I found the following exercise on the 12th page of Presentations of Groups:

  1. Let $F$ be free of rank $r$. Show that $F/F'$ is free abelian of rank $r$.

First of all, $F'$ is not defined. I'm assuming it stands for a general normal subgroup of $F$.

So, what seems very strange here is that $F$ is not supposed to be abelian, but if I prove that $F/F'$ has the same rank that $F$ has, then, $F/F'$ and $F$ are isomorphic, right? But how can a non-necessarily abelian group be isomorphic to an abelian one?

It seems that someone has already posted a question here related to this: Let free group, F and F/F' have same rank.

But the answers don't address the issue I'm pointing out here.

  • $\begingroup$ This says that $F_r/[F_r,F_r]\cong \mathbb{Z}^r$, where $F_r$ denotes the free group of rank $r$, i.e., with $r$ generators. Of course, $F_r$ is not isomrophic to $\mathbb{Z}^r$ for $r>1$. $\endgroup$ – Dietrich Burde Jun 27 '17 at 13:46

$F'$ denotes the derived subgroup of $F$, which is just another name for the commutator subgroup $[F,F]$. Thus $F/F'$ is just the abelianization of $F$. This should help you understand the answers in the other thread, or alternatively the answers to the question Abelianization of free group is the free abelian group.

  • $\begingroup$ Ok. I got it. On the other hand, it is still strange that this factor group is abelian and isomorphic to $F$. If two groups are isomorphic, they are both abelian or both non-abelian, right? $\endgroup$ – Hilder Vitor Lima Pereira Jun 27 '17 at 13:35
  • 3
    $\begingroup$ Yes, but $F/F'$ and $F$ are not isomorphic. The point is that $F$ is free of rank $r$ and $F/F'$ is free abelian of rank $r$. These are two different universal properties, so it doesn't imply that they must be isomorphic. $\endgroup$ – Arnaud D. Jun 27 '17 at 13:38
  • $\begingroup$ Ok, I see. The book didn't explicitly defined free abelian groups, just made a comment about that. And it now even comment about commutator subgroup. So I think this is why I get confused on that. Thank you for clarifying those points. $\endgroup$ – Hilder Vitor Lima Pereira Jun 27 '17 at 13:47

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.