CHAPTER ONE

Groups

1.1 Definition and Examples

It takes patience to appreciate the diverse ways in which groups arise, but one of these ways is
so familiar that we can use it to ease our way into the basic definition. To this end, recall the
following three things about the set of integers with respect to the operation addition. First
addition is associative. Second, 0 is an identity. And third, relative to 0, each integer has an
inverse (its negative). They show that the integers with addition form a group, in the sense of
the following definition.
1.1.1 Definition We say that a system (G, *), consisting of a non-empty set G and a binary
operation * on a G is a group if the following four axioms are satisfied:
G1: a, b  g  a* b  G … (Closure)
G2: a, b, c  G, (a * b) * c = a * (b * c) … (Associativity)
G3: e  G such that, a  G, a * e = a = e * a …(Existence of identity )
G4: a  G,  a  G, such that a * a = e = a * a … (Existence of inverses)

1.1.2 Definition A group (G, *) is said to be abelian (or commutative) if the following axiom
is also satisfied:
G5: a, b  G, a * b = b * a …. (Commutativity)

1.1.2 Examples of groups

a) (Z, +) is a group, Where Z is the set of integers. Because, the sum of any two
integers is an integer (Z, .) is not a group, Axiom G4 fails.
b) (Q+, .) is a group.

1.2 Simple Properties of Groups

1.2.1 Theorem: Let (G, *) be a group. Then the identity element of G is unique.

Proof: Suppose e and e are both identity elements of (G, *), Then,
e * e = e, since e is an identity element.
Also
e * e = e, since e is an identity element.
Thus e = e

1.2.2 Theorem: (Cancellation Laws). Let a, b, c be arbitrary elements of a group (G, *).
Then the following are true.
i) a * b = a * c  b = c
ii) a * c = b * c  a = b

Proof: (i) By G4;  G such that, a  a = (a  a = e

Now
a*b=a*c
 a * ( a * b) = a * ( a * c) … (* is a binary operation)
 ( a * a) * b = ( a * a) * c … (Associativity)
 e * b = e * c …. (Def. of inverse)
 b = c … ( Def. of identity)
ii) is also similarly proved.
1
1.2.3 Theorem: Let a be an arbitrary element of a group (G, *).
Then, an inverse of a is unique.

Proof Suppose a and a~ are both inverses of a. Then by G4.

a * a =a * a = e
and
a * a~ = a~ * a = e
Therefore, a * a = a * a~ . Hence, by the cancellation law, a = a~ . Thus the inverse of a is
unique.
Remark
We usually use the symbol 1 to denote the identity element of a group and the symbol a-1 for the
inverse of an element a, whenever we use the multiplicative notation for the binary operation.
When the additive notation is used for the binary operation, we denote the identity element by 0
and the inverse of a by –a. Unless specified otherwise, we shall henceforth use the
multiplicative notation.

The next two theorems establish some properties of inverses.

1.2.4 Theorem If G s a group, then (a-1)-1 = a for all a  G.

Proof (a-1)-1 is the inverse of a-1. But since aa-1 = a-1a = 1, then a is also the inverse of a-1. By
the uniqueness of the inverse of a-1, we get (a-1)-1 = a.

1.2.5 Theorem If G is a group, then (ab)-1 = b-1a-1 for all a,b G.

Proof By definition (ab)-1 is the inverse of ab. But
(ab) (b-1 a-1) = [(ab) b-1] a-1 …. (associativity)
= [a (b b-1)]a-1 …. (associativity)
= (a 1) a-1 …. (def, of b-1)
= a a-1 …. (def. of 1)
= 1 ….. (def. of a-1)
-1 -1
Similarly, (b a ) (ab) = 1
Hence by uniqueness of the inverse of ab we get:
(ab)-1 = b-1 a-1

One method of classifying groups is by consideration of the number of elements in the group.
Hence we make the following definition.

1.2.6 Definition The number (possibly infinite) of elements in a group G is called the order
of G and is denoted by o(G). We say G is a finite group (an infinite group) if its order is finite
(infinite).

1.2.7 Definition Let a be an element of a group G with identity element 1. Define:

a0 = 1
a=a
a2 = aa
and whenever ak is defined for a positive integer k, definite
ak+1 = a.ak. Then, by mathematical induction, an is defined for each non-negative integer n.
Furthermore, define
a-n = (a-1)n
for every positive integer n. Then an is defined for all integers n.

2
Next we see that this definition of exponents satisfies the usual rules of exponents.

1.2.8 Theorem Let G be a group and let a be an arbitrary element of G. Then:

1) am an = am+n for all integers m and n.

2) (am)n = amn for all integers m and n.
Proof (1) We proceed by induction on n. If n = 1, ama1 = am+1 is true by definition of am+1.
Assume that amak = am+k for some positive integer k. Then,

amak+1 = am(aka1) …. def. of ak+1

= (amak) ….. associativity
= am+ka1 … hypothesis
= am+k+1 … def. of am+k+1

This shows that am an = am+n holds for all integers m and for all positive integers n. If n = 0, the
above equality holds, since a0 = 1. Suppose n is negative. Then –n is positive. Hence,
aman = (a-1)-m * (a-1)-n … def.
= (a-1)-m-n …… above induction proof.
= (a-1)-(m+n) ….. notation
= am+n …. def.

2) Again we proceed by induction on n.

If n = 1, (am)1 = am = am.1 is true by def. of (am)1.
Assume (am)k = amk for some positive integer k.
Then, (am)k+1 = (am)k (am)1 … def. of (am)k+1
= amk am …. hypothesis and case n = 1
= amk+m ….. case (1)
= am(k+1) …. distributives in Z
The case n  0 can easily be checked.

Note: In cases in which we use additive notation for the binary operation, we make use of
multiples in place of powers; that is, na takes the place of an.

1.3 More Examples of Groups

1.3.1 The Group of 2  2 Matrices over Z
a b
Let M = The set of 2  2 matrices over Z, i.e., the set of all symbols of the form   where
c d
a, b, c, d  Z. Define an addition on M by setting:

a b  e f  a  e b  f
  +   =  
c d g h c  g d  h 
 0 0
The identity is   and the inverse of
 0 0
a b  a  b
  is   .
c d c  d
Claim: M is an abelian group with addition so defined.

3
1.3.2 The Group of Integers Modulo n
Let Z = The set of integers, and let n be a fixed positive integer. We define a relation  on Z as
follows: a, bZ, a  b (mod n) if
a – b is a multiple of n.

It is an easy exercise to show that  is an equivalence relation on Z.

Zn is a group with the addition defined above.

Notation: We shall denote the element of the group Zn by 0, 1, 2, ..., n - 1.

1.3.3 The group of Permutations

Let A be any non-empty set. Let S(A) be the set of all one-to-one mappings of A onto itself.
S(A) is called the set of all permutations of A.

We want to show that with composition of mappings as a binary operation, S(A) is a group. We
show this step by step

1. To show that  o  S(A) for all , B element of S(A), we need only show that  o  is
one-to-one and onto.
(i) One-to-oneness of  o 
 o (a) =  o (b)  ((a)) = ((b))
 (a) = (b)  a = b, showing
that  o  is one-to-one.
(ii) Ontoness of  o 
Suppose b  A. Since  is onto x  A such that (x) = b. Since  is onto a  A
such that, (a) = x. Therefore,  o (a) = ((a)) = (x) = b. Hence,  o  is onto.
2. We next show that  o ( o ) = ( o ) o . ( o ( o ))
(a) = (((a))), while (()o )(a) = ( o ) ((a)) = (((a))) Therefore, ( o ( o ))(a) =
(( o ) o )(a) for all a  A. Hence,  o ( o )) = ( o ) o .

3. The identity mapping 1A: A  A is the identity element,

i.e.,  o 1A = 1A o  = 

4. Since  is onto, for each a  A a  A

such that (a) = a. By one-to-oneness of , this a is unique. Hence the correspondence
associating a to a is a mapping. We denote this mapping -1.
Notice that  o -1 (a) = (-1(a)) = (a) = a = 1A(a) and -1o (a) = -1((a)) = -1(a) =
a = 1A(a).
Therefore,  o -1 = -1 o  = 1A.
Hence, from (1), (2), (3) and (4) wee see that S(A) is a group.

1.3.4 The Symmetric Group

Let A = {a1, a2, …, an}. A permutation of a1, a2, … an is a one-to-one mapping from A onto A.
As shown in 4.3.3, the set of all permutations of a1, a2, …, an is a group where the operation is
composition of mappings. This group is denoted by Sn and is called the symmetric group of
order n. Even though Sn is a special case of S(A), it is sufficiently important in the study of
groups to merit special consideration. We illustrate the principles involved with S4.

4
(a) Consider the mapping  defined on the set {a1, a2, a3, a4}, by:
a1  a2
: a2  a3
a3  a1
a4  a4

 is a permutation on a1, a2, a3, a4. It is advisable to develop a shorthand notation for the
mapping . The notation

 1 2 3 4
 =  
 2 3 1 4

will be used to represent the permutation . Let the permutation  be represented by:
1 2 3 4
 =  
 3 4 1 2
Since .  are mappings, we can compose them as usual to obtain
 o  and  o :
 1 2 3 4   1 2 3 4  1 2 3 4 
 o  =   o   =  
 2 3 1 4   3 4 1 2  1 4 2 3 

1 2 3 4  1 2 3 4  1 2 3 4
 o  =   o   =  
 3 4 1 2  2 3 1 4  4 1 3 2
Observe that  o    o .

b) Returning to Sn, we can easily see that in general a permutation  on the n symbols a1, a2
…, an can be represented by:
1 2  n
 =  
 i1 i2  in 
where i1, i2, …, in is a simple rearrangement of 1, 2, …, n.

c) What is the order of Sn?

The order of Sn is the number of permutations of 1, 2, …, n and it is n(n-1)(n-2) … 3.2.1 =
n!

For instance, S3 has 3! = 6 elements. S3 = {, , , , , } where

1 2 3   1 2 3  1 2 3
 =   ;  =   ;  =  
1 2 3   2 1 3  3 2 1

1 2 3   1 2 3 1 2 3
 =   ;  =   ;  =  
1 3 2   2 3 1 3 1 2

1.4 Subgroups
1.4.1 Definition A non-empty subset H of a group G is a subgroup if H is a group with the
binary operation of G.

5
Remarks: (1) Every group G is a subgroup of itself. The subset consisting of just the identity
element 1, i.e., {1} is also a subgroup of G. Hence, every group G consisting of more than one
element has at least two subgroups: G and {1}. They are called trivial subgroups of G.
(2) If G is a group, then all subgroups of G other than G itself and {1} are called proper
subgroups of G.

Notation: If H is a subgroup of G, we write H  G. If H  G and H  G, we write H < G. In

the latter case, H is a proper subgroup of G.

Examples
(1) (Q+, .) is a proper subgroup of (+, .).
(2) Consider the group Z4. The following is the addition table for Z4.

+ 0 1 2 3
0 0 1 2 3
1 1 2 3 0
2 2 3 0 1
3 3 0 1 2

Observe that the only proper subgroup of Z4 is {0, 2}

(3) What are the subgroups of (Z, +)? We have the trivial subgroup {0}. For every positive
integer m, (mZ, +) is a subgroup of (Z, +). In fact these are the only subgroups of Z, as
show in the following theorem.

1.4.2 Theorem: (mZ, +), where m runs through the set of positive integers, is the only non-
zero proper subgroup of (Z, +).

Proof: Let H be a subgroup of Z and H  (0).

Then H contains a non-zero integer and, being a subgroup, must contain a positive integer (since
if n  H, then –n  H also). Let m be the smallest positive integer in H. Then m Z  H, since a
typical element of mZ is of the form mn where n  Z. We need only show that H  mZ. Let h
 H. By Division Algorithm  q, r  Z such that h = mq + r, where 0  r < m. Since h, n  H
and r = h – mq, it it follows that r  H. Since m is the smallest positive integer in H, it follows
that r = 0. Thus h = mq  mZ. Hence,
H = mZ.

1.4.3 Theorem. Let G be a group. a non-empty subset H of G is a subgroup of G if and only

if:
(1) a, b  H  ab  H
(2) a  H  a-1  H

Proof: : If H is a subgroup, (1) and (2) clearly hold.

: Suppose (1) and (2) hold. Then we need only check G2 and G3. Let a, b, c,  H.
Then a, b, c,  G.
Since G is a group, a(bc) = (ab)c. Hence G2 holds for H. Let a  H. Then by (2), a-
1
H. By (1) we obtain
1 = aa-1H. This proves G3. Therefore H is a group and hence a subgroup of G.

6
1.4.4 Corollary. Let G be a group. A non-empty subset H of G is a subgroup of G if and only
if:
a, b  H  a-1b  H.
Proof: : Suppose H is a subgroup of G. Let a, b  H.
By G4, a-1  H. By G1, a-1b H.
: Suppose a-1b  H  a, b  H. Let a  H.
Then 1 = a-1aH. Hence, a-1 = a-11  H, showing that (2) of Theorem 4.4.3
holds. Let a, b  H.

Then by (2), a-1, b  H. Hence, ab = (a-1)-1b  H. Thus (1) of Theorem 4.4.3

also holds. Therefore, H is a subgroup of G.

1.4.5 Theorem Let G be a group and let a  G. Then

H = {an|n Z} is the subgroup of G and is the smallest subgroup of G which contains a.

Proof: Let x, y be arbitrary element of H. Then x = am, y = ak for some m, k  Z. Then xy =

amak = am+k  H.
Notice that x-1 = a-m. Hence x-1  H.
Thus by Theorem 4.4.3, H is a subgroup of G. H is the smallest subgroup of G
containing a because any subgroup of G containing a must contain H.

1.4.6 Definition A group H is said to be cyclic if there exists an element a  H such that
every element of H can be written as an for some integer n. H is called the cyclic group
generated by a, and is denote by <a>. a is called a generator of H. If H is a subgroup of
G, then H is called a cyclic sub-group of G.

Examples
1) The group (Z, +) is a cyclic group. Both 1 and -1 are generators for the group.
2) (Z4, +) is also a cyclic group and both 1 and 3 are generators, i.e., <1> = <3> = Z4.
3) Consider the group (Z, +). For any n  Z, <n> = nZ.
1.4.7 Theorem. Let G be a finite cyclic group generated by a. If o(G) = n, then:
i) am  1 for any positive integer m < n.
ii) {a, a2, …, an-1, an = 1} is precisely the set of elements belonging to G.

Proof: (i) Suppose m < n and am = 1. Let x  G. Then x = ak for some integer k. By Division
Algorithm  unique integers q, r such that k = qm + r and 0  r < m. Then
x = ak = aqm+r = (am)q.ar = 1q.ar = 1.ar = ar

Hence, every element of G is of the form ar, where 0  r < m. Hence G has at most m elements,
where m < n. Contradiction!

ii) We know that the elements a, a2, …, an-1, an all belong to G. Next we show that these
elements are distinct. Suppose ai = aj for some positive integers i < j  n. Then multiplying
by a-1, we get:

1 = ao = ai-i = aia-i = aj. a-i = aj-i

and, since i < j  n, we must have 0 < j – i < n.
But by (i) above, am  1 for m < n. Contradiction! Thus the elements a, a2, …, an-1, an are
all distinct.

7
 Since o(G) = n, then G must be precisely the set
{a, a2, …, an-1, an}.

However, since G is a group, 1  G. But am  1 if m < n.

Hence, an = 1.

1.4.8 Theorem. Every subgroup of a cyclic group is cyclic.

Proof: Let G = <a>. Let H be a subgroup of G.

If H = {1}, then H is certainly cyclic with generator 1.
Hence, assume H  {1}. Since H is a subgroup of G, every element of H is of the form a k
for some integer k.
Let m = the least positive integer such that am  H. Let ak be an arbitrary element of H. By
Division Algorithm,  integers q, r such that k = mq + r and 0  r < m. Then r = k – mq,
and hence
ar = ak+(-mq) = aka-mq = ak(am)-q.

Since amH, then (am)-q must also be in H. Moreover, since

ak  H, then ak(am)-q must be in H. Hence, ar  H. Since m is the least positive integer
such that amH, we must have r = 0. Hence, k = mq. Hence, H is a cyclic group with
generator am.

1.4.9 Corollary Suppose a finite cyclic group G = <a> has order n. Then the subgroups of G
are precisely the subgroups generated by am where m divides n.

Proof: Left as exercise.

1.5 Cosets and Lagrange’s Theorem

In this section we will study more closely the relationship between a group and its subgroup.

1.5.1 Definition Let G be a group and let H be a subgroup of G. If a G, then Ha = {ha| h 
H} is called the right coset of H in G determined by a. The element a is often called a
representative of the right coset Ha.

Similarly, aH = {ah|h H} is called the left coset of H in G. Of course, if G is abelian Ha = aH

for all a  G.

Remarks:
1) We shall consider only right coset here; however with minor alternations, the discussion
here could be repeated for left cosets.
2) Note that H1 = H so that H itself is a right coset of H in G.

We will see later that the set of all right cosets of a subgroup H in a group G decomposes G into
disjoint subsets. Before proceeding with the proof of this, let us consider an example.

Example.
Consider Z4 = {0, 1, 2, 3} and its sub-group. H = {0, 2}. We will compute all the right cosets
of H in Z4.
H0 = H2 = {0, 2}
H1 = H3 = {1, 3}

8
There are two distinct right cosets of H in Z4. Observe that, in fact, these distinct right cosets of
H form a partition of Z4. We will soon see that this is the case for any group G and its sub-
group H.

1.5.2 Definition Let G be a group and H a subgroup of G. For a, b,  G we say a is congruent

to b modulo H, written as a  b(mod H), if ab-1  H.

Note: “Congruence modulo H” is a relation on G.

1.5.3 Theorem The relation a  b(mod H) is an equivalence relation on G.

Proof: 1) Since H is a subgroup, aa-1 = eH. Hence, a a(mod H).

2) Suppose a  b(mod H),i.e.,ab-1H. Then ba-1=(ab-1)-1 H. Hence, b  a(mod H).
3) Suppose a  b(mod H) and b  c(mod H), i.e., ab-1,
bc-1H. Now, ac-1 = (ab-1) (bc-1) H. Hence,
a  c(mod H).

Therefore, the reflexive, symmetric and transitive properties hold and a  b(mod H) is an
equivalence relation on G.

Remark: If G = Z, and H = nZ, then the relation a  b(mod H), i.e., ab-1. H, under the
additive notation, reads as a – b is a multiple of n. This is the usual number theoretical
congruence modulo n.

1.5.4 Lemma: Let [a] = the equivalence class of a under “Congruence modulo H”. Then [a]
– Ha. That is, [a] is precisely the right coset of H in G determined by a.

Proof: [a] = {x  G|x  a(mod H)}

= {x |xa-1 = h for some h  H}
= {x  G|x = ha for some h  H}
= Ha
Since the set of right cosets of H in G are simply the equivalence classes of an equivalence
relation on G, we immediately obtain the following theorem.

1.5.5 Theorem Let G be a group and let H be a subgroup of G. Then the set of all right cosets
of H in G form a partition of G. Hence every element of G belongs to one and only one right
coset of H in G, i.e., any two right cosets of H in G either are identical or have no elements in
common.
1.5.6 Lemma There is a 1.1 correspondence between any two right cosets of H in G.Proof:
Let Ha and Hb be any two arbitrary right cosets of H in G. Define a mapping : Ha  Hb by
(ha) = hb for all h  H. Then  is onto since for each hb  Hb, ha  Ha such that (ha) =
hb. Supposes (ha) = (ha), i.e., hb = hb. Then by cancellation law,
h = h, and so ha – ha. Therefore,  is 1-1. Hence,  is a 1-1 correspondence between Ha and
Hb.

Remark: Any two right cosets of H in G have the same number of elements. Moreover, since
H is a right coset, each right coset of H in G contains the same number of elements as H,
namely, o(H).

9
1.5.7 Theorem (Lagrange) If G is a finite group and H subgroup of G, then o(H)|o(G).

Proof: Let n = 0(G), m = 0(H). Let k = The number of distinct right cosets of H in G. By
4.5.5. and 4.5.6, any two distinct right cosets of H in G have no elements in common and each
has m elements. Hence, n = mk.

1.5.8 Definition Let G be a group and H a subgroup of G. The number of distinct right cosets
of H in G is called the index of H in G. The symbol iG(H) denotes the index of H in G.

Note: By Lagrange’s theorem, in case G is a finite group.

o(G )
iG(H) = .
o( H )

1.5.9 Definition If G is a group and a G, the order (or period) of a is the least positive integer
m such that am = 1. If no such integer exists, we say that a is of infinite order. We use the
notation o(a) for the order of a. Note that o(a) = o(H) where H = <a>.

Lagrange’s theorem has several corollaries. We list a few.

1.5.10 Corollary If G is a finite group of order n, then an = 1 for all a  G.

Proof: Let a  G and H = <a>. Let m = o(H). By Lagrange’s theorem, n = mk for some
positive integer k. We know am = 1. Hence, an = amk = (am)k = 1k = 1.

1.5.11 Corollary If G is a finite group and a  G, o(a)|o(G).

Proof: We know o(a) = o(H) where H = <a>. We also know by Lagrange’s Theorem
o(H)|o(g).

1.5.12 Corollary If G is a finite group of order p where p is a prime number, then G is cyclic
and every element of G except the identity is a generator of G.

Proof: Let 1  a G. By 4.5.11, o(a)|p. Since p is prime and o(a)>1, o(a) = p. Hence the
cyclic subgroup generated by any element of G other than the identity 1 has order p and
therefore must be all of G.

1.6 Normal Subgroups and Quotient Groups

1.6.1 Definition A subgroup H of a group G is said to be a normal subgroup of g is xhx-1H
x  G and h  H.

1.6.2 Definition xHx-1 = {xhx-1: h H} for x  G.

1.6.3 Corollary A subgroup H of a group g is normal
 xHx-1 H, x  G.

Notation. To say H is a normal subgroup of G is expressed symbolically as H  G. If H  G

and H  G, we write H  G.

Example1. {1} and G are normal subgroups of every group G, where 1 is the identity element
of G. These are called trivial normal subgroups of group G.
Example 2 Let M2 be the set of all non-singular 2  2 matrices with real entries.

10
 Then M2 forms a group under matrix multiplication (please check).
Further, let N be the subset of M2 of all matrices of unit determinant. Then N is a
subgroup of M2 (please verify). To prove that N  M2, let A  N and let X  M2
and let |A| stand for determinant of A. Then:
|x-1AX| = |X|-1|A| |X|
= |X|-1 |X| |A|
= |A|
=1
 x-1 AX N
Hence by definition, N  M2.

Properties of Normal Subgroups

1.6.4 Theorem Let H be a subgroup of G. Then

H  G  xHx-1 = H x  G.

Proof: xHx-1 = H x  G
 xHx-1  H
 H  G (Corollary 4.6.3)
Conversely, let H  G
 xHx-1  H (Corollary 4.6.3) (i)
 x Hx = x H(x )  H (using (i))
-1 -1 -1 -1

 H = x(x-1Hx)x-1  xHx-1 (because x-1Hx  H)

 H = xHx-1 x  G

1.6.5 Theorem: Let H  G. Then H  G  Hx = xH x  G.

Proof: H  G  xHx-1 = H x  G (by Theorem 4.6.4)

 (xHx-1)x = Hx
 xH = Hx, x  G
Conversely, let xH = Hx x  G.
Thus for x  G and h  H, we have
xhx-1 = (xh)x-1
= (h1x)x-1, (xh = h1x for some h1  H as (xH = Hx)
= h1(xx-1)
= h1  H
 xhx-1 H x  G and h  H
 H  G (by definition (4.6.1) of normal subgroup)

1.6.6 Lemma: If H is a subgroup of group G, then HH = H.

Proof: By definition, HH = {h1h2|h1H, h2H}

Now H < G.
 h1h2  H h1, h2  H
Therefore, h1h2  HH  h1h2  H
Hence HH  H
Again, H = H1, 1 is the identity element of H
 HH
Therefore, H  HH
Hence HH = H
11
1.6.7 Theorem: Let H  G. Then,
H  G  product of two right cosets of H in G is again a right coset of H in G.

Proof: Given: H  G. Let a, b  G.

Then, HaHb = H(aH)b
= H(Ha)b (because H G  Ha = aH a  G
= (HH)ab (Theorem 4.6.5)
= Hab (Lemma 4.6.6) which is a right coset of H in G.
Conversely, let Ha and Hb be two right cosets of H in G. Also let HaHb = Hc where Hc is
another right coset of H in G. Since no two right cosets of H in G are the same and no term is
repeated in any coset, then if HaHb = Hc, then c = ab.
Therefore, HaHb = Hab
 (HaHb)b-1 = (Hab)b-1
 HaH1 = Ha1
 HaH1 = Ha1
 HaH = Ha  (HaH)a-1 = (Ha)a-1
 HaHa-1 = H = HH (Lemma 4.6.6)
 H(aHa-1) = HH
 aHa = H
 H  G (Theorem 4.6.4)

1.6.8 Theorem: Let G be a group. Any subgroup of index 2 is normal in G.

Proof: Let H < G of index 2. Then iG(H) = 2. In other words, there are 2 right cosets of H in
G. Let x  G but x  H. Then H  Hx =  and G = H  Hx (right coset
decomposition). It follows that Hx = G\H.
Again there are two left cosets of H in G and H  xH =  whenever x  H, x  G.
Therefore, G = H  xH and xH = G\H
Therefore, xH = Hx = G\H.
Hence H  G.

Example: Consider symmetric group S3 on {1, 2 3}. The elements of S3 are:

1 2 3  1 2 3 
P0 =   = 1 u1 =   = (2 3)
1 2 3  1 3 2 

 1 2 3  1 2 3
P1 =   = (1 2 3) u2 =   = (1 3)
 2 3 1   3 2 1 

1 2 3  1 2 3
P2 =   = (1 3 2) u3 =   = (1 2)
3 1 2  2 1 3

Considers H = {P0, P1, P2}

lease verify from multiplication table that H  G.
0(S3) = 6, 0(H) = 3
Therefore, iG(H) = 2
Hence H  S3 (using Theorem 4.6.8)

We can also prove that H G without using Theorem 4.6.8.

12
Right cosets: HP0 = HP1 = HP2 = H = {P0, P1, P2}
Hu1 = Hu2 = Hu3 = {u1, u2, u3}

Left cosets: P0H = P1H = H = {P0, P1, P2}

u1H = u2H = u3H = {u1, u2, u3}
Therefore, Hx = xH, x  S3.
Hence H  G

Next we consider H = {P0, u1}

It can easily be checked from multiplication table that H < G. Then,

Right cosets Left cosets

H = {P0, u1,} H = {P0, u1}
Hu2 = {u2, P2} = HP2 u2H = {u2, P1} = HP1
Hu3 = {u3, P1} = HP1 u3H = {u3, P2} = HP2

Therefore, Hu2  u2H and Hu3  u3H.

Hence H is not a normal subgroup of S3.

Similarly, one can check that subgroups {P0, u2} and {P0, u3} of S3 are also not normal
subgroups of G. In fact, it can be easily verified that S3 has no normal subgroup of index 3.

1.6.9 Definition Let g be a group and let N  G. Then we define: G/N = {Nx: x  G}. In
other words, G/N is the collection of all right cosets of N in G.

1.6.10 Theorem: Let G be a group and let N  G. Then we define: G/N = {Nx: x  G}. In
other words, G/N is the collection of all right cosets of N in G.

Let  = n1a and b1 = n2b for some n1, n2  N

 a1 = n1a and b1 = n2b for some n1, n2  N
 n = n11  and b = n21 b1
 ab = n11a1 n21b1 
= n11 a n n 21 b1
 
= n11 n3 a1 b1 = n11n3 a1b1  Na1b1
 ab  Na1b1
but ab  Nab
Therefore, Nab = Na1b1
hence 0 is well-defined.
i) Closure: Na o Nb = Nab  G/N (because Nab is also a right coset). Hence G/N is closed
under. o
ii) Associativity: Na o (Nb o Nc) = Na o (Nbc)
= Na (bc)
= N (ab)c
= Nab o Nc
= (Na o Nb) o Nc Na, Nb, Nc  G/N.
Hence G/N is associative.

iii) Existence of Identity:

Let 1 be the identity of G. Then N1 = N.

13
 N1 o Na = N1a = Na and Na o N1 = Na1 = Na Na  G/N.
Hence N1, i.e. N is the identity element of G/N.

iv) Existence of Inverses:

Let Na  G/N. Then Na o Na-1 = Naa-1 = N1 = N and Na-1 o Na = Na-1a = N1 = N
Therefore, (Na)-1 = Na-1

Hence < G/N, o > forms a group.

1.6.11 Definition: The group < G/N, o > is called the quotient group or the factor group of
G by the normal subgroup N. A factor group G/N is called proper if N  G and N 
{1}.
o(G )
1.6.12 Corollary: If G is a finite group and N  G, then o(G/N) =
o( N )
Proof: Follows from corollary to Lagrange’s theorem.

1.6.13 Theorem: Every subgroup of an abelian group is normal.

Proof: Exercise

Example 1: Let G = Z. Then <Z, +> is a group.

Let N = 3Z = {…, -9, -6, -3, 0, 3, 6, 9, …}
N is a cyclic subgroup generated by 3.
Therefore, G is abelian group
Hence, N  G
Since G is abelian, right cosets of N in G will be written as N + a in place of Na.

Example 2: Let H be a subgroup of order 2 in the group Z8. Z8 is a cyclic group generated by
1, i.e., Z8 = <1> Let G = Z8. Then the only subgroup of G of order 2 is
H = {0, 4} (please verify).
Since G is abelian, H  G.

The cosets of H are:

H0 = 0 + H = {0, 4}
H1 = 1 + H = {1, 5}
H2 = 2 + H = {2, 6}
H3 = 3 + H = {3, 7}
Therefore, G/H = {H0, H1, H2, H3}

The operation table of G/H is the following:

 H0 H1 H2 H3
H0 H0 H1 H2 H3
H1 H1 H2 H3 H0
H2 H2 H3 H0 H1
H3 H3 H0 H1 H2

For example, H1  H = (1 + H)  (1+ H) = 2 + H = H2

H1  H3 = (1 + H)  (3 + H) = 4 + H = {0,4) = H0
It follows from the operation table that G/H forms a group.

14
In other words, Z8/H, where H = {0, 4}, is the quotient group of Z8.

Example: What is the group G/N when G = S3 and

N = {P0, P1,P2} in the usual notation?
We have seen that N  G and iG(N) = 2
Therefore, G/N = {N, Nx} where Nx = {u1, u2, u3} and x is any one of the elements u1, u2, and
u3. The Multiplication table of G/N is given below:

. N0 N1 where N0 = N
N0 N0 N1 N1 = Nx
N1 N1 N0

For example, N1 . N1 = Nx . Nx = Nx2 = N = N0 (since x2 = P0 xG)

1.7 Homomorphisms
1.7.1 Definition: Let G, G be groups. A mapping : G  G is called a homomorphism if
(ab) = (a) (b) a, b,  G.

Example:
1. Let G = any group with identity element 1. :G  G, defined by (a) = 1 a  G, is a
homomorphism.
2. Let G = any group, : G  G, defined by (a) = a a  G, is a homomorphism
3. Let + = The group of positive real numbers under multiplication .
Let  = The group of real numbers under addition.
 = +  , defined by (a) = log a, is a homomorphism.

1.7.2 Definition: Let : G  G be a homomorphism of groups and let 1, 1 are identity

element of G and G respectively. Then

a) ker () = {a  G |  (a) = 1 }, is called the kernel of 

b) Im() = { a  G | (a) = a for some a  G} is called the image of .

1.7.3 Lemma Let : G  G be homomorphism. Let 1, be identity element of G and G

respectively. Then
a) (1) = 1
b) (a)-1 (a)-1
c) ker () is a normal subgroup of G.

Proof: (a) a  G, (a) = (a.1) = (a). (1).

On the other hand, since 1 is the identity in G , by cancellation law, (1) = 1
(b) Let a  G. Since G is a group, 1 = a-1a = aa-1.
Now apply :
1 = (1) = (a-1)(a) = (a)(a-1).

Hence, (a-1) is the inverse of (a) in G , and therefore (a-1) = (a)-1.

(c) Let a,  ker(). Then (a) = 1 = (b). Therefore (ab) = (a)(b) = 1 . 1 = 1 ,

showing ab  ker(). Also, (a-1) = (a)-1 = 1 -1 = 1 showing a-1  ker(). Hence
15
 ker() is a subgroup of G. Now let a  G and h  ker(). Then, (aha-1) =
(a)(h)(a-1) = (a). 1 .(a-1 = 1 . Therefore, aha-1  ker(). Hence, ker() is a
normal subgroup of G.

1.7.4 Definition: Let :G  G be a homomorphism.

a) If  is onto,  is called an epimorphism.
b) If  is one-to-one,  is called a monomorphism.
c) If  is both an epimorphism and a monomorphism, then  is called an isomorphism. If
 is an isomorphism, then G and G are said to be isomorphic, and we write G  G .

1.7.5 Theorem: Let : G  G be a homomorphism. Then

1)  is an epimorphism if and only if Im() = G .
2)  is a monomorphism if and only if ker() = {1}, where 1 is the identity element in G.

Proof: (1) is obvious

(2) Suppose  is a monomorphism. Let a  ker . Then (a) = 1 , where 1 is the
identity element in G . But (1) = 1 . Hence, (a) = (1)  a = 1. Therefore
ker  = {1}
Suppose ker  = {1}. Let a, b  G such that (a) = (b).
Then 1 = (a)(b)-1 = (ab-1) = (ab-1)  ab-1  ker   ab-1 = 1  a = b. hence,
 is a monomorphism.
1.7.6 Theorem: Let : G  G be an ephimorphism. Then G  G/ker().

Proof: Define a map : G.ker()  G by (Ka) = (a), where K = ker().

Claim:  is well-defined, i.e., b  Ka then (a) = (b). But b  Ka  a  b(mod K)  ab-1 

K  ab-1 = k for some k  K. Therefore, a = kb for some k  K. Hence, (a) =
(k)(b) = 1 .x/(b) = (b). Thus  is well-defined.

Now (KaKb) = (Kab) = (ab) = (a)(b) = (Ka)(Kb). Hence,  is a homomorphism.

Suppose (Ka) = (Kb)  (a) = (b)  (ab-1) = (a)(b)-1 = 1 ab-1  ker()  a 

b(mod K)  Ka = Kb. Hence  is
one-o-one. Clearly  is onto, since  is onto.

Hence,  is an isomorphism.

1.7.7 Theorem: Suppose G is a group and H a normal subgroup of G. Then : G  G/H

defined by (a) = Ha  a  G is an epimorphism and ker() = H.

Proof: (ab) = Hab = HaHb =(a)(b) showing that  is a homomorphism. Clearly  is onto.
Hence  is an epimorphism.

Suppose a  ker(). Then Ha = (a) = H  a  H. Hence, ker()  H. Suppose h  H. Then

(h) = Hh = H  h  ker(). Hence, H  ker(). Therefore, ker() = H.

1.7.8 Definition: The epimorphism : G G/H defined by

(a) = Ha is called the canonical epimorphism.

16
1.8 Automorphisms
1.8.1 Definition: Let G be any group. An isomorphism space : G  G is called an
Automorphism of G.

In other words, a map : G  G is to be Automorphisms iff:

i)  is homomorphism;
ii)  is one-to-one; and
iii)  is onto.
Example: 1) The identity mapping 1G of G onto itself is clearly an Automorphisms of G.
2) Let G be any abelian group. Define a mapping
: G G by (x) = x-1 x  G.  is an Automorphisms of G.
Obviously  is onto and  is one-to-one. Moreover, (x1x2) = (x1x2)-1 = x 21 x11
= x11 x 21 = (x1)(x2) x1,x2  G.
Therefore,  is homomorphism. Hence  is Automorphism of G.

Furthermore, if xo  G such that x0  x01 , then (x0) = x01  x0. So   1.

1.8.2 Theorem: The set of all Automorphisms of group of G forms a group.

Proof: Let A(G) be the set of all a automorphism of G. We know that S(G) is the group of all
permutations of G. Then we shall prove that A(G) < S(G). Obviously A(G)  S(G). Clearly
the identity map 1G is automorphism of G.

Therefore, 1G  A(G). Hence A(G)  .

For elements of A(G), we can use composition of mappings as defined for group S(G).

i) Let 1, 2  A(g)  1, 2  S(G)

 21  S(G)
 21 is one-one and onto
Thus, x1, x2  G, we have:
(21)(x1x2) = 2(1(x1x2)) = 2(1x11x2)
= 21(x1)21(x2)
Therefore, 21 is a homomorphism of G and hence 21 is automorphism of G i.e.
21  A(g).

ii) Let   A(G)    S(G)  -1  S(G)  -1 is 1 – 1 and onto

Also   1 x1 1 x 2  = (-1x1) (-1x2)
= -1(x1)-1(x2)
= 1G(x1)1G (x2) = x1x2

Thus -1(x1)-1(x2) = -1(x1x2)

Therefore, -1 is automorphism and hence -1  A(G) < S(G) Thus A(G) is a group in its own
right. We call A(G) the full group of automorphisms of G.

1.8.3 Theorem: For g  G, define a map ig: G  G by ig(x) = g xg-1 x  G. Then ig is an

automorphism of G.

17
Proof:
i) Given y  G. Let x = g-1 y g
Then ig(x) = ig(g-1 y g) = g(g-1 y g)g-1 = y
Therefore, ig is onto.
ii) Let ig(x1) = i(x2)  gx1g-1  gx2g-1  x1 = x2
Therefore, ig is one-one
Hence ig is an automorphism of G.
iii) ig(x1x2) = g(x1x2)g-1 = g(x1g-1gx2)g-1
= (g x1g-1) (g x2g-1) = ig(x1)ig(x2)
Hence ig is homomorphism.

1.8.4 Definition: For each g  G, the automorphism

ig: G  G defined as ig(x) = g  g-1 x  G is called an inner automorphism of G. Any other
automorphism of G is called outer-automorphism of G.

Example: Let G be non-abelian group,

  a pair a, b  g such that ab  ba
Now ia(b) = aba-1  b
Therefore, ia  1G
Thus for non-abelian group G, there always exist non-trivial inner automorphisms of G.

Let J(G) = {ig  A(G) | g  G}

= the set of all inner automorphisms of the elements of group G.

1.8.5 Theorem: J(G) < A(G). In other words, J(G) is a group in its own right.

Proof: Clearly J(G)  A(G) and J(G)  

i) Let ig, ih  J(G) where g, h  G.
Then x  G, (ihig)(x) = ih(ig(x)) = ih(g  g-1) = h(g  g-1)h-1
= (hg)x(hg)-1 = ihg(x)
Hence ihig = ihg  J(G)
ii) Let ig  J(G)  ig  A(G)
Then x  G, ig(g-1xg) = g(g-1xg)g-1 = x
Therefore, igig-1(x) = x (because g-1 g = g-1x(g-1)-1
= i g 1 ( x )
Therefore, i g i g 1 = i1, where 1 = The identity element in G.
Similarly, i g 1 i g = i1.
Therefore (ig)-1 = i g 1  J(G)
Hence J(G) < A(G).
J(G) is called the group of inner automorphisms of G.

Remark: Let G be abelian group. Then obviously the only inner automorphism is the
identity, since then gxg-1 = x,x  G. All non-trivial automorphisms of G are outer
automorphisms.

1.8.6 Definition: The set of all elements commutative with all elements of a group G is called
its center and is denoted by C(G). In other words,
C(G) = {x  G: |gx = xg, g  G}.

18
1.8.7 Lemma: For any group G, the center C(G) is an abelian and hence a normal subgroup
of G.

Proof: (i) Let c1, c2  C(G) = gc1 = c1g  gc2 = c2g g  G.

Then (c1c2)g = c1(c2g0 = c1(gc2) = (c1g)c2 = g(c1c2)
Therefore, c1c2  C(G)

(ii) Let c  C(G)  cg = gc, g  G

 gc-1 = c-1g, g  G
Therefore, c-1  C(G)
Hence C(G) < G.
Obviously C(G) is abelian.
Hence C(G) is a normal subgroup of G.

Example: By elementary calculations, it can be shown that the group S3 has trivial centre.
If G is abelian group, then C(G) = G.

1.8.8 Theorem: J(G)  G/C(G).

Proof: Consider the mapping:

: G  J(G) defined by (g) = ig where ig  J(G).
i) Obviously  is onto
ii) g, h  G, e have
(gh) = igh = igih (proved in theorem (4.8.6)).
= (g) (h)
Therefore,  is homomorphism
Therefore, : G  J(G) is homomorphism and onto. Hence, by first fundamental theorem of
homomorphism, J(G)  G/K, where K is the kernel of homomorphism , i.e. K ={g G|(g)
= i1}, where i1 is the identity map of J(G)
We shall prove that K = C(G).
Let g  K  g  G  (g) = i1
Therefore, ig = i1
 ig(x) = i1(x), x  G
 g xg-1 = 1  1-1
 xg = gx, x  G
 g  C(G)
Hence K  C(G).

Conversely, let c  C(G)  xc = cx, x  G

 x = cxc-1 = ic(x), xG
 ic = i1
 (c) = i1
 c  K
Hence C(G)  K
It follows that K = C(G)
Hence, it follows that J(G)  G/C(G)

19
1.9 Cayley’s Theorem.
1.9.1 Theorem: Every group is isomorphic to a group of permutations.

Proof: Let G be a given group.

Step I. We shall produce a suitable set G. Let us think of G just as a set.
Let S(G) be the group of all permutations of G.
For g  G, we define a map Pg: G  G given by
Pg(x) = gx, x G
i) Let y G. Then Pg(g-1y) = g(g-1y) = y
Therefore, Pg is onto.
ii) Let Pg(x1) = Py(x2), x1, x2  G
 gx1 = gx2
 x1 = x2
Therefore, Pg is 1 – 1
Therefore, Pg is a permutation of G. Hence Pg  S(G).
Let G = {Pg  S(G) | g  G}

Step II: Se next show that G < S(G).

i) Let Pg, Ph  G. The x  G, we have
(PhPg)(x) = Ph(gx) = h(gx) = (hg)x = Phg(x).
Therefore, PhPg = Phg  G

ii) P1(x) = 1x = x, where 1 is the identity of G.

Therefore, P1 is the identity of G

iii) Let Pg  G, where g  G

Then, Pg 1 Pg = Pg 1 g = P1
and Pg Pg 1 = P1
Therefore, (Pg)-1 = Pg 1  G
Hence G < S(G)

Step III: To prove that G  G.

Define a map : G  G by (g) = Pg  G g  G

i) It follows that  is onto

ii) Let (g) = (h)
 Pg = Ph
 Pg(x) = Pg(x)
 gx = hx
g=h
Therefore,  is 1 – 1

iii) For g, h  G
(gh) = Pgh = pgph = (g)(h)
Therefore,  is homomorphism

20
 It follows that  is isomorphism of G onto
G Hence G  G

1.9.2 Definition: The group G defined above is called left regular representation of G.

Further consider : G  G defined by g(x) = xg. Then each g  S(G). Let G = {g
 S(G)|g G}. Then G is group and G  G. The group G is called right regular
representation of G.

The multiplication tables for G and G shows that G and G are isomorphic where G is the
group of permutations as shown.

21