Consider the following:

Let $A$ and $B$ be groups and $\phi:\ A\ \longrightarrow\ B$ a group homomorphism. Let $T$ be a normal subgroup of $\operatorname{im}\phi$, and let $Q = \lbrace x \in A :\ \phi(x) \in T \rbrace $.

I have managed to show that $Q$ is a normal subgroup of A. Now I conjecture:

$$ \frac{A}{Q} = \frac{\text{im}(\phi)}{T} $$

But I'm not sure how to show this in general.

  • 1
    $\begingroup$ In general, $T$ need not be a subgroup of $\operatorname{im}\phi$, so this cannot be true! $\endgroup$ Sep 23 '15 at 7:36
  • $\begingroup$ I re-read what I had done and I realized I didn't need it to be a normal subgroup of B, but just the image of $\phi$ $\endgroup$ Sep 23 '15 at 7:38
  • $\begingroup$ @Servaes thanks for pointing that out! $\endgroup$ Sep 23 '15 at 7:39

Given the map $\phi$, by the first isomorphism theorem we have an isomorphism $$\overline{\phi}:\ A/\ker\phi\ \longrightarrow\ \operatorname{im}\phi.$$ Because $T$ is a normal subgroup of $\operatorname{im}\phi$ we have a surjective group homomorphism $$\pi:\ \operatorname{im}\phi\ \longrightarrow\ \operatorname{im}\phi/T.$$ Because $\phi$ and $\pi$ are both surjective, also $\psi:=\pi\circ\phi$ is a surjective group homomorphism $$\psi:\ A\ \longrightarrow\ \operatorname{im}\phi/T,$$ so $\operatorname{im}\psi=\operatorname{im}\phi/T$, and $\ker\psi=Q$ by definition. So again by the first isomorphism theorem we have a group isomorphism $$\overline{\psi}:\ A/Q\ \longrightarrow\ \operatorname{im}\phi/T.$$

  • $\begingroup$ This is clever, I didn't think of reusing the first isomorphism theorem to prove its own generalization $\endgroup$ Sep 23 '15 at 7:48
  • 2
    $\begingroup$ I guess this shows that your generalisation is in fact equivalent to the first isomorphism theorem :) $\endgroup$ Sep 23 '15 at 7:49

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.