures just as groups, vector fields, or modules over a ring, the preimage of the unit element (or $0$ if grp operation is $+$) uniquely determines the preimage of all other elements. – Berci Oct 27 '12 at 10:53
Basically, all you need is a Mal'cev operation. – Zhen Lin Oct 27 '12 at 11:30

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.