Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Homework 7

2.5.7 Claim: If N ≤ G then the following are equivalent.

(a) N is normal.

(b) Each left coset of N is also a right coset.

(c) For each a ∈ G, aN = N a.

Proof: First we will show that if N is normal, then each left coset of N is also a right coset
of N . Suppose a ∈ G, and consider the left coset aN . Since N is normal, aN a−1 = N .
That is, the set aN a−1 = {ana−1 | n ∈ N } = N . Now consider the set aN = {an | n ∈ N }.
We know ana−1 = n0 for some n0 ∈ N . Hence an = n0 a. Therefore, aN ⊆ N a. Now
suppose we take an element na ∈ N a. Then since a−1 na = a−1 n(a−1 )−1 = n00 for some
n00 ∈ N , na = an00 . So N a ⊂ aN . This shows the left coset aN is also a right coset N a.
Now suppose we know that N is a subgroup of G and all its left cosets are also right
cosets. Say we have a left coset aN . Then there exists b ∈ G so that aN = N b. Since
a ∈ aN , for some n ∈ N , a = nb. Hence b−1 a = n ∈ N , and so N b = N a. Therefore
aN = N a.
Finally, suppose we know that aN = N a for all a ∈ A. Then if n ∈ N there is an
n inN so that an = n0 a. This implies that an−1 = n0 ∈ N so aN a−1 ⊆ N . Similarly, if
0

n ∈ N , then a−1 na ∈ N by the previous argument, so n = a(a−1 na)a−1 ∈ aN a−1 . This


shows that aN a−1 = N and N is normal.

2.5.13

(a) Claim: The center of a group G is a normal subgroup of G

Proof: Let Z(G) be the center of a group G. First, we perform the subgroup test
to determine that Z(G) is a subgroup of G. Z(G) is nonempty, since e ∈ Z(G). Let
a, b ∈ Z(G), and g ∈ G. Then (ab)g = a(bg) by associativity. And a(bg) = a(gb) since
b ∈ Z(G). Further, a(gb) = (ag)b = (ga)b by associativity, and using that a ∈ Z(G).
Hence (ab)g = g(ab), and ab ∈ Z(G).
This shows that Z(G) is closed under multiplication. Now we check that Z(G) is closed
under inverse taking. If a ∈ Z(G) and g ∈ G, then a−1 g = (g −1 a)−1 = (ag −1 )−1 since
a ∈ Z(G). And (ag −1 )−1 = ga−1 . Hence a−1 ∈ Z(G). This shows that Z(G) is a
subgroup of G.

1
Now we show that Z(G) is normal. Let g ∈ G. Then gZ(G)g −1 = {gag −1 | a ∈ Z(G)}.
But gag −1 = agg −1 = a since a ∈ Z(G). Hence gZ(G)g −1 = {a | a ∈ Z(G)} = Z(G).
This proves that Z(G) is normal.

(b) Claim: Consider the group S3 , whose elements are e, (12), (13), (23), (123), (132).
First, e ∈ Z(G) since the identity element of a group is always in the center of the group.
Next, we see (12) and (23) are not in Z(S3 ) since (12)(23) = (123) and (23)(12) = (132).
Similarly, (13) is not in Z(S3 ) since (13)(12) = (123) and (12)(13) = (132). Next, (123)
is not in Z(S3 ) since (12)(123) = (23) and (123)(12) = (13). And (132) is not in Z(S3 )
since (13)(132) = (23) and (132)(13) = (12). Therefore, Z(S3 ) = {e}.

2.6.1 Claim: If f : X → Y is a surjective map, then the relation x1 ∼ x2 if and only


if f (x1 ) = f (x2 ) is an equivalence relations. This relation partitions X into “fibers”
f −1({y}).

Proof: Suppose f : X → Y is a surjective map. First, f (x1 ) = f (x1 ) so x1 ∼ x1 . Next,


if f (x1 ) = f (x2 ) then f (x2 ) = f (x1 ), so x1 ∼ x2 implies that x2 ∼ x1 . Finally, if x1 ∼ x2
and x2 ∼ x3 , then f (x1 ) = f (x2 ) = f (x3 ), so x1 ∼ x3 . Hence ∼ is an equivalence relation
on X.
Let x ∈ X. The equivalence class of x is the set {z ∈ X | f (z) = f (x)}. Since f is
surjective, for each y ∈ Y there is an x ∈ X such that f (x) = y. Hence the equivalence
classes of X correspond to the points of Y under the one-to-one map y ∈ Y goes to
f −1 ({y}) = {x ∈ X | f (x) = y}, which are called the fibers of Y .

2.7.9 Claim: Let G be a group, and let C be the subgroup generated by all elements of
the form xyx−1 y −1 where x, y ∈ G. Then C is normal and G/C is abelian. Further, if H
is a normal subgroup of G and G/H is abelian, then C ⊆ H.

Proof: The elements of C are products of commutators. That is, c ∈ C can be written
c = c1 c2 · · · cn where ci = xi yi x−1 −1
i yi . We must show that for any g ∈ G, gCg
−1 = C.

It is enough to show that gCg −1 ⊆ C. Let c ∈ C, and consider gcg −1 . Then gcg −1 =
gc1 c2 · · · cn g −1 = (gc1 g −1 )(gc2 g −1 ) · · · (gc−1
n ), so we must show that for each i, gci g
−1 =
−1 −1 −1
gx1 yi xi yi g ∈ C.
We can rewrite gxi yi x−1 −1 −1
i yi g = (gxi g −1 )(gyi g −1 )(gx−1 −1 −1 −1
i g )(gyi g ) be inserting
products gg −1 = e into the original product. Then u = gxi g −1 and v = gyi g −1 are
elements of G, so gci g −1 = uvu−1 v −1 ∈ C. This shows C is normal.
Now consider the quotient G/C. Let a, b ∈ G, and consider the product aCbC = (ab)C.
We would like to show that (ab)C = (ba)C so that G/C is abelian. This holds if and only
if (ab)(ba)−1 ∈ C. But (ab)(ba)−1 = aba−1 b−1 ∈ C since it is a commutator. So G/C is
abelian.

2
Finally, suppose H is a normal subgroup of G, and G/H is abelian. Then if x, y ∈ G,
xHyH = yHxH. This holds if and only if xy(yx)−1 ∈ H. That is, each commutator
xyx−1 y −1 must be in H. Since C is the group that contains exactly all commutators and
their products, C ⊆ H.

You might also like