# Abstract Nonsense

## Review of Group Theory: The Commutator Subgroup and the Abelianization of a Group

Point of post: In this post we discuss the notion of the commutator subgroup of a group $G$ and prove some basic, yet important, facts about it. We then discuss the abelianization of a group (the group mod the commutator subgroup).

Motivation

Very often in group theory we are dealt groups which, to our consternation, are not abelian. That said, being abelian and not abelian is often too black-and-white of a measure of commutativity. Thus, one may ask if there is some in-between quality that measures the ‘abelianess’ or lack thereof of the group. There is–this is the commutator subgroup. In essence the commutator subgroup is precisely such that when the ambient group is quotiented out by it everything of the form $xyx^{-1}y^{-1}$ goes to the identity. This makes sense since in a perfect world, an abelian world that is, everything of this form would already be the identity. Thus, quotienting out by the subgroup generated by the set of all $xyx^{-1}y^{-1}$ (i.e. the commutator subgroup) collapses all the defiant breakers of the golden abelian rule. Thus, in essence the bigger the commutator subgroup is the further from abelian the group is.

$\text{ }$

Commutator Subgroup

Let $G$ be any group. We define the map $[\cdot,\cdot]:G\times G\to G$ given by $[x,y]=xyx^{-1}y^{-1}$ to be the commutator bracket. Then the subgroup generated by the set $\left\{[x,y]:x,y\in G\right\}$ is called the commutator subgroup of $G$ and is denoted $\left[G,G\right]$. Our first claim is that $[G,G]$ is a normal subgroup of $G$. Indeed it clearly suffices to check that $g[x,y]g^{-1}\in [G,G]$ for every $g,x,y\in G$. Recall the inner automorphism $i_g$ and the fact that it is, surprisingly, an automorphism. Thus, one sees that $gxyx^{-1}y^{-1}g^{-1}=i_g(xyx^{-1}y^{-1})=i_g(x)i_y(y)i_g(x)^{-1}i_g(y)^{-1}\in [G,G]$ and since $g,x,y\in G$ were arbitrary it follows that $[G,G]\unlhd G$ as claimed.

Abelinization

Now that we know $[G,G]\unlhd G$ we may speak freely of quotient groups. We call the quotient group $G/[G,G]$ the abelinization of $G$ and denote it $G^{\text{ab}}$. The first interesting theorem pertaining to $G^{\text{ab}}$ is:

$\text{ }$

Theorem: Let $G$ be any group and $N\unlhd G$. Then, $G/N$ is abelian if and only if $[G,G]\leqslant N$.

Proof: Suppose first that $[G,G]\leqslant N$ and let $\pi:G\to G/N$ be the canonical projection. We note then that for any $x,y\in G$ one has that $\pi(x)\pi(y)\pi(x)^{-1}\pi(y)^{-1}=\pi\left(xyx^{-1}y^{-1}\right)=N$ (since $xyx^{-1}y^{-1}\in[G,G]\subseteq N$) and so $\pi(x)\pi(y)=\pi(y)\pi(x)$. Then, since $\pi$ is surjective the conclusion follows.

$\text{ }$

Conversely, suppose that $G/N$ is abelian. Then, for any $x,y\in G$ one has that $\pi(x)\pi(y)=\pi(y)\pi(x)$ or
$\pi\left(xyx^{-1}y^{-1}\right)=N$ which is true if and only if $xyx^{-1}y^{-1}\in N$. Since $x,y\in G$ were arbitrary it follows that $[G,G]\subseteq N$ as desired. $\blacksquare$

$\text{ }$

Corollary: $G^{\text{ab}}$ is abelian.

$\text{ }$

$\text{ }$

The really interesting theorem regarding $G^{\text{ab}}$ though is:

Theorem: Let $G$ and $H$ be any two groups. Then a homomorphism $\phi:G\to H$ has the property that $\text{im}(\phi)\leqslant H$ is abelian if and only if $\phi$ can be factored through $\text{G}^{\text{ab}}$ by $\pi$, in other words if and only if there exists a homomorphism $\theta:\text{G}^{\text{ab}}\to H$ such that $\phi=\theta\circ\pi$.

Proof: Clearly if $\phi$ can be factored in such a way then $\text{im}(\phi)=\text{im}(\theta)$ and $\text{im}(\theta)$ is the image of an abelian group under a homomorphism, and thus trivially abelian.

$\text{ }$

Conversely, suppose that $\text{im}(\phi)$ is abelian. Consider then the map $\displaystyle \theta:G^{\text{ab}}\to H$ by $g[G,G]\mapsto \phi(g)$. This is well-defined since if $g[G,G]=g'[G,G]$ then $g'=gk$ for some $k\in [G,G]$. But, since $\text{im}(\phi)$ is abelian it’s fairly easy to see that $\phi(k)=e$ and so $\phi(g')=\phi(g)$. Furthermore, it’s evident that $\theta$ is a homomorphism and $\phi=\theta\circ\pi$. $\blacksquare$

$\text{ }$

$\text{ }$

References:

1.  Dummit, David Steven., and Richard M. Foote. Abstract Algebra. Hoboken, NJ: Wiley, 2009

February 27, 2011 -

1. […] Recall from our discussion of the abelinization of a group that for every group and every homomomorphism such that is […]

Pingback by Representation Theory: The Number of Degree One Irreps « Abstract Nonsense | February 28, 2011 | Reply

2. […] Also, we’ve proven that the number of degree one irreps of is equal to the order of the abelinization of . In this post we’ll prove the supremely interesting result that the degree of any irrep […]

Pingback by Representation Theory: The Dimension Theorem « Abstract Nonsense | March 3, 2011 | Reply

3. […] be a group. Recall the definition of the commutator subgroup of a group and recursively define by and . We call then the […]

Pingback by Review of Group Theory: Solvable Groups « Abstract Nonsense | March 11, 2011 | Reply

4. […] a last note, we know that where is the abelianization of and is the commutator […]

Pingback by Representation Theory: The Linear Group of a Group G and the Dual Group « Abstract Nonsense | April 12, 2011 | Reply

5. […] We recall that the dual group  has order equal the the abelianization . But, by inspection we see that and so and so (the commutator subgroup). It evidently follows […]

Pingback by University of Maryland College Park Qualifying Exams (Group Theory and Representation Theory) (August-2003) « Abstract Nonsense | May 1, 2011 | Reply

6. […] he number of degree one equivalence classes of irreps of a finite group is equal to where is the abelianization of . This proof followed pretty closely the proof in Barry Simon’s book, and to be honest I […]

Pingback by Clearer Proof for the Number of Degree One Irrep Classes « Abstract Nonsense | May 6, 2011 | Reply

7. […] be more specific, let us denote this functor by . Then, for one has that is nothing more than the abelianization of , and given in we can get a map by noting that by composing with the canonical projection we […]

Pingback by Functors (Pt. I) « Abstract Nonsense | December 27, 2011 | Reply

8. […] functor  is right adjoint to the abelianization functor sending a group to its abelianization and sending a group map to the map we get from the characterization of applied to the […]

Pingback by Adjoint Functors (Pt. II) « Abstract Nonsense | April 15, 2012 | Reply