# The Center of a Group as the centralizer of a subgroup.

• May 25th 2010, 12:35 PM
davismj
The Center of a Group as the centralizer of a subgroup.
http://i50.tinypic.com/2yud85t.png

I'm sure this is true. I'm not sure this is the only such subgroup, and I'm not sure this is the subgroup they're thinking of. What do you think?

If notation isn't clear:

$N(g) = \{n \in G : ng = gn\}$
• May 25th 2010, 06:13 PM
tonio
Quote:

Originally Posted by davismj
http://i50.tinypic.com/2yud85t.png

I'm sure this is true. I'm not sure this is the only such subgroup, and I'm not sure this is the subgroup they're thinking of. What do you think?

If notation isn't clear:

$N(g) = \{n \in G : ng = gn\}$

But what $g\in G$ do you choose for $N(g)$ ?!? This should make it clear your choice isn't correct...

I think the answer is most trivial: $Z(G) =C(G)$ , so the center of a group G is the centralizer of the subgroup G of G...**sigh**...yes, I know: it's lame.

Tonio
• May 25th 2010, 07:38 PM
davismj
Quote:

Originally Posted by tonio
But what $g\in G$ do you choose for $N(g)$ ?!? This should make it clear your choice isn't correct...

I think the answer is most trivial: $Z(G) =C(G)$ , so the center of a group G is the centralizer of the subgroup G of G...**sigh**...yes, I know: it's lame.

Tonio

Sorry, I meant for all g in G. That is, for any n that is not in any N(g). I think thats true, since if n is in N(g), ng = gn for some g in G, so the result is trivial.

Your answer is obvious. It's certainly true, but I hardly see why they'd want do highlight this particular point. But it is the most logical choice.
• May 26th 2010, 03:47 AM
tonio
Quote:

Originally Posted by davismj
Sorry, I meant for all g in G. That is, for any n that is not in any N(g). I think thats true, since if n is in N(g), ng = gn for some g in G, so the result is trivial.

I think you're confused (or perhaps I am: go figure!) : you define $N(g):=\{n\in G\;;\;ng=gn\}=:C(g)=$ the centralizer of the element $g\in G$ (try no to use the letter N for this since it is used for another thing, the normalizer). But an element $z\in Z(G)$ belongs to the centralizer of each and every element of G! So it can't be that an element in the center is NOT in some element's centralizer

Tonio

Your answer is obvious. It's certainly true, but I hardly see why they'd want do highlight this particular point. But it is the most logical choice.

.
• May 26th 2010, 07:05 AM
davismj
Quote:

Originally Posted by tonio
.

After looking up wikipedia, it seems my book wanted me to learn them backwards. Here is what my book indicated to me:

The normalizer of a in G is a set $N(a) = \{g \in G : ag = ga\}$

The centralizer of a subset $H \subset G$ is a set $C(H) = \{g \in G : gh = hg, \forall h \in H\}
$

According to Wikipedia, I've got this backwards.