Part I: Groups and Subgroups

Satya Mandal
University of Kansas, Lawrence KS 66045 USA

January 22

1 Intorduction and Examples

This sections attempts to give some idea of the "nature of abstract algebra".
I will give a summary only. Please glance through the whole section in the
textbook. Follwing are some of the main points:

1. The section provides a prelude to "binary operations", which we define

in the next section.

2. To do this it discusses multiplication of complex numbers.

3. It gives Euler Formula that

eiθ = cos θ + i sin θ

4. Given any complex number z ∈ C we can write

z = |z| eiθ

5. It discusses the algebra of the Unit Circle.

(a) The unit circle

U = {z ∈ C : |z| = 1} = {z ∈ C : z = eiθ where θ ∈ R}

1
 (b) Note, for z, w ∈ U , the product zw ∈ U . We say the unit circle U
is closed under multiplication.
(c) Define the map

f : [0, 2π) −→ U where f (θ) = eiθ .

Then, f is a bijection.
(d) In fact, f (x + y) = f (x)f (y) sends sum to the product. Here,
addition x + y in [0, 2π) is defined "modulo 2π".

6. We discuss the algebra of Roots on Unity. Fix a positive integer n.

(a) Let Un be the set of all solutions of the equation z n = 1 (in C)

2πi
(b) write ζ = e n . Then

Un = {ζ 0 , ζ 1 , ζ 2 , . . . , ζ n−1 }

(c) Define the map

2πri
ϕ : Zn −→ Un by ϕ(r) = ζ r = e n

is a bijection. It needs a proof that ϕ is well defined.

(d) In fact, ϕ(x + y) = ϕ(x)ϕ(y) sends sum to the product.

7. Also, Un ⊆ U , the unit circle.

2
2 Binary Operation
Examples of "binary operations" are addition and multiplication, in all the
situations where we worked with them:

Z, Zn , R, C, Mn (R), Mn (C)

where Mn (R), Mn (C) denote the set of matrices of size n×n, with coefficients
in R or C. Similarly, multiplication on U, Un are binary operations. They are
called binary operations, because to each ordered pair (x, y) they associate
another element x + y or xy.
We give a formal definition of "binary operations".

Definition 2.1. Let S be a set. A binary operation ∗ on S is a mapping

∗ : S × S −→ S. For now, we use the notation x ∗ y := ∗(x, y).

Definition 2.2. Suppose ∗ is a binary operation on S and H be a subset

of S. We say that H is closed under ∗, if for any x, y ∈ H we also have
x ∗ y ∈ H. Notationally,

if x, y ∈ H =⇒ x ∗ y ∈ H.

Reading Assignment: §I.2 Examples 2.2-2.10.

Example 2.3 (§I.2, 2.7). Let F be the set of all continuous real valued
functions on R. We give four binary operations:

1. Sum (f + g)(x) = f (x) + g(x)

2. Product (f g)(x) = f (x)g(x)

3. Composition (f og)(x) = f (g(x))

4. Subtraction (f − g)(x) = f (x) − g(x)

5. Note division f /g is not always defined (unless g(x) 6= 0 ∀x). So,

division is not a binary operation on F .

3
§1. Properties of binary operations

Definition 2.4. A binary operation ∗ on S is said to be commutative,

if x∗y =y∗x ∀ x, y ∈ S.

Remark or examples. As far as I can see, matrix multiplication and com-

position are the only "natural" binary operations that are not commutative.
Most of the counter examples are artificially constructed.

1. On Z, Zn , R, C both addition and multiplication are commutative.

2. On Mn (R), Mn (C) additions are commutative. But multiplcation is

NOT commutative. For example
     
0 1 0 1 0 1 0 1
6=
1 0 0 0 0 0 1 0

More generally,
     
0 1 a b a b 0 1
6=
1 0 x y x y 1 0
   
x y b a
LHS = and RHS = .
a b y x

3. Let F be the set of all continuous functions on R. Then

(a) Addition + is commutative.

(b) Substraction is NOT commutative.
(c) The composition is NOT commutative. Let f (x) = ex and g(x) =
2
x2 . Then f og(x) = ex and gof (x) = e2x . So, f og 6= gof .

Definition 2.5. A binary operation ∗ on S is siad to be associative

if a ∗ (b ∗ c) = (a ∗ b) ∗ c ∀ a, b, c ∈ S.

Remaks and Examples.

4
 1. First, only when an operation is associative, we do not need to use
parentheses to specify order of multiplication. we can write a ∗ b ∗ c for
both a ∗ (b ∗ c), (a ∗ b) ∗ c.
2. I do not know (well I do) any natural example of binary operations,
that is not associative.

Theorem 2.6. Let F(S) be the set of all functions f : S −→ S. Then,

the compositions o is a binary operation on F(S). The composition is an
associative binary operation.

Proof. It is straight forward. Look at the text book.

Corollary 2.7. Multiplication on Mn (R), Mn (C) are associative.

Proof. Let L(Rn ) be the set of all linear functions Rn −→ Rn . So, L(Rn ) ⊆
F(Rn ). So, composition is associative in L(Rn ).
Recall, there is an 1-1 and onto correspondence between

ϕ : Mn (R) −→ L(Rn )

such that ϕ(AB) = ϕ(A)ϕ(B). Now, we will use the associative property of
the composition in L(Rn ). We have

ϕ((AB)C) = ϕ(AB)ϕ(C) = [ϕ(A)ϕ(B)]ϕ(C)

= ϕ(A)[ϕ(B)ϕ(C)] = ϕ(A)[ϕ(BC)] = ϕ(A(BC).

Since, ϕ is 1-1, we have (AB)C = A(BC). So, the matrix product is asso-
ciative. The proof is complete.

5
2.1 Tables
For a finite set S, tables can be used to describe a binary operation.
Reading Assignment; Read examples 2.14-2.25.
Let me describe the addition and multiplication on Z4 by tables:

Addition T able P roduct T able

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

Let me do the same for Z5 :

Addition T able P roduct T able

· 0 1 2 3 4 · 0 1 2 3 4
0 0 1 2 3 4 0 0 0 0 0 0
1 1 2 3 4 0 1 0 1 2 3 4
2 2 3 4 0 1 2 0 2 4 1 3
3 3 4 0 1 2 3 0 3 1 4 2
4 4 0 1 2 3 4 0 4 3 2 1

6
3 Isomorphic Binary Structures
Abstract

We define isomorphic Binary Structures. Main point is, if two

binary structures are isomorphic, then propertes of one translate over
to properties of the other, via the isomorphism. So, if we know one we
know the other. We do not have to study two of them seperately.

Definition 3.1. By a binary structure hS, ∗i , we mean a set S

with a binary operation ∗ on it.

Definition 3.2. Let hS, ∗i and hT, ∗′ i be two binary structures.

1. A map ϕ : S −→ T is called (a map of) or a homomorphism

of binary structures

if ϕ(x ∗ y) = ϕ(x) ∗′ ϕ(y) ∀ x, y ∈ S.

2. A map ϕ : S −→ T is called an isomorphism of binary struc-

tures

if ϕ(x ∗ y) = ϕ(x) ∗′ ϕ(y) ∀ x, y ∈ S.

and if ϕ is a bijection.
(Emphasis in this section is on isomorphic structures; not on
homomorphisms)

Example 3.3. Let U = {z ∈ C : |z| = 1} be the unit circle. Then,

with usual multiplication, hU, ·i is a binary structure.
On the interval [0, 2π) the addition "modulo 2π provides a binary
structure ([0, 2π), +). The map

ϕ : [0, 2π) −→ U def ined by ϕ(t) = eit

is an isomorphism of binary structures.

7
Example 3.4. Let n be a fixed positive mumber. Then,
2kπi
ψ : Zn −→ Un def ined by ψ(k) = e n (= ζ n )

is an isomorphism of binary structures.

Example 3.5. The mapping

exp : hR, +i −→ h(0, ∞), ·i def ined by exp(t) = et

is an isomorphism of binary structures. Its inverse

ln : h(0, ∞), ·i −→ hR, +i t 7→ ln t

is also an isomorphism of binary structures.

Definition 3.6. Let hS, ∗i be a binary structure. An element e ∈ S

is called an identity element for ∗

if e ∗ x = x ∗ e = x ∀ x ∈ S.

Theorem 3.7. Let hS, ∗i be a binary structure. Then, hS, ∗i has at

most one identity element.

Proof. Suppose e, ǫ be identity elements in S. We will prove that

e = ǫ.
ǫ = ǫe because e is identity.

Also
e = ǫe because ǫ is identity.

So, ǫ = e. The proof is complete.

Theorem 3.8. Suppose ϕ : S −→ T is an isomorphism of two binary

structures hS, ∗i and hT, ∗′ i. Let e ∈ S be the identity for ∗. Then
ϕ(e) is an identity in hT, ∗′ i.

8
Proof. For x ∈ T we have to prove x ∗′ ϕ(e) = ϕ(e) ∗′ x = x. Since ϕ
is onto, ϕ(a) = x for some a ∈ S. We have

e ∗ a = a ∗ e = a. Apply ϕ : ϕ(e) ∗′ ϕ(a) = ϕ(a) ∗′ ϕ(e) = ϕ(a).

Which is ϕ(e) ∗′ x = x ∗′ ϕ(e) = x. So, ϕ(e) is an identity in T . The

proof is complete.

We look at a few binary structures that are not isomorphic.

Example 3.9 (13.15). 1. hQ, +i and hZ, +i are not isomorphic.

2. (Added): hQ, ·i and hZ, ·i are not isomorphic.

Proof.

1. This is because hQ, +i is "divisible" by any positive integer n. It

is divisible by 3 means, give any y ∈ Q there is an element x ∈ Q
such that x + x + x = y, namely x = y/3. But hZ, +i does not
enjoy this property.

2. For the second statement note all nonzero elements is Q has an

inverse, while that is not true for Z.

Example 3.10 (Added). hR, ·i is not isomorphic to hM2 (R), ∗i where

∗ is usual multiplication. Among other things, the first one is commu-
tative and the second one it not commutative.

Example 3.11 (3.17). hR, ·i and hC, ·i are not isomorphic.

9
4 Groups
Definition 4.1. A Group hG, ∗i is a binary structure such that the
following axioms holds:
1. Associativty holds:

(a ∗ b) ∗ c = a ∗ (b ∗ c) ∀ a, b, c ∈ G.

2. G has an itentity element e, which means

e∗a=a∗e=a ∀a∈G

3. (Inverse)

F or each a ∈ G ∃ a′ ∈ G ∋ a ∗ a′ = a′ ∗ a = e.

This a′ is called an/the inverse of a.

Remarks.
1. To check hG, ∗i is a group, we check the (0) G is closed under ∗,
(1) ∗ is associative, (2) G has an identity, (3) each element has
an inverse.
2. Notation. We usually denote the group hG, ∗i by G, when ∗ is
understood.
3. The notation a′ is not very normal. For most of the groups,
the operation ∗ is denoted by addition + or multiplication (like
x · y of xy). If we use multiplicative notations, then a′ is usually
denoted by a−1 . If we use addive notation, then a′ is usually
denoted by −a. The additive notation + is used, only when ∗ is
commutative.

Definition 4.2. A Group hG, ∗i is said to be an abelian group, if ∗

is commutative.

Example 4.3 (4.2). The unit circle U and the roots of unity Un are
groups under multiplication.

Reading Assignment: Read Example 4.4-4.14.

10
4.1 Elementary Properties of Groups
Theorem 4.4. Let G be a group. Then left and right cancellation
holds. That means,

f or x, y, z ∈ G x∗z = y∗z =⇒ x=y (right Cancelleation)

and

f or x, y, z ∈ G z∗x = z∗y =⇒ x=y (left Cancelleation)

Proof. Let x ∗ z = y ∗ z. Multiply this equation by inverse z ′ of

z, on the right. We get (x ∗ z) ∗ z ′ = (y ∗ z) ∗ z ′ . By associativity
x ∗ (z ∗ z ′ ) = y ∗ (z ∗ z ′ ), So, x ∗ e = y ∗ e or x = y. This establishes
the right cancellation.
To prove the left cancellation, multiply the equation z ∗ x = z ∗ y
by z ′ on the left. (Exercise: complete it). The proof is complete.

Theorem 4.5. Let G be a group and a, b ∈ G. Then

1. The equation ax = b has a unique solution.

2. The equation xa = b has a unique solution.

Proof. Let a′ be an/the inverse of a. Then, x = a′ ∗ b is a solution of

the equation ax = b, becuase

a ∗ (a′ ∗ b) = (a ∗ a′ ) ∗ b = e ∗ b = b.

So, the equation a ∗ x = b has a solution x = a′ ∗ b. Now suppose the

equation a ∗ x = b has two solutions x = x1 , x2 . So, a ∗ x1 = b and
a ∗ x2 = b. So, a ∗ x1 = a ∗ x2 . By left cancellation, x1 = x2 . So, the
equation a ∗ x = b has exactly one solution. So, the statement (1) is
established. We prove statement (2) similarly (exercise).

Theorem 4.6. Let G be a group. Then

1. G has exactly one identity e.

11
 2. Given x ∈ G there is exactly one element x′ such that

x ∗ x′ = x′ ∗ x = e.

This (unique) x′ is called the inverse of x.

Proof. By definition of group, G has an identity e ∈ G such that

x ∗ e = e ∗ x = x for all x ∈ G. The uniqueness follows from the
uniqueness of identity for binary structures. (Please rewrite the proof).
So, (1) is established.
Suppose x ∈ G. By the third property of groups, there is one
element x′ ∈ G such that

x ∗ x′ = x′ ∗ x = e.

Suppose x′ ∈ G also satisfy the same property, i.e

x ∗ x” = x” ∗ x = e.

Then, clearly x ∗ x′ = x ∗ x”. So, by left cancellation x′ = x”. So, the

uniqueness of the "inverse" of x is established.

Notations: Suppose G is a group.

1. When we use the multiplicative noation, the inverse of a will be

denoted by a−1 . When we use the additive noation "+", the
inverse of a will be denoted by −a.

Corollary 4.7. Let G be a group and a, b ∈ G. Then, (a ∗ b)−1 =

b−1 ∗ a−1 . (Recall, for inverses of matrices, we have seen the same.)

Proof. We have

(a∗b)∗(b−1 ∗a−1 ) = ((a∗b)∗b−1 )∗a−1 = (a∗(b∗b−1 ))∗a−1 = (a∗e)∗a−1 = a∗a−1 = e.

Similarly, (b−1 ∗ a−1 ) ∗ (a ∗ b) = e. So, (a ∗ b) = b−1 ∗ a−1 . The proof

is complete.

12
4.2 Finite Groups
Example 4.8. 1. Any singleton set {e} can be given a group struc-
ture by defining e ∗ e = e.

2. Also, the subset {0} of Z is a group under addition.

3. Also, the subset {1} of Z is a group under multiplication.

4. All these groups are isomorphic (as in binary structures).

Example 4.9. 1. Any doubleton set {e, a} can be given a group

structure by defining multiplication by the table

∗ e a
e e a
a a e

2. Z2 is a group with two elements.

3. h{1, −1}, ·i is a group with two elements.

4. These groups are isomorphic.

5. In fact, any group G with two elements is isomprphic to Z2 (give

a proof).

Example 4.10. Suppose G is a group of order three. Then G ≈ Z3 .

Proof. Let G = {e, a, b}, where e is the identity.

1. First, ab 6= a and ab 6= b. So, ab = e.

2. Claim a2 = b. This is because, a2 6= e and a2 6= a.

3. Therefore, G = {e, a, a2 }. Now, a3 6= a, a3 6= a2 . So, a3 = e.

∼
4. So, the mapping ϕ : Z3 −→ G given by

ϕ(0) = e, ϕ(1) = a, ϕ(2) = a2

is an isomprphism.

The proof is complete.

13
Example 4.11. Suppose G is a group of order four. We will show
that either Z4 ≈ G or G is the Klein group (to be defined).
Proof. Write G = {e, a, b, c} where e is the identity. There are two
cases:

1. First, ab = e or

2. ab = c.

1. Suppose ab = e. In this case, we will prove Z4 ≈ G.

(a) Then, c is its own inverse or c2 = e.

(b) Claim a2 = c. To see this, first note a2 6= e, a2 6= a. Further,
if a2 = b then a3 = e. Then, it would follow a3 = e. That
/ {e, a, a2 , c} = G, which is impossible.
would imply ac ∈
Therfore, a2 = c. Similarly, b2 = c.
(c) So, a2 = b2 and hence a3 = b. So, G = {e, a, a2 , a3 }.
(d) Also note a4 = c2 = e.
∼
(e) So, the mapping ϕ : Z4 −→ G given by

ϕ(0) = e, ϕ(1) = a, ϕ(2) = a2 , ϕ(2) = a3

is an isomprphism.
The proof is complete.

2. Now suppose ab = c. In this case, a is its own inverse and b is its

own inverse. So, c is its own inverse. So, a2 = b2 = c2 = e. So,
the multiplication table loos like:

P roduct T able
· e a b c
e e a b c
a a e c
b b e
c c e

14
 By cancellation property, no repeatation is allowed in any row or
column. So, the multiplication table is completed as follows.

P roduct T able
· e a b c
e e a b c
a a e c b
b b c e a
c c b a e

This group is called the Klein Group.

3. So, there are only two distinct groups of order 4.

4.3 Failure of Cancellation

Example 4.12. 1. Recall, in R cancellation fails for multiplcation.
The zero is the problem: 0 ∗ x = 0 ∗ y = 0 for all x, y ∈ R.

2. For matrices, the cancellation property fails for multiplication:

We have
" #" # " # " #" #
1 1 1 1 0 0 2 2 1 1
= =
1 1 −1 −1 0 0 2 2 −1 −1

So, cancellation property fails for matrix product.

15
5 Subgroups
First, we set up some notations:

1. Normally, we use addition + or multiplication (like x · y or xy)

to denote the binary operation ∗.

2. Only when the group G is known to be abelian, we use additive

"+" notation.

3. If we use the additive notation +, then the identity is denoted

by zero 0. The inverse of a is denoted by −a.

4. If we use the multiplicative notation, then the identity is denoted

by "one" 1. The inverse of a is denoted by a−1 .

5. Suppose n ≥ 0 is a non-negative integer. In the additive notation,

na = n · a := a + a + · · · + a denotes sum of a with itself n times.
Also −na = −(na). In multiplicative notation, an := a · a · · · a
product od a with itself n times. Also a−n := (an )−1 .

Definition 5.1. For a group G, order of G is defined to be the

number of elements in G. It is denoted by |G|. Obvioulsy, a group can
have infinite order. For example |Zn | = n and |Z| = ∞.

5.1 Subgroups
Definition 5.2. Let G be a group. A subset H of G is called a
subgroup of G, if H itself is a group under the operation inherited
from G. For a subset H to be a subgroup G following should be
satisfied:

1. H is closed under the binary operation in G. That means,

a, b ∈ H =⇒ ab ∈ H.

2. The identity e of G is in H.

3. For a ∈ H =⇒ a−1 ∈ H.

16
 4. (Remark. We do not need to check associativity in H, because
it is inherited directly from G).

If H is a subgroup of G, we write H ≤ G. Further if, H 6= G then we

say H is a proper subgroup of G.

The Trivial Subgroups:

Let G be a group. Then, {e} and G are two of its trivial subgroups.

Example 5.3. Following are subgroups:

1.
hZ, +i ≤ hQ, +i ≤ hR, +i ≤ hC, +i

Each one on the left is a subgroup of any one on the right.

2.
h{1, −1}, ·i ≤ hQ∗ , ·i ≤ hR∗ , ·i ≤ hC∗ , ·i

Each one on the left is a subgroup of any one on the right.

3.
hUn , ·i ≤ hU, ·i ≤ hC∗ , ·i

Each one on the left is a subgroup of any one on the right.

4. h{0, 2}, +i is a subgroup of hZ4 , +i.

More generally, let n = kr be a positive integer, k > 0, r > 0.
Then, h{0, k, 2k, . . . , (r − 1)k}, +i is a subgroup of hZn , +i. Note

h{0, k, 2k, . . . , (r − 1)k}, +i ≈ Zr .

So, one may loosely say Zr is a subgroup of Zn .

5. Let C[0, 1] be set of all continuous functions on the interval [0, 1].
Then hC[0, 1], +i is a group. Let H be the set of all functions
f ∈ C[0, 1] which vanishes on (.25, .75). Then, H is a subgroup
of C[0, 1]. In fact, given any subset X ⊂ [0, 1], the set

Z(X) = {f ∈ C[0, 1] : f|X = 0} is a subgroup.

17
 6. Let GLn (R) be the set of all invertible matrices of order n. (We
know GLn (R) = {A ∈ Mn (R) : det A 6= 0}.) Then, GLn (R) is a
group.

(a) Let SLn (R) = {A ∈ Mn (R) : det A = 1}. Then SLn (R) is a
subgroup of GLn (R).
(b) Let On (R) be the set of all orthogonal matrices. (i. e. A ∈
GLn (R) such that AAT = In .) Then On (R) is a subgroup
of GLn (R).
(c) Let SOn (R) = {A ∈ On (R) : det A = 1}. Then SOn (R) is a
subgroup of On (R).

7. Similarly, let GLn (C) be the set of all invertible matrices of order
n. (We know GLn (C) = {A ∈ Mn (C) : det A 6= 0}.) Then,
GLn (C) is a group.

(a) Let SLn (C) = {A ∈ Mn (C) : det A = 1}. Then SLn (C) is a
subgroup of GLn (C).
(b) Let Un (C) be the set of all unitary matrices. (i. e. A ∈
GLn (C) such that AĀT = In .) Then Un (C) is a subgroup
of GLn (C).
(c) Let SUn (C) = {A ∈ Un (C) : det A = 1}. Then SUn (C) is a
subgroup of Un (C).

5.2 Cyclic Subgroups

Theorem 5.4. Let G be group and a ∈ G. Then H = {an : n ∈ Z}
is a subgroup of G. In fact, H is the smallest subgroup of G that
contains a.

Proof. First, recall for a negative integer k < 0 we define ak :=

(a−k )−1 . Now H is closed under product: for m, n ∈ Z we have
am · an = am+n ∈ H. The identity e = e0 ∈ H. For an ∈ H, we have
(an )−1 = a−n ∈ H. So, H is a subgroup.

18
 Now, suppose K is another subgroup of G that contains a. Since
K is closed under multiplication an ∈ K for all non-negative integers
n. Again, for negative integers m we have am = (a−m )−1 ∈ K. So,
an ∈ K, ∀ n ∈ Z. So, H ⊆ K. This establishes that H is the smallest
subgroup of G that contains a. The proof is complete.

Definition 5.5. Let G be a group and a ∈ G.

1. Then, H = {an : n ∈ Z} is called the cyclic subgroup of G

generated by a. This H is denoted by hai.

2. If G = hai for some a ∈ G, then we say that G is a cyclic group.

3. Remark. So, a cyclic group is a group that is generated by one

element. In future, we will consider groups generated by a set of
elements.

Example 5.6. 1. hZ, +i is cyclic, generated by 1 or −1.

hZ, +i = h1i = h−1i

2. hZn , +i = h1i is cyclic. In fact, given any integer k so that

gcd(k, n) = 1 we have hZn , +i = hki. (Exercise. Give a proof.)

3. The Klein group is not cyclic.(Exercise. Give a proof.)

4. Un , the nth roots of unity is cyclic. It is generated by the premi-

2πi
tive root ζ = e n . (Exercise. Give a proof.)

19
6 Cyclic Groups
Abstract:Any cyclic group is either isomorhic to hZ, +i or isomorhic
to hZn , +i for some integer n ≥ 1.

6.1 Elementary Properties

Theorem 6.1. Every cyclic group is abelian.

Proof. Let G = hai be a cyclic group generated by a. Then, for

x, y ∈ G we have x = am , y = an for some m, n ∈ Z. So,

xy = am · an = am+n = yx.

The proof is complete.

Theorem 6.2 (Division Algorithm). Suppose m > 0 is fixed positive

integer. Then, for any integer n ∈ Z there are unique integers q, r such
that
n = mq + r with 0 ≤ r < n.

Proof. Exercise.

Theorem 6.3. Let G be cyclic group. Then, any subgroup H of G is

also cyclic.

Proof. Write G = hai. If H = {e} then H = hei is cyclic. Now

assume H 6= {e}. Write

S = {n ∈ Z+ : an ∈ H}.

Since H 6= {e}, the set S is non-empty. Let m be the smallest integer

in S. We calim the g = am generates H. Notationally,

H = ham i = hci

Obviously, ham i ⊆ H. Now, let x = an ∈ H. Then n = mq + r

for some integer 0 ≤ r < m. Then an = (am )q ar and ar =

20
(am )−q an ∈ H. Since 0 ≤ r < m, by minimality of m, we have r = 0
and n = mq. So, an = (am )q ∈ ham i. So, H ⊆ ham i. The proof is
complete.

Corollary 6.4. Let H be subgroup of hZ, +i. Then H = nZ, where n

is the smallest positive integer in H. This n will be called the positive
generator of H.

Proof. It follows directly from the above theorem (and its proof.)
The proof is complete.

Exercise 6.5. Let r, s be two positive integers. Recall the definition

of the greatest common divisor gcd(r, s). Prove that gcd(r, s) is the
positive generator of the subgroup H = {nr + ms : m, n ∈ Z}.

6.2 The Structure of Cyclic groups

Theorem 6.6. Let G be a cyclic group with generator a.

1. If G has finite order n then G is isomorphic to hZn , +i.

2. If G is infinite then G is isomorphic to hZ, +i.

Proof. Suppose G = hai. Suppose G is finite. Then, there are integers

r < s such that ar = as and hence as−r = e. So, am = e for some
integer m > 0. So, the set {m : am = e with m > 0} is nonempty.

Let n = min{m : am = e with m > 0}

Define ϕ : Zn −→ G by assigning ϕ(k) = ak , where k = 0, 1, 2, . . . , n −

1.
First, ϕ is onto. To see this let x = am ∈ G. By division algorithm
m = nq + r for some 0 ≤ r ≤ n − 1. So, x = am = anq+r = (an )q ar =
ϕ(r). It is established that ϕ is onto. Now, to prove that ϕ is one to
one let ϕ(r) = ϕ(s) for some 0 ≤ r ≤ s ≤ n − 1. So, ar = as and
hence as−r = e. Since 0 ≤ s − r ≤ n − 1, by minimality of n we have
s − r = 0. So, ϕ is one to one. Also, note ϕ(x + y) = ϕ(x)ϕ(y) for all

21
x, y ∈ Zn . So, it is established that G is isomorphic to Zn , as group
structures.
Now suppose G is infinite. Define

ϕ : Z −→ G by ϕ(r) = ar .

It follows ϕ(r + s) = ar+s = ar as = ϕ(r)ϕ(s). So, ϕ is a well defined

homomorphism of the binary structures. In fact, ϕ is onto, because
G = {ar : r ∈ Z}. Now we prove that ϕ is one to one. Suppose
ϕ(r) = ϕ(s). So, ar = as . We assume r ≤ s. So, as−r = e. Write
m = s − r ≥ 0. If m > 0, using division algorithm, it follows G =
{e, a, a2 , . . . , am−1 }. Since G is infinite, this is not possible. So, r = s
and ϕ is one to one. So, ϕ is an isomorphism. The proof is complete.

22
6.3 Subgroups of Finite Cyclic Groups
Theorem 6.7. Let G = hai be a finite cyclic group of order n. Let
n
b = as and H = hbi. Order of H is |H| = d where d = gcd(s, n)

Proof. Read from the textbook.

1. In fact, we may assume

G = hZn , +i = h{0, 1, 2, . . . , n − 1}, +i with a = 1.

2. The statement of the theorem means, if b = s and H = hsi, then

n
H has gcd(s,n) elements.
n
3. In the easy case, if s|n then d = gcd(s,n) = n/s and

H = h{0, s, 2s, . . . , ((d − 1)s)}, +i

n
4. Again, if s 6 |n, then with d = gcd(s,n) , whe have

H = h{0, s, 2s, . . . , ((d − 1)s)}, +i

The proof is complete.

Reading Assignment: Read Examples 6.13, 6.15-6.17.

23
7 Generating Sets
Abstract: Given a group and a subset S ⊆ G, we define the smallest
subgroup H of G containing S. This H is called the subgroup of G
generated by S.
For a group G and a ∈ G the subgroup generated by a was the
cyclic group H = hai.

Theorem 7.1. Let G be a group. Suppose Hi is a set of subgroups of

T
G indexed by i ∈ I. Then the intersection H = i∈I Hi is a subgroups
of G.

Proof. We have check three conditions (we do not need to check as-
sociativity).
1. First, we need to show H is closed under multiplication.

x, y ∈ H =⇒ (x, y ∈ Hi ∀ i ∈ I) =⇒ (x · y ∈ Hi ∀ i ∈ I)

because Hi are subgroups. So, x · y ∈ H and H is closed under

multiplication.
2. We do not have to check associativity of multiplication in H,
because it is inherited from G.
3. Again, since Hi are subgroups

(e ∈ Hi ∀ i ∈ I) =⇒ e ∈ H.

So, e ∈ H, which satisfies the property of the identity in H.

4. Inverse: let a ∈ H.

a ∈ H =⇒ (a ∈ Hi ∀ i ∈ I) =⇒ (a−1 ∈ Hi ∀ i ∈ I)

because Hi are subgroups. So, a−1 ∈ H, which satisfies the

property of inverse in H.
So, H is a subgroup of G. The proof is complete.

24
Definition 7.2. Let G be a group and S = {ai : i ∈ I} ⊆ G.
1. Then, the smallest subgroup G(S) of G is called the subgroup
generated by S. So,
\
G(S) = {H ≤ G : S ⊆ H}

Note that there is one subgroup, namely G, of G that contains

S. So, the right hand side in not an empty-intersection.
2. If G = G(S) we say that G is generated by S. We also say that
G is generated by {ai }.
3. If there is a finite set S = {a1 , a2 , . . . , an } that generates G then
we say that G is finitely generated. If there is no such finite
set, we say G is infinitely generated.

Theorem 7.3. Let G be a group and S = {ai : i ∈ I} ⊆ G is a subset.

Let G(S) be the subgroup generated by S.
1. Write S −1 = {a−1
i : i ∈ I}. Then, G(S) consists of all the
"words" (of finite length) written with S∪S −1 = {ai , a−1
i : i ∈ I}.

2. Note, such a "word" looks like w = x1 x2 · · · xn where xj = ai or

xj = a−1 −1
i for some i. When adjacent "letters" are ai and/or ai ,
we can combine them and write w = y1n1 y2n2 · · · yrnr where yj = ai
for some i.

Proof. We only need to prove (1). Let H be the set consisiting of all
such "words". Then
1. S ⊆ H because ai is an word of lenght one.
2. e = ai a−1
i ∈ H.

3. Let w = x1 x2 · · · xn is a "word" xj = ai or xj = a−1

i for some i.
−1 −1 −1
Then w−1 = x−1
n xn−1 · · · x2 x1 is a "word" of the same kind.
So, w ∈ H.
So, H is a subgroup of G containing S. Now, if K is a subgroup of G
containing S, then each such "word" is in K. So, H ⊆ K. So, H is
the smallest such group and H = G(S). The proof is complete.

25
Corollary 7.4. Suppose G and S be as above (7.3). Assume G is
abelian. Then,

G = {an1 1 an2 2 · · · anr r : r ≥ 0, ni ∈ Z, ai ∈ S are distinct}.

In additive notation:

G = {n1 a1 + n2 a2 + · · · + nr ar : r ≥ 0, ni ∈ Z, ai ∈ S are distinct}.

Proof. Follows directly from (7.3), because we can switch the ele-
ments. The proof is complete.

26