0
$\begingroup$

From Hungerford's "Algebra":

enter image description here

What type of tools does one have to tackle a problem like this? I seem at a loss at how to show a group is free at all. One can consider each group as the homomorphic image of a free group, but how does one work in the other direction?

$\endgroup$
  • 1
    $\begingroup$ I would use universal properties. That is, for any group $G$ and function $f:X\setminus Y \to G$, I would show that $f$ extends to a unique homomorphism $F \to G$ whose kernel contains $Y$, hence $H$. $\endgroup$ – Slade Oct 13 '14 at 0:11
1
$\begingroup$

The group $F/H$ is (isomorphic to) the free group on $X \setminus Y$, since it has the correct universal property: If $G$ is a group with underlying set $|G|$, then there are natural bijections $$\hom(F/H,G) \cong \{\phi \in \hom(F,G) : H \subseteq \ker(\phi)\}$$ $$= \{\phi \in \hom(F,G) : Y \subseteq \ker(\phi)\} \cong \{f \in \hom(X,|G|) : f|_Y = 1_G\}$$ $$ \cong \hom(X \setminus Y,|G|).$$

$\endgroup$
-1
$\begingroup$

Take any function $X \setminus Y \to G$ where $G$ Is an arbitrary group. Then extend it is any way you want to a function $X \to G$. This induces a unique group homomorphism $F \to G$ which in turn induces a group homomorphism $F/H \to G$. A bit of work shows that this extension from $X \setminus Y$ to $F/H$ is unique, so the universal property of the free group is satisfied. So $F/H$ is free on $X \setminus Y$.

This all makes sense since we $X \setminus Y$ is (morally speaking) its own image in $F/H$.

$\endgroup$
  • $\begingroup$ This is not correct. $X \to G$ has to send $Y$ to $1_G$, only then we get an extension to $F/H \to G$. Also, this "bit of work" uses universal properties anyway. $\endgroup$ – Martin Brandenburg Oct 13 '14 at 0:44

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.