Isomorphism Theorems

Definition: Let (G1 , o1 ) and (G2 , o2 ) be groups. A map f : G1 → G2 is

called a homomorphism if f (ao1 b) = f (a)o2 f (b) for all a, b ∈ G1 .

• An injective homomorphism is called a monomorphism.

• A surjective homomorphism is called an epimorphism.
• A bijective homomorphism is called an isomorphism.

• A homomorphism from a group (G, o) to itself is called an endomor-

phism.
• An isomorphism from (G, o) to itself is called an automorphism of the
group.

Definition: Let f be a homomorphism from a group (G1 , o1 ) to a group

(G2 , o2 ). Then, f 1 (e2 ) = a ∈ G1 : f (a) = e2 is called the Kernel of f and is
denoted by kerf .
Theorem: (Fundamental Theorem of Homomorphism) Let f be a
surjective homomorphism from a group G1 to a group G2 . Then, f induces
a unique isomorphism f : G1 /kerf → G2 such that the following diagram is
commutative i.e. f = f ◦ ν.

f
G1 G2
f
ν

G1 /kerf

where ν : G1 → G1 /kerf is a quotient homomorphism.

Proof. Let H = kerf . Let h ∈ H and a ∈ G1 . Then f (a ◦1 h) = f (a) ◦2 f (h) =

f (a) ◦2 e2 = f (a). Thus each element of a ◦1 H has image (under f ) f (a).
Therefore f induces a map f : G/H → G0 defined by f (a ◦1 H) = f (a). Since
f is onto, f is also onto.
Next, we show that f is a homomorphism.
Let a, b ∈ G1 . Then f (aH ∗ bH) = f (abH) = f (a ◦1 b) = f (a) ◦2 f (b) =
f (aH) ◦2 f (bH). Thus, f is a homomorphism. Next, we show that f is one-one.
For this, let a ∈ G1 such that a ◦1 H ∈ kerf . Then (f (a) =)f (a ◦1 H) = e2 .
This implies that a ∈ kerf = H. So, a ◦1 H = H = id of G1 /H. Therefore, f
is one-one. Thus, f is an isomorphism.
Now, let a ∈ G1 , then (f ◦ ν)(a) = f (ν(a)) = f (a ◦1 H) = f (a). Thus,
f ◦ ν = f.
For the uniqueness, let φ : G1 /H → G2 be also a homomorphism such that
φ ◦ ν = f . Then, for any a ∈ G1 , f (a ◦1 H) = f (ν(a)) = (f ◦ ν)(a) = f (a) =
(φ ◦ ν)(a) = φ(ν(a)) = φ(a ◦1 H). Thus, f = φ i.e. f is unique.
Theorem: (First Isomorphism Theorem) Let f be a surjective homo-
morphism from G1 to G2 and H1 E G1 such that kerf ⊆ H1 . Let H2 = f (H1 ).

1
Then f induces a unique isomorphism f : G1 /H1 → G2 /H2 such that the fol-
lowing diagram is commutative, i.e. f ◦ ν1 = ν2 ◦ f , where ν1 and ν2 are the
corresponding quotient maps.

f
G1 G2
ν1 ν2

f
G1 /H1 G2 /H2

Proof. Since f is onto and H1 E G1 , hence H2 = f (H1 ) is normal in G2 .

Consider the map φ = ν2 ◦f . Since f and ν2 both are surjective homomorphism,
hence φ is also surjective homomorphism. Now, kerφ = {g ∈ G1 : φ(g) = H2 } =
{g ∈ G1 : ν2 (f (g)) = H2 } = {g ∈ G1 : f (g) ◦2 H2 = H2 } = {g ∈ G1 : f (g) ∈
H2 } = f −1 (H2 ) = H1 . Thus, by Fundamental theorem of homomorphism, φ
induces a bijective homomorphism f : G1 /H1 → G2 /H2 such that φ = f ◦ ν1 =
ν2
◦f .
Corollary: Let H and K be normal subgroups of G such that H ⊆ K.
Then, K/H E G/H and (G/H)/(K/H) ∼= G/K.
Proof. Let ν : G → G/H be the quotient homomorphism. Then kerν = H ⊆ K
and ν(K) = {ν(k) : k ∈ K} = {kH : k ∈ K} = K/H. Therefore, by First
Isomorphism theorem, G/K ∼= (G/H)/(K/H).
Theorem: (Noether 2nd Isomorphism Theorem) Let H and K be
subgroups of G and K E G. Then, H ∩ K E H and H/(H ∩ K) ∼
= HK/K.
Proof. Let g ∈ H and h ∈ H ∩ K. Then ghg −1 ∈ K, because K is normal
in G. Further, g, h ∈ H and H is a subgroup, hence ghg −1 ∈ H. Thus,
ghg −1 ∈ H ∩ K and so H ∩ K E H. Now, consider the map f : H → HK/K
defined by f (h) = hK. Clearly, f is onto map. Next, since f is the restriction
of the quotient homomorphism ν : G → G/K, hence f is a homomorphism.
Now, kerf = {h ∈ H : f (h) = K} = {h ∈ H : hK = K} = {h ∈ H :
h ∈ K} = H ∩ K. Therefore, by Fundamental theorem of homomorphism
H/(H ∩ K) ∼ = HK/K.
Applications of the Isomorphism theorems
1. A cyclic group of order n is isomorphic to Zn .
Hint: Let G be a cyclic group of order n and a ∈ G be a generator. Define
a map f : Z → G by f (k) = ak . Then f is a homomorphism (verify) and
kerf = {k ∈ Z : f (k) = ak = e} = {k ∈ Z : n = o(a)|k} = {k ∈ Z : k = nu},
for some u ∈ Z = nZ. Thus, by fundamental theorem of homomorphism,
Z/nZ ∼ = G.
2. Show that R/Z ∼ = S 1 , where R is the additive group of real numbers.
Hint: Define a map f : R → S 1 by f (x) = e2πix . Then, f is a surjective
homomorphism (verify) and kerf = {x ∈ R : e2πix = 1} = Z. By fundamental
theorem of homomorphism, R/Z ∼ = S1.
3. Show that R/Q ∼ 1
= S /P , where R, Q are the additive groups of real
numbers and rational numbers respectively and P is the group of roots of unity.

2
 Hint: Define a map f : R → S 1 by f (x) = e2πix . Verify that f is an onto
homomorphism. Let H = Q, then f (H) = P . Thus, by First isomorphism
theorem, R/Q ∼
= S 1 /P .

Symmetric Group

Definition: Let X be a set. Let Sym(X) denote the set of all bijective
maps from X to X. Then, Sym(X) is a group with respect to the composition
of maps. This group is called the Symmetric group or the Transformation
group or the Permutation group on X.
If X = {1, 2, . . . , n}, then Sym(X) is denoted by Sn and is called Symmetric
group of degree n. Thus, Sn is the group of all bijective maps from {1, 2, . . . , n}
to itself, the operation being the composition of maps. An element f ∈ Sn can
be represented by
 
1 2 3 ··· n
f (1) f (2) f (3) · · · f (n)

Definition: A permutation α ∈ Sn is called a cycle of length r ≥ 1 if there

exists a subset {i1 , i2 , . . . , ir } of {1, 2, . . . , n} containing r distinct elements such
that α(i1 ) = i2 , α(i2 ) = i3 , . . . , α(ir1 ) = ir , α(ir ) = i1 and α(j) = j for all
j ∈ {1, 2, . . . , n} − {i1 , i2 , . . . , ir }. The cycle α is denoted by (i1 i2 . . . ir ). A
cycle of length 2 is called a transposition.
Two cycles (i1 i2 . . . ir ) and (j1 j2 . . . js ) are said to be disjoint if {i1 , i2 , . . . , ir }∩
{j1 , j2 , . . . js } = ∅
Proposition: Any two disjoint cycles commute.
Proposition: If α and β are disjoint cycles, then < α > ∩ < β >= {I}.
Proposition: Let α ∈ Sn be a cycle of length r. Then, o(α) = r.
Proposition: Let {1 , 2 , . . . , r } be a set of pairwise disjoint cycles of lengths
m1 , m2 , . . . , mr respectively. Let = 12 . . . r . Then, o() = lcm[m1 , m2 , . . . , mr ]
the least common multiple of m1 , m2 , . . . mr .
Theorem: Every nonidentity permutation can be written as a product of
disjoint cycles. Further, any two representations of a nonidentity permutation
as product of disjoint cycles is same up to rearrangement of cycles.
Proof. Let f be a nonidentity permutation in Sn . Then, there exists i1 ∈
{1, 2, . . . , n} such that f (i1 ) 6= i1 . Since o(Sn ) = n!, f n! = I. Hence, f n! (i1 ) =
i1 . Let l1 be the least positive integer such that f l1 (i1 ) = i1 . Thus, i1 , f (i1 ), f 2 (i1 ), . . . , f l1 −1(i1 )
are all distinct. Now, given k ∈ Z, by the division algorithm, there exist q, r ∈ Z
such that k = l1 q + r, where 0 ≤ r ≤ l1 − 1. But, then f k (i1 ) = f ql1 +r (i1 ) =
f r (i1 ). It is clear from the above observation that the effect of the permuta-
tion f on symbols in {i1 , f (i1 ), f 2 (i1 ), . . . , f 11 1 (i1 )} is same as that of the cycle
C1 = (i1 f (i1 )f 2 (i2 ) . . . f l1 1 (i1 )). If f = C1 , there is nothing to do. If not, there
exists i2 ∈ {1, 2, . . . , n} − {i1 , f (i1 ), f 2 (i1 ), . . . f l1 1 (i1 )} such that f (i2 ) 6= i2 .
As before consider the cycle C2 = (i2 f (i2 )f 2 (i2 ) . . . f l2 1 (i2 )), where l2 is the
smallest positive integer such that f l2 (i2 ) = i2 . Clearly, C1 and C2 are disjoint
cycles. If f = C1C2, there is nothing to do. If not proceed. This process

3
stops after finitely many steps giving f as product of disjoint cycles, because
the symbols are finitely many. Finally, we prove the uniqueness. Suppose that
f 6= I and f = C1 C2 . . . Cr = C10 C20 . . . Cs0 , where Ci and Cj are disjoint for
i 6= j , and also Ck0 and Cl0 are disjoint for k 6= l. Let i ∈ {1, 2, . . . , n} such that
f (i) 6= i. Then, there exist k, l(1 ≤ k ≤ r, 1 ≤ l ≤ s) such that Ck (i) 6= i and
also Cl0 (i) 6= i. Since a cycle is completely determined by any symbol moved by
it, hence Ck = Cl0 . Since disjoint cycle commutes, we may assume that Ck = C1
and Cl0 = C10 . Canceling C1 and C10 , using induction and the fact that prod-
ucts of nonidentity disjoint cycles can never be identity, we find that r = s and
Ci = Ci0 for all i.
 
1 2 3 4 5 6 7 8 9 10
Examples: 1. = (16103827459) a
6 7 8 5 9 10 4 2 1 3
single cycle.
 
1 2 3 4 5 6 7 8
2. = (13)(4567)
3 2 1 5 6 7 4 8

Exercise:
• S7 contains no elements of order 8. For, a permutation is of order 8 if and
only if it can be written as product of disjoint cycles such that the least
common multiple of lengths of these cycles equals 8. We also observe that
l.c.m of certain numbers equals 8 only if at least one of them is 8. Thus,
we cannot have any such permutation in S7 . Similarly, S7 contains no
element of order 15, for a permutation of order 15 contains at least one
cycle of length 5 and at least one cycle of length 3, or it should contain a
cycle of length 15 in its decomposition as product of disjoint cycles.
• Does S6 contains an element of order 10?

• Does S14 , S13 , S12 has an element of order 42?

Definition: Let and be two nontrivial permutations in Sn . We say that
and are of same form if
(i) the number of cycles in the representation of as product of disjoint cycles is
same as that in the representation of as product of disjoint cycles, and
(ii) there is a bijective correspondence between the set of cycles used in the
representation of and that in the representation of so that the lengths of the
corresponding cycles are same.
Thus, (123)(47)(5698) and (4523)(89)(167) are permutations of same form
where as (123)(47) and (4523)(89) are not of the same form.
Theorem: Two cycles in Sn are conjugate if and only if they have same
length.
Theorem: Two permutations in Sn are conjugate if and only if they are of
same form.