Contents

Course Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1 Cyclic Groups 7

2 Permutation Groups 15

3 Cosets and Normal Subgroups 25

4 Homomorphisms 33

5 Direct Products 41

6 Action of a Group on a Set 45

1
Benedict Vasco Normenyo
 Course Outline

Department of Mathematics
University of Ghana
Second Semester of 2017-2018 Academic Year

COURSE CODE : MATH 354

COURSE TITLE : Abstract Algebra I
CREDITS : Three(3)
COURSE INSTRUCTOR:

Name Mr. Benedict Vasco Normenyo

Office Department of Mathematics Room 29
E-mail [email protected]
Office Hours Tuesday 9:30am - 1:20pm, Wednesday 9:30am - 11:20am
Lecture Hours Tuesday 1:30pm - 3:20pm, Wednesday 11:30am - 1:20pm
PREREQUISITES : Students registered for this course studied logic, the concept of math-
ematical proof, set theory and the concept of a function when they took MATH 121. A
pass in MATH 121 ensured that they formed part of the MATH 224 class. In MATH
224, elementary number theory and group theory were treated. A pass in MATH 224
qualifies a student to register for this course. We shall build on the foundations laid in
MATH 121 and MATH 224. It is therefore very important that students revise the very
essential concepts they were exposed to in those courses.

LEARNING OUTCOMES : By the end of this course, students should be able to:

1. define cyclic a group

2. give examples of cyclic groups group.

3. give examples of groups.

4. define the order of an element of a group.

5. give the relations between the order of an element of a group and the order of the
group.

6. prove some theorems about cyclic groups.

7. compute subgroups of a cyclic group and draw a subgroup lattice.

8. define a permutation of a set.

9. prove that a given mapping is a permutation.

10. compute products and inverses of permutations.

2
Benedict Vasco Normenyo
 11. state and prove Cayley’s theorem.

12. compute orbits of a permutation.

13. define a cycle.

14. express a permutation as a product of transpositions.

15. determine the parity of a permutation.

16. state the definition of a coset.

17. state and prove Lagrange’s theorem.

18. prove some results involving cosets.

19. compute cosets.

20. define normal subgroup.

21. state and prove the three equivalent characterizations of normal subgroups.

22. define a factor group.

23. define a homomorphism.

24. show that a given map is a homomorphism/isomorphism or otherwise.

25. prove some important results about homomorphisms and factor groups.

26. state and prove the fundamental homomorphism theorem.

27. state the definitions of external and internal direct products.

28. prove certain characterisations of external and internal direct products.

29. find the order of an element in external direct products.

30. prove certain results relating to direct products.

31. define an action of a group on a set.

32. show that a given map is an action of a group on a set.

33. define/compute isotropy (stabilizer) subgroup of x in a group G, where x ∈ X and

X is a set.

34. define/compute orbit in a set X under a group G.

35. state and prove the orbit-stabilizer theorem.

36. derive the class equation.

3
Benedict Vasco Normenyo
ASSIGNMENTS AND EVALUATION: At the end of each chapter, exercises are pro-
vided. Selected questions from the exercises will constitute assignments. There will be
two class tests and an end-of-semester examination. Assessment will be conducted as
follows:

1. Assignments and Tutorials: 20%

2. Two class tests: 20%

3. End of semester exam: 60%

INSTRUCTIONAL METHOD : In each week, there will be three hours of lectures and
one hour of tutorials. During tutorials, solutions to the assignments and class tests will
be provided and discussed. Students are expected to solve questions in the exercises not
given as assignments as they may be called to provide such solutions on the board for
class discussion.

RECOMMENDED BOOKS

1. J.B. Fraleign, A First Course in Abstract Algebra,7th edition,

Addison Wesley, 2003.

2. J.J. Rotman, A First Course in Abstract Algebra with Applications, 3rd edition,
Pearson Prentice Hall, 2006.

3. I.N. Herstein, Abstract Algebra, 2nd edition, Macmillan, 1990.

COURSE CONTENTS :

1. Cyclic Groups (Feb 5 - Feb 18)

(a) Order of an element and some related results.

(b) Definition of a cyclic group with examples.
(c) Some relevant results relating to cyclic groups.
(d) Subgroups of a cyclic group and their subgroup lattice.

2. Permutations (Feb 19 - Mar 4)

(a) Definition of a permutation with examples.

(b) Products and inverses of permutations.
(c) Cayley’s theorem.
(d) Orbits of a permutation and cycles.
(e) Transpositions and parity of permutations.

3. Cosets and Normal Subgroups (Mar 5 - Mar 25)

(a) Definition of cosets and normal subgroups.

4
Benedict Vasco Normenyo
 (b) Lagrange’s theorem.
(c) Characterisations of normal subgroups.
(d) Factor groups.

4. Homomorphisms (Mar 26 - Apr 8)

(a) Definition of homomorphism with examples.

(b) Properties of homomorphisms.
(c) Isomorphisms.
(d) The fundamental theorem of homomorphisms.

5. Direct Products (Apr 9 - Apr 22)

(a) External direct products.

(b) Internal direct products.

6. Action of a Group on a Set (Apr 23 - May 4)

(a) Definition and examples.

(b) Isotropy (Stabilizer) subgroup.
(c) Orbits
(d) Orbit-Stabilizer theorem.
(e) The class equation.

PLAGIARISM POLICY : Plagiarism in any form is unacceptable and shall be treated

as a serious offence. Appropriate sanctions, as stipulated in the Plagiarism Policy, will
be applied when students are found to have violated the Plagiarism policy. The policy is
available at http://www.ug.edu.gh/aqau/policies-guidelines. ALL students are expected
to familiarize themselves with the contents of the Policy.

5
Benedict Vasco Normenyo
Contents

6
Benedict Vasco Normenyo
Chapter 1
Cyclic Groups

Let g be an element of a group G. Let e be the identity element in G. We define the

powers g n of g, for n ∈ Z, as follows.

g0 =e
g n = ggg...g (n factors) , if n > 0
−n −1 n −1 −1 −1 −1
g = (g ) = g g g ...g (n factors) , if n > 0

In additive notation, we have that

0g =0
ng = g + g + g + ... + g (n summands) , if n > 0
−ng = n (−g) = (−g) + (−g) + (−g) + ... + (−g) (n summands) , if n > 0

Definition 1.1. (Order of an element) If G is a group and g ∈ G, then g is said to

be of finite order if there exists n ∈ Z+ such that g n = e. If such an integer exists, then
the smallest positive integer n such that g n = e is called the order of g and is denoted by
o(g). If g is not of finite order, then we say that g is of infinite order and write o(g) = ∞.

Example 1.1. Consider the following.

(a) Let G be (Z3 , +3 ). Then o(1) = 3, since 1 6= 0, 1 +3 1 = 2 6= 0 and 1 +3 1 +3 1 = 0.
(b) Let G be (Z, +). Then o(1) = ∞, since 1 6= 0, 1 + 1 = 2 6= 0 and 1 + 1 + 1 = 3 6= 0,
n(1) 6= 0 for n > 0.

(c) Let G be Q+ , · . Then o(2) = ∞, since 2 6= 1, 22 = 4 6= 1, 23 = 8 6= 1, 2n 6= 1 for
n > 0.

Theorem 1.1. Let G be a group and x ∈ G. Then the following hold.

1. o(x) = o(x−1 ).
2. If o(x) = n and xm = e, then n | m.
3. If o(x) = n and gcd(m, n) = d, then o(xm ) = nd .
PROOF : Let G be a group and x ∈ G.
(1) Let o(x) = n. Then n is the least positive integer such that xn = e. Thus,

(xn )−1 = e−1 = e, and so we get

n
(x−1 ) = e.
n
That is, n is the smallest positive integer such that (x−1 ) = e. This means that
o(x−1 ) = n. Therefore, o(x) = o(x−1 ).

7
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups

(2) Let o(x) = n and xm = e. Then n ≤ m. By the division algorithm, There exists
unique q, r ∈ Z with 0 ≤ r < n such that m = nq + r.

xm = e =⇒ xnq+r = e
=⇒ xnq xr = e
=⇒ (xn )q xr = e
=⇒ (e)q xr = e
=⇒ exr = e
=⇒ xr = e
Since n is the least positive integer such xn = e, we have that r = 0. This means that
m = nq, and so n | m.
(3) Let o(x) = n and gcd(m, n) = d. We get (xm )n/d = (xn )m/d = em/d = e. Suppose
that there exists k ∈ Z+ such that k < nd and (xm )k = e (that is, xmk = e). The result
obtained in (ii) implies that n | mk, and so there exists s ∈ Z such that mk = sn. This
leads to m n n m
k=s , which imples that k.
d d d d
Since gcd md , nd = 1, we must have that nd |k . That is nd ≤ k. This contradict the

assumption that k < nd .

Definition 1.2. (Cyclic Group) A group G is called cyclic if there exists x ∈ G such
that G = {xn | n ∈ Z}. The element x is then called a generator for G.
Notation : The set {xn | n ∈ Z} is denoted by hxi. Thus, G is cyclic with x as a generator
if and only if G = hxi. In additive notation, hxi = {nx | n ∈ Z}.
Example 1.2. (a) Z1 is the trivial group {0}. Clearly, Z1 = h0i. If n ≥ 2, then Zn = h1i
and 1, 1 + 1, 1 + 1 + 1, ..., 1 + 1 + 1 + ... + 1(n terms) gives us all the elements in Zn . For
instance, Z4 = {0, 1, 2, 3}. (b) (Z, +) is cyclic with generator 1, that is,
Z = h1i = {n(1) | n ∈ Z}.

Example 1.3. The group (Q, +) is not cyclic. Clearly, 0 is not a generator. Let q ∈ Q
with q 6= 0. Now, 2q ∈ Q but 2q ∈
/ hqi since 2q = 12 q with 12 ∈
/ Z. This means that Q 6= hqi.

Theorem 1.2. Let G = hxi. If o(x) = ∞, then xj 6= xk for j 6= k, and consequently G

is infinite. If o(x) = n, then xj = xk if and only if j ≡ k (mod n) and consequently the
distinct elements of G are e, x, x2 , ..., xn−1 .
PROOF : Let G = hxi. Suppose that o(x) = ∞ and that xj = xk for j 6= k, say j > k.
Then xj−k = e with j −k > 0. Thus x has a finite order, which contradicts the assumption
that o(x) = ∞. Therefore, xj 6= xk for j 6= k and G is infinite.
Now, suppose that o(x) = n. Then

xj = xk ⇐⇒ xj−k = e
⇐⇒ n | (j − k)
⇐⇒ j ≡ k (mod n).

8
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups
Definition 1.3. (Order of a group) The order of a group G, denoted by |G|, is the
number of elements in G.

Corollary 1.1. If G = hxi, then |G| = o(x).

Theorem 1.3. If G is a cyclic group, then G is Abelian.

PROOF : Let G be a cyclic group. Then G = hxi for some x ∈ G. Let a, b ∈ G. Then
a = xm and b = xn where m, n ∈ Z. We have that

ab = (xm )(xn ) = xm+n = xn+m = (xn )(xm ) = ba.

Hence G is Abelian.

Theorem 1.4. Let G be a group and let a ∈ G. Then H = hai = {an | n ∈ Z} is

a subgroup of G and is the smallest subgroup of G that contains a. (That is, every
subgroup of G containing a contains H).

PROOF : Let G be a group and let a ∈ G such that H = hai = {an | n ∈ Z}. Clearly,
H 6= {} since a ∈ H. Let aj , ak ∈ H. We find that aj ak = aj+k ∈ H and (aj )−1 = a−j ∈ H
since j + k, −j ∈ Z. Hence H is a subgroup of G. Now, suppose that I is a subgroup of
G containing a. We want to prove the statement P (n) : an ∈ I ∀ n ∈ Z.
P (1) is true since a ∈ I. Suppose that P (k) is true for k ≥ 2, that is, ak ∈ I. Then
we obtain ak+1 = ak a ∈ I, since ak , a ∈ I and I is a subgroup of G. This means that
P (k + 1) is true. Hence P (n) is true for all n ∈ Z+ .
Now, let m = −n ∈ Z− , where n ∈ Z+ . Then am = a−n = (a−1 )n ∈ I since a−1 ∈ I and
so (a−1 )n ∈ I for n ∈ Z+ . Finally, a0 = e ∈ I, I being a subgroup of G. We have shown
that an ∈ I for all n ∈ Z, hence the result follows.

Definition 1.4. Let G be a group and let a ∈ G. Then the subgroup hai = {an | n ∈ Z}
is called the cyclic subgroup of G generated by a.

Theorem 1.5. Let G be a cyclic group. Then every subgroup of G is cyclic.

PROOF : Let G be a cyclic group with G = hai. If a = e, then the result easily follows.
So suppose that a 6= e. Let H be a subgroup of G. If H = {e}, then H is cyclic with
H = hei. Suppose that H 6= {e}. Then there exists t ∈ Z+ such that at ∈ H. Let m be
the smallest integer in Z+ such that am ∈ H (m exists by the Well-Ordering Principle).
Now, let b ∈ H. Then b = an for some n ∈ Z, since H ≤ G. By the division algorithm,
there exists unique q, r ∈ Z such that n = mq + r with 0 ≤ r < m. This leads to
an = amq+r = (am )q ar , from which we get ar = (am )−q an . Since an , am ∈ H and H is
a subgroup of G, we see that ar = (am )−q an ∈ H. Since m is the least positive integer
such that am ∈ H, and 0 ≤ r < m, we find that r = 0. Hence n = mq, which means that
b = (am )q for some q ∈ Z. Therefore, H = ham i.

Example 1.4. Let G = (Z, +) and a = 2. Then hai = h2i = (2Z, +). In general,
nZ = hni. Observe that 6Z < 3Z and 4Z < 2Z.

9
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups
Example 1.5. Consider the Klein 4-group V = {e, a, b, c} with a2 = b2 = c2 = e,
ab = ba = c, ac = ca = b and bc = cb = a. The only subgroup that contains none of a, b
or c is {e} = hei. If a subgroup contains just one of a, b or c, then it is either hai, hbi or
hci. If a subgroup contains two of a, b, c, then it contains the third as well and so must be
V . Thus, the subgroups are hei, hai, hbi, hci and V . The Klein 4-group is an example of
an Abelian non-cyclic group with the property that all of its proper subgroups are cyclic.
Below is a subgroup lattice or a subgroup diagram for the Klein 4-group.

hai hbi hci

hei

Figure 1.1: A subgroup diagram for the Klein 4-group

Theorem 1.6. Let G be a cyclic group with generator a. If the order of G is infinite,
then G is isomorphic to Z. If G has a finite order n, then then G is Isomorphic to Zn .

PROOF : Let G = hai.

Case I : Let the order of G be infinite. Then for all m ∈ Z+ , am 6= e. Suppose that
ah = ak where h > k. Then ah−k = e. This contradicts the assumption that am 6= e
for all m ∈ Z+ . Hence we deduce that no two distinct exponents give the same element.
This implies that for all b ∈ G, there exists a unique m ∈ Z such that b = am . Let us
define the map
φ : G −→ Z by φ(ai ) = i.
Let u, v ∈ G such that u = v. Then u = as and v = at where s = t. It follows that
φ(u) = φ(as ) = s and φ(v) = φ(at ) = t. Thus, φ(u) = φ(v) and so φ is well defined.
Let x, y be any two elements in G. Then x = ai and y = aj for some unique i, j ∈ Z. We
get
φ(xy) = φ(ai aj ) = φ(ai+j ) = i + j = φ(ai ) + φ(aj ) = φ(x) + φ(y).
So, φ is a homomorphism.
Suppose that φ(x) = φ(y). Then i = j, meaning x = y. Hence φ is one-to-one.
Let c be any element in Z. Choose an element ac in G. Then φ(ac ) = c. This implies
that φ is onto. Therefore φ is an isomorphism.
Case II : Let the order of G be finite. Then there exists m ∈ Z+ such that am = e. Let
n be the smallest positive integer such that an = e (That is, |G| =o(a) = n). Let s ∈ Z.

10
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups
Then ∃! q, r ∈ Z such that s = nq + r where 0 ≤ r < n. Now,

as = anq+r = (an )q ar = eq ar = ar .

Suppose that 0 < k < h < n and ah = ak . Then ah−k = e and 0 < h − ka < n, which
contradicts our assumption that n is the least positive integer such that an = e. This
shows that a0 = e, a1 , a2 , a3 , ..., an−1 are all distinct and constitute all elements in G.
Consider the map

ψ : G −→ Zn defined by ψ(ai ) = i, for i = 0, 1, 2, ...., n − 1.

We see that ψ is one-to-one and onto. Also,

ψ(ai aj ) = ψ(ai+n j ) = i +n j = ψ(ai ) +n ψ(aj ).

Therefore, ψ is an isomorphism.

Theorem 1.7. Let G = hxi be a cyclic group of order n. Then the following hold.
(i) For any positive integer m, G has a subgroup of order m if and only if m | n.
(ii) If m | n, then G has a unique subgroup of order m.
(iii) Two powers xr and xs of x generate the same subgroup of G if and only if
gcd(r, n) = gcd(s, n).

PROOF : Let G = hxi be a cyclic group of order n. Then o(x) = n.

(i) Let m ∈ Z+ . Suppose that m | n. Then m n
∈ Z+ and xn/m ∈ G. So hxn/m i is a
subgroup of G by Theorem 1.4. Furthermore, we have that

hxn/m i = o xn/m = gcd nn, n , by (iii) of Theorem 1.1

( m)
n
= (n/m)
= m.

Hence hxn/m i is a subgroup of order m. Conversely, assume that G has a subgroup H

of order m. By Theorem 1.5 H is cyclic, so let H = hxk i for some k ∈ Z+ . Then
n
m = o(xk ) = gcd(k,n) , and so m · gcd(k, n) = n. Hence m | n.
(ii) Suppose that m | n. Then G has a subgroup of order m. Let H1 = hxk1 i and
H2 = hxk2 i be two subgroups of G of order m where . Then xk1 and xk2 are the smallest
positive powers of x that lie in H1 and H2 respectively. We have that

|H1 | = m = gcd(kn1 ,n)

|H2 | = m = gcd(kn2 ,n) ,

which leads to gcd(k1 , n) = gcd(k2 , n). If t is any integer such that xt ∈ H1 , then k1 | t.
In particular, k1 | n since xn = e ∈ H1 . Thus gcd(k1 , n) = k1 . Similarly, gcd(k2 , n) = k2 .
Thus k1 = k2 , and so H1 = H2 .

11
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups
(iii) Let xr , xs ∈ G. Then we have that

hxr i = hxs i ⇐⇒ |hxr i| = |hxs i|

⇐⇒ o(xr ) = o(xs )
n n
⇐⇒ gcd(r,n) = gcd(s,n)
⇐⇒ gcd(r, n) = gcd(s, n)

Corollary 1.2. Let d(n) be the number of positive divisors of n. If G = hxi is a cyclic
group of order n and d1 , d2 , ..., dd(n) are the distinct positive divisors of n, then xd1 , xd2 ,
..., xdd(n) are the distinct subgroups of G.

PROOF : Since di | n =⇒ gcd(di , n) = di , we see that di =

6 dj =⇒ gcd(di , n) 6= gcd(dj , n).
Hence x 6= x . Thus the subgroups x , x , ..., xdd(n) are distinct and give
di dj d1 d2

all d(n) subgroups of G.

Example 1.6. Let G = Z12 . Then G = h1i and |G| = 12. The positive divisors of 12
are 1, 2, 3, 4, 6, 12. The distinct subgroups of G are

h1(1)i = h1i h2(1)i = h2i h3(1)i = h3i

h4(1)i = h4i h6(1)i = h6i h12(1)i = h0i

Observe that h5i = h1i = G since gcd(5, 12) = gcd(1, 12). Similarly, h7i = h1i = h11i.
We also have that h8i = h4i since gcd(8, 12) = gcd(4, 12). Below is a subgroup lattice or
a subgroup diagram of Z12 .

h1i

h2i h3i

h4i h6i

h0i

Figure 1.2: A subgroup lattice for Z12 .

Theorem 1.8. Let G = hxi be an infinite cyclic group. Then hei, hxi, hx2 i, ... are all
the distinct subgroups of G.

12
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups
Exercise One

1. Find the number of elements in the indicated cyclic group.

(a) The cyclic subgroup of Z30 generated by 25

(b) The cyclic subgroup of Z42 generated by 30
(c) The cyclic subgroup hii of the group C∗ of nonzero complex numbers under
multiplication
(d) The cyclic subgroup of the group√C∗ of nonzero complex numbers under mul-
tiplication generated by (1 + i)/ 2
(e) The cyclic subgroup of the group C∗ of nonzero complex numbers under mul-
tiplication generated by 1 + i.

2. Let G = hxi be a cyclic group of order 144. How many elements are there in the
subgroup hx26 i?

3. List all the elements of Z45 that are of order 15.

4. Show that Zp has no proper nontrivial subgroups if p is a prime number.

5. Let G be a group and suppose a ∈ G generates a cyclic subgroup of order 2 and is

the unique such element. Show that ax = xa for all x ∈ G.
h
2 i
Hint : Consider xax−1 .

6. Obtain all subgroups of Z18 and give their subgroup diagram.

7. Let G be an abelian group and let H and K be finite cyclic subgroups with |H| = r
and |K| = s.

(a) Show that if r and s are relatively prime, then G contains a cyclic subgroup
of order rs.
(b) Generalizing part (a), show that G contains a cyclic subgroup of order the
least common multiple of r and s.

8. Show that a group with no proper nontrivial subgroups is cyclic.

9. If G is a group and a, b ∈ G, prove that ab and ba have the same order.

13
Benedict Vasco Normenyo
Chapter 1. Cyclic Groups

14
Benedict Vasco Normenyo
Chapter 2
Permutation Groups

Definition 2.1. (Permutation) Let X be a non-empty set. A one-to-one and onto

mapping δ : X −→ X is called a permutation of X.

If the set X is finite (say number of elements in X is n), then the total number of
permutations of X is the number of ways of filling in the table

x x1 x2 x3 ··· xn−1 xn
δ(x) ? ? ? ? ?

in a one-to-one and onto manner.

Example 2.1. Let X = {1, 2, 3}. Then the permutations are the following.

x 1 2 3 x 1 2 3 x 1 2 3
δ1 (x) 1 2 3 δ2 (x) 1 3 2 δ3 (x) 2 1 3

x 1 2 3 x 1 2 3 x 1 2 3
δ4 (x) 2 3 1 δ5 (x) 3 1 2 δ6 (x) 3 2 1

In general, if a set X has n elements, then there are n! possible permutations of X. We

let X be the set {1, 2, 3, · · · , n} and denote by Sn the set of all possible permutations of
X. Let δ ∈ Sn . Then instead of the table, we represent δ by
 
1 2 3 ··· n
δ(1) δ(2) δ(3) δ(n)
Thus, the elements of S3 as described in Example 2.1 are represented as
     
1 2 3 1 2 3 1 2 3
δ1 = δ2 = δ3 =
1 2 3 1 3 2 2 1 3
     
1 2 3 1 2 3 1 2 3
δ4 = δ5 = δ6 =
2 3 1 3 1 2 3 2 1

Definition 2.2. (Product of Permutations) Let f, g ∈ Sn . Then the product f g of f

and g is the function composition f ◦ g of f and g.

15
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
Example 2.2. From Example 1.7 we get

δ1 δ2 (1) = δ1 (δ2 (1)) = δ1 (1) = 1

δ1 δ2 (2) = δ1 (δ2 (2)) = δ1 (3) = 3
δ1 δ2 (3) = δ1 (δ2 (3)) = δ1 (2) = 2.

This means that

    
1 2 3 1 2 3 1 2 3
δ1 δ2 = = .
1 2 3 1 3 2 1 3 2
Similarly, we get
     
1 2 3 1 2 3 1 2 3
δ5 δ6 = , δ5 δ4 = , δ 6 δ2 = ,
2 1 3 1 2 3 3 1 2
and so on.

The inverse of the permutation

 
1 2 3 ··· n
δ(1) δ(2) δ(3) δ(n)

is  
δ(1) δ(2) δ(3) · · · δ(n)
.
1 2 3 n

Example 2.3. From Example 1.7 let us consider

 
1 2 3
δ4 = .
2 3 1

Then we obtain    
2 3 1 1 2 3
δ4−1 = = .
1 2 3 3 1 2
It is easy to check that δ4 δ4−1 = δ1 .

Theorem 2.1. Let X be a non-empty set and let SX be the collection of all permutations
of X. Then SX is a group under permutation multiplication.
PROOF : Let f and g be permutations of X. Then f g is the function f ◦ g : X −→ X.
Let (f g)(x) = (f g)(y), where x, y ∈ X. Then

f [g(x)] = f [g(y)]
g(x) = g(y), since f is one-to-one
x = y, since g is one-to-one.

g f
Hence, f g is one-to-one. Note that X −→ X −→ X. Let y ∈ X(codomain of f ). Then
there exists z ∈ X(domain of f ) such that f (z) = y, since f is onto. Similarly, there

16
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
exists x ∈ X(domain of g) such that g(x) = z, since g is onto. Now, choose
x ∈ X(domain of f g). Then (f g)(x) = f [g(x)] = f (z) = y. Hence f g is onto.
Therefore, f g is a permutation of X. Function composition is associative, so permutation
multiplication is associative. The permutation that maps every element of X to itself is
the identity element. A permutation is a bijection and so has inverse.

Definition 2.3. (Symmetric Group) Let X = {1, 2, 3, ..., n}. The group of all permu-
tations of X is called the symmetric group on n letters and is denoted by Sn .

Example 2.4. let us consider the equilateral triangle with vertices labelled 1, 2, 3. Let

ρi represent clockwise rotations of the triangle.

µi represent mirror images in bisectors of angles.

Then we obtain the following.

   
1 2 3 1 2 3
ρ0 = µ1 =
1 2 3 1 3 2
   
1 2 3 1 2 3
ρ1 = µ2 =
2 3 1 3 2 1
   
1 2 3 1 2 3
ρ2 = µ3 =
3 1 2 2 1 3

S3 is also the group D3 of symmetries of an equilateral triangle. The nth dihedral group
Dn is the group of symmetries of the regular n-gon. Some denote Dn by D2n .

Example 2.5. Consider a square with vertices 1, 2, 3, 4.

4 3

1 2

Let ρi represent rotations;

µi represent mirror images in perpendicular bisection of sides;
δi represent diagonal flips.
Then we have the following.
   
1 2 3 4 1 2 3 4
ρ0 = µ1 =
1 2 3 4 2 1 4 3
   
1 2 3 4 1 2 3 4
ρ1 = µ2 =
2 3 4 1 4 3 2 1

17
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
   
1 2 3 4 1 2 3 4
ρ2 = δ1 =
3 4 1 2 3 2 1 4
   
1 2 3 4 1 2 3 4
ρ3 = δ2 =
4 1 2 3 1 4 3 2
This is the group D4 of symmetries of the square. It is also called the octic group.

Definition 2.4. Let f : A −→ B be a function and let H be a subset of A. The image

of H under f is {f (h)|h ∈ H} and is denoted by f [H].

Lemma 2.1. Let G and G0 be groups and let φ : G −→ G0 be a 1-1 function such that

φ(xy) = φ(x)φ(y) for all x, y ∈ G.

Then, φ[G] is a subgroup of G0 and φ provides an isomorphism of G with φ[G].

PROOF : Let a, b ∈ φ[G]. Then, a = φ(m), b = φ(n) for some m, n ∈ G. Now, as e ∈ G,

we have that φ(e) ∈ φ[G]. Thus, φ[G] is not empty. Also, ab = φ(m)φ(n) = φ(mn).
Hence, ab ∈ φ[G] since mn ∈ G. Furthermore, a−1 = (φ(m))−1 = φ(m−1 ) ∈ φ[G], since
m−1 ∈ G.

Theorem 2.2. (Cayley’s Theorem) Every subgroup is isomorphic to a group of per-

mutations.

PROOF : For x ∈ G, let

λx : G −→ G be defined by λx (g) = xg ∀ g ∈ G.

Let c ∈ G (Codomain of λx ). Then, there exists x−1 c ∈ G (Domain of λx ) such that

λx (x−1 c) = x(x−1 c) = c. Thus, λx is onto. Let λx (a) = λx (b). Then xa = xb, which
implies that a = b by left cancellation. Hence, λx is 1-1. Therefore, λx is a permutation
of G; that is, λx ∈ SG .
Define φ : G −→ SG by φ(x) = λx , ∀x ∈ G. Suppose that φ(x) = φ(y). Then

λx = λy
λx (a) = λy (a), for all a ∈ G
xa = ya
x = y, by right cancellation.

Thus, φ is 1-1. Let g ∈ G.

(λx λy )(g) = λx (λy (g)) = λx (yg) = x(yg)

= (xy)g, by associativity in G
= λxy (g)

18
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
This implies that λxy = λx λy , and so we get φ(xy) = φ(x)φ(y). By Lemma 2.1, we see
that φ[G] = {λx |x ∈ G} is a group of permutations and G is isomorphic to φ[G].

Definition 2.5. (Orbits) Let σ ∈ SA where A is a set. Define the equivalence relation
∼ on A by:
∀ a, b ∈ A, a ∼ b ⇐⇒ b = σ n (a) for some n ∈ Z.
The equivalence classes in A determined by the equivalence relation ∼ are called the
orbits of σ.

 
1 2 3 4 5 6 7 8
Example 2.6. Find the orbits of the permutation σ = in S8 .
3 8 6 7 4 1 5 2
SOLUTION :

σ 0 (1) = i(1) = 1
σ 1 (1) = 3
σ 2 (1) = σ(3) = 6
σ 3 (1) = σ(σ 2 (1)) = σ(6) = 1
σ 4 (1) = σ(σ 3 (1)) = σ(1) = 3
σ 5 (1) = σ(3) = 6
σ 6 (1) = σ(6) = 1
σ σ σ σ σ
This shows that, 1 −−→ 3 −−→ 6 −−→ 1 −−→ 3 −−→ · · · and so the orbit containing 1 is
{1, 3, 6}. Similarly, we find that the other orbits of σ are {2, 8} and {4, 7, 5}.

Remark 2.1. Since the identity permutation, i, of A leaves each element of A fixed; that
is, i(a) = a ∀ a ∈ A, the orbits of i are one-element subsets {a} of A.

In Example 2.6 we obtained the orbits {1,3,6},{2,8} and {4,7,5}. We can represent the
orbits with circles:

Figure 2.1:

Each of the circles correspond to a permutation. The first one corresponds to

 
1 2 3 4 5 6 7 8
µ= .
3 2 6 4 5 1 7 8

µ has orbits {1,3,6},{2},{4},{5},{7},{8}. A permutation such as µ is called a cycle.

Definition 2.6. A permutation σ ∈ Sn is called a cycle if it has at most one orbit

containing more than one element. The length of a cycle is the number of elements in its
largest orbit.

19
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
Notation : Let x1 , x2 , ..., xr where 1 ≤ r ≤ n, be r distinct elements of {1, 2, ..., n}.
Suppose that σ ∈ Sn such that

x1 −→ x2 , x2 −→ x3 , ..., xr−1 −→ xr , xr −→ x1

and leaves all elements of {1, 2, ..., n} other than x1 , x2 , ..., xr fixed. Then, we denote σ
by (x1 , x2 , ..., xr ), which can also be written as

(xr , x1 , x2 , ..., xr−1 ) or (xr−1 , xr , x1 , ..., xr−2 ) and so on.

This is sometimes called a r-cycle since it has length r. From our discussion above, we
see that µ can be written as µ = (1, 3, 6) or µ = (6, 1, 3) or µ = (3, 6, 1).

Theorem 2.3. Let f ∈ Sn . Then, there exists disjoint cycles f1 , f2 , ..., fn in Sn such that
f = f1 ◦ f2 ◦ ... ◦ fm .

PROOF : Choose some x1 ∈ {1, 2, ..., n}. Suppose that x2 = f (x1 ), x3 = f (x2 ) and so on.
Since {1, 2, ..., n} is finite, there must be a first element in the sequence x1 , x2 , ... which is
the same as a previous element. Say this element is xk and xk = xj , j < k. We claim that
j = 1. Suppose that j > 1. Then xk = xj implies that f (xk−1 ) = f (xj−1 ). This leads to
xk−1 = xj−1 since f is one-to-one, which contradicts the minimality of k. Thus, the first
k − 1 elements of the sequence x1 , x2 , ... are distinct and xk = x1 . Let f1 = (x1 , ..., xk−1 ).
Then, f = f1 ◦ h1 , where h1 permutes the elements of {1, 2, ..., n} other than x1 , ..., xk−1 .
Repeating our argument on h1 , we have h1 = f2 ◦ h2 , where f2 is a cycle disjoint from f1
and h2 permutes the elements of {1, 2, ..., n} not contained in either f1 or f2 . Continuing,
we obtain hm , the identity permutation, having nothing left to permute. Then,

f = f1 ◦ f2 ◦ ... ◦ fm ◦ hm = f1 ◦ f2 ◦ ... ◦ fm .

 
1 2 3 4 5 6 7 8
Example 2.7. Consider the permutation f = in S8 .
3 5 7 4 2 8 1 6
 
1 2 3 4 5 6 7 8
f = (1, 3, 7)
1 5 3 4 2 8 7 6
 
1 2 3 4 5 6 7 8
= (1, 3, 7)(2, 5)
1 2 3 4 5 8 7 6
= (1, 3, 7)(2, 5)(6, 8).

Definition 2.7. A cycle of length 2 (that is, 2-cycle) is called a transposition.

Theorem 2.4. If n ≥ 2, then any cycle in Sn can be written as a product of transposi-

tions.

PROOF : A 1-cycle is the identity. It can be written as (1,2)(1,2). For an r-cycle with
r ≥ 2,
(x1 , x2 , ..., xr ) = (x1 , xr )(x1 , xr−1 )(x1 , xr−2 ) · · · (x1 , x3 )(x1 , x2 ).

20
Benedict Vasco Normenyo
Chapter 2. Permutation Groups  
1 2 3 4 5 6 7 8
Example 2.8. Consider f = . We can write f as a product
3 5 7 4 2 8 1 6
of transpositions as follows.

f = (1, 3, 7)(2, 5)(6, 8) = (1, 7)(1, 3)(2, 5)(6, 8).

Observe that (1, 3, 7) = (4, 7)(1, 7)(1, 4)(1, 3).

Definition 2.8. (Parity of a Permutation) A permutation of a finite set is even or odd

according to whether it can be expressed as a product of an even number of transpositions
or the product of an odd number of transpositions respectively.

Theorem 2.5. If n ≥ 2, then the collection of all even permutations of {1, 2, ..., n} forms
a subgroup of order n!/2 of the symmetric group, Sn .

PROOF : Since n ≥ 2, Sn has the transposition (1,2) and i = (1, 2)(1, 2) is an even
permutation.The product of two even permutations is again even. If σ is a product of
transpositions, the product of the same transposition taken in reverse order is σ −1 . Thus,
if σ is even, then σ −1 is even. Hence, the collection of all even permutations of {1, 2, ..., n}
forms a subgroup of Sn .
Now we want to show that the number of even and odd permutations in Sn are equal.
Let An be the set of even permutations in Sn and Bn , the set of odd permutations in Sn .
Let τ be any fixed transposition in Sn , say τ = (1, 2).
Define λτ : An −→ Bn by λτ (σ) = τ σ, where σ ∈ An . Since σ ∈ An , (1, 2)σ ∈ Bn .
Let σ, µ ∈ An and suppose that λτ (σ) = λτ (µ). Then (1, 2)σ = (1, 2)µ, which means that
σ = µ since Sn is a group. Thus, λτ is one-to-one. Since τ = (1, 2) = τ −1 , we find that
for all ρ ∈ Bn , there exists τ −1 ρ ∈ An such that λτ (τ −1 ρ) = τ (τ −1 ρ) = ρ. Thus, λτ is
onto. Hence, |An | = |Bn |, implying the subgroup An is of order n!/2.

Definition 2.9. (Alternating Group) The subgroup of Sn consisting of the even per-
mutations of n letters is the alternating group, An , on n letters.

21
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
Exercise Two

1. Compute the indicated product involving the following permutations in S6 :

     
1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
σ= ,τ = ,µ = .
3 1 4 5 6 2 2 4 1 3 6 5 5 2 4 3 1 6
(a) τ σ (b) τ 2 σ (c) µσ 2 (d) σ −2 τ (e) σ −1 τ σ

2. Complete the table below for the group D3 of symmetries of an equilateral triangle.

ρ0 ρ1 ρ2 µ1 µ2 µ3
ρ0
ρ1
ρ2
µ1
µ2
µ3

3. Complete the table below for the group D4 of symmetries of a square.

ρ0 ρ1 ρ2 ρ3 µ1 µ2 δ1 δ2
ρ0
ρ1
ρ2
ρ3
µ1
µ2
δ1
δ2

4. Obtain the orbits of the permutation

 
1 2 3 4 5 6 7 8
θ= .
5 3 7 1 8 4 2 6

5. Compute each of the indicated products involving the following permutations in S9

and determine its parity by writing it as a product of transpositions.
   
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
σ= τ=
2 1 8 7 3 6 9 5 4 5 7 6 8 9 1 2 4 3
 
1 2 3 4 5 6 7 8 9
ρ=
8 9 7 5 2 1 3 6 4

(a) τ 2 σ

22
Benedict Vasco Normenyo
Chapter 2. Permutation Groups
(b) ρσ 2
(c) σ −2 τ

6. Let σ ∈ SA , where A is a set. Define the relation ∼ on A by:

∀ a, b ∈ A, a ∼ b ⇐⇒ b = σ n (a) for some n ∈ Z.

Show that ∼ is an equivalence relation on A.

23
Benedict Vasco Normenyo
Chapter 2. Permutation Groups

24
Benedict Vasco Normenyo
Chapter 3
Cosets and Normal Subgroups

Theorem 3.1. Let H be a subgroup of a group G. Let the relation ∼L be defined on G

by
a ∼L b ⇐⇒ a−1 b ∈ H.
Let the relation ∼R be defined on G by

a ∼R b ⇐⇒ ab−1 ∈ H.

Then, ∼L and ∼R are both equivalence relations on G.

PROOF : Let H be a subgroup of a group G. Define ∼L by

a ∼L b ⇐⇒ a−1 b ∈ H.

Let a, b, c ∈ G. We have that a−1 a = e ∈ H, since H is a subgroup of G. It follows

that a ∼L a, which means that ∼L is reflexive. Suppose that a ∼L b. Then a−1 b ∈ H.
This leads to (a−1 b)−1 = b−1 a ∈ H, since H is a subgroup. Thus, b ∼L a and so ∼L is
symmetric. Now suppose that a ∼L b and b ∼L c. Then a−1 b ∈ H and b−1 c ∈ H. This
leads us to a−1 c = (a−1 b)(b−1 c) ∈ H. Hence a ∼L c and so ∼L is transitive. Therefore,
∼L is an equivalence relation on G. Similarly, ∼R is also an equivalence relation on G.

Notation : For ∼L , the equivalence class containing a ∈ G is given by

ā = {x ∈ G|a ∼L x}
= {x ∈ G|a−1 x ∈ H}
= {x ∈ G|a−1 x = h for some h ∈ H}
= {x ∈ G|x = ah for some h ∈ H}
ā = {ah|h ∈ H}.

The equivalence class ā is denoted by aH. Thus, aH = {ah|h ∈ H}. For ∼R , the
equivalence class containing a is, Ha = {ha|h ∈ H}.

Definition 3.1. (Left and Right Cosets) Let H be a subgroup of a group G. The
subset aH = {ah|h ∈ H} of G is the left coset of H containing a, while the subset
Ha = {ha|h ∈ H} is the right coset of H containing a.

Example 3.1. Obtain all the distinct left cosets of the subgroup 3Z of Z.

25
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
SOLUTION : The left coset of 3Z containing m is m + 3Z

m = 0 =⇒ 3Z = {..., −9, −6, −3, 0, 3, 6, 9, ...}

m = 1 =⇒ 1 + 3Z = {..., −8, −5, −2, 1, 4, 7, 10, ...}
m = 2 =⇒ 2 + 3Z = {..., −7, −4, −1, 2, 5, 8, 11, ...}

Lemma 3.1. Let G be any group and let H be a subgroup of G. Let Ha and Hb be
right cosets of H in G. Then, there is a 1-1 correspondence between the elements of Ha
and those of Hb.
PROOF : Let G be any group and let H < G. Let φ : Ha → Hb be defined by
φ(ha) = hb ∀ h ∈ H. Let φ(h1 a) = φ(h2 a). Then,

h1 b = h2 b
h1 = h2 , by right cancellation
h1 a = h2 a,

from which we conclude that φ is one-to-one. Let hb ∈ Hb where h ∈ H. Then, we choose

ha ∈ Ha such that φ(ha) = hb. Hence, φ is onto. Hence the mapping φ : Ha → Hb is a
bijection, (This means that |Ha| = |Hb| = |H|).

Theorem 3.2. (Lagrange’s Theorem) Let G be a finite group and let H be a subgroup
of G. Then, the order of H divides the order of G (that is, |H| divides |G|).
PROOF : Let G be a finite group and let H be a subgroup of G. Suppose there are k
distinct right cosets (or left cosets) of H in G. Since the right cosets form a partition of
G, we must have that

G = Ha1 ∪ Ha2 ∪ · · · ∪ Hak , where a1 , a2 , ..., ak ∈ G,

from which we obtain

|G| = |Ha1 | + |Ha2 | + · · · + |Hak | = |H| + |H| + · · · + |H| = k|H|.

| {z }
k times

Therefore, |H| divides |G|.

Corollary 3.1. Every group of prime order is cyclic.
PROOF : Let G be a finite group with |G| = p, where p is a prime number. Since p ≥ 2,
there exists e 6= a ∈ G and o(a) ≥ 2. Let H = hai. By Lagrange’s theorem |H| divides
|G|, which leads to |H| = |G| as |G| is prime. We get hai = G, and so G is cyclic.
Theorem 3.3. Let G be a finite group and let x ∈ G. Then, o(x) divides |G|. Conse-
quently, x|G| = e ∀ x ∈ G.
PROOF : Let G be a finite group and let x ∈ G. Let H = hxi. Then, H is a subgroup of
G and |H| = o(x). By Lagrange’s theorem, |H| divides |G|. Therefore, o(x) divides |G|.
Let us take o(x) = n, then |G| = mn, for some m ∈ Z+ . We then have the following:

x|G| = xmn = (xn )m = em = e.

26
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
Definition 3.2. (The Index of a Subgroup) Let H be a subgroup of a group G. The
number of distinct right (or left) cosets of H in G is called the index of H in G, denoted
by (G : H) or [G : H].

Theorem 3.4. Suppose H and K are subgroups of a group G such that K 6 H 6 G

and suppose (H : K) and (G : H) are both finite. Then, (G : K) is finite and
(G : K) = (G : H)(H : K).

PROOF : We have that |H| = |K|(H : K) and |G| = |H|(G : H). This means that
|G| = |K|(H : K)(G : H). This implies that (G : K) = (H : K)(G : H).

|G|
Remark 3.1. It is important to note that (G : H) = |H|
.

Definition 3.3. (Normal Subgroup) A subgroup H of a group G is normal if its left

and right cosets coincide; that is, if gH = Hg for all g ∈ G. We write H / G to mean
that H is a normal subgroup of G.

Theorem 3.5. Let H be a subgroup of G. Then, the following are equivalent conditions
for H to be a normal subgroup of G.

(i) ghg −1 ∈ H ∀ g ∈ G and h ∈ H;

(ii) gHg −1 = H ∀ g ∈ G;

(iii) gH = Hg ∀ g ∈ G [gh1 = h2 g].

PROOF : Let H be a subgroup of G.

To prove (i)=⇒(ii) : Suppose (i) is true. Let x ∈ gHg −1 . Then, x = ghg −1 for some
h ∈ H. Thus, x ∈ H by (i). Hence, gHg −1 ⊆ H. Let x ∈ H. Then, x = h for some
h ∈ H. We get

x = (g −1 g)h(g −1 g)
= g −1 (ghg −1 )g
= g −1 h1 g, where h1 = ghg −1 ∈ H
= (g −1 )h1 (g −1 )−1
= rh1 r−1 , where r = g −1 ∈ G.

This means that x ∈ gHg −1 and so H ⊆ gHg −1 . This gives us gHg −1 = H.

To prove (ii)=⇒(iii) : Suppose that (ii) is true. Let x ∈ gH. Then, x = gh for some
h ∈ H. It follows that

x = gh(g −1 g) = (ghg −1 )g = h1 g, where h1 = ghg −1 ∈ H.

27
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
This means that x ∈ Hg, and so gH ⊆ Hg. Now let y ∈ Hg. Then, y = hg for some
h ∈ H.

y = (gg −1 )hg
= g(g −1 hg)
= g(g −1 h(g −1 )−1 )
= g(rhr−1 ), where r = g −1 ∈ G
= gh2 , h2 = rhr−1 ∈ H

Thus, we have y ∈ gH and so Hg ⊆ gH. Therefore, gH = Hg.

To prove (iii)=⇒(i) : Suppose that (iii) is true. Then, gh = h0 g for some h, h0 ∈ H. This
means that ghg −1 = h0 gg −1 = h0 ∈ H. Therefore ghg −1 ∈ H.

Notation : If H / G, then G/H = {xH | x ∈ G} denotes the set of left cosets of H in G.

Theorem 3.6. Let H be a subgroup of a group G. Then, left coset multiplication is

well-defined by the equation (aH)(bH) = (ab)H if and only if H is a normal subgroup of
G.

PROOF : Suppose that (aH)(bH) = (ab)H gives a well-defined binary operation on left
cosets. Let a ∈ G. Now, let x ∈ aH.
Choosing representatives x ∈ aH and a−1 ∈ a−1 H, we have

(xH)(a−1 H) = (xa−1 )H.

Choosing representatives a ∈ aH and a−1 ∈ a−1 H, we have

(aH)(a−1 H) = eH = H.

Since the binary operation is well-defined, we get (xa−1 )H = H. This implies that
xa−1 ∈ H, so xa−1 = h for some h ∈ H. This yields x = ha, from which we get x ∈ Ha.
Hence, aH ⊆ Ha. Similarly, Ha ⊆ aH. Thus, Ha = aH, implying that H / G.
Conversely, suppose that H / G.
Choosing a ∈ aH and b ∈ bH, we have

(aH)(bH) = (ab)H.

Choosing ah1 ∈ aH and bh2 ∈ bH, where h1 , h2 ∈ H, we have

(ah1 H)(bh2 H) = (ah1 bh2 )H.

Now, h1 b ∈ Hb = bH and so h1 b = bh3 for some h3 ∈ H. We have that

ah1 bh2 = abh3 h2 = (ab)(h3 h2 ).

Thus, (ab)(h3 h2 ) ∈ (ab)H. That is, ah1 bh2 ∈ (ab)H. Therefore, (ab)H = (ah1 bh2 )H
(Recall that cosets are equivalence classes). Therefore, coset multiplication is well-defined
by (aH)(bH) = (ab)H.

28
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
Corollary 3.2. Let H be a normal subgroup of G. Then, the cosets of H form a group
G/H under the binary operation (aH)(bH) = (ab)H.

PROOF : Let H / G. By Theorem 3.6, (aH)(bH) = (ab)H is a binary operation on G/H.

Let a, b, c ∈ G. Then we have that

(aH)[(bH)(cH)] = (aH)[(bc)H] = (a(bc))H

= ((ab)c)H
= [(ab)H](cH)
= [(aH)(bH)](cH).

Thus, coset multiplication is associative. From

(aH)(eH) = (ae)H = aH

and
(eH)(aH) = (ea)H = aH
we deduce that H is the identity element in G/H. Furthermore,

(a−1 H)(aH) = (a−1 a)H = eH = H

and
(aH)(a−1 H) = (aa−1 )H = eH = H
show that (aH)−1 = a−1 H. Therefore, G/H is a group under coset multiplication defined
by (aH)(bH) = (ab)H.

Definition 3.4. (Quotient Group or Factor Group) The group G/H is called the
quotient group or factor group of G by H.

Note : |G/H| = [G : H]

29
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
Exercise Three

1. Let H be a subgroup of a group G. Let the relation ∼R be defined on G by

a ∼R b ⇐⇒ ab−1 ∈ H.

Show that ∼R is an equivalence relation on G.

2. Let G be a finite group and let H be a subgroup of G. Show that the number of
left cosets of H in G is equal to the number of right cosets of H in G.

3. Find the partition of Z6 into cosets of the subgroup H = {0, 3}.

4. Let H be the subgroup hµ1 i = {ρ0 , µ1 } of S3 . Find the partitions of S3 into left
cosets of H and right cosets of H.

5. Let H and K be subgroups of a group G. Define ∼ on G by a ∼ b if and only if

a = hbk for some h ∈ H and some k ∈ K.

(a) Prove that ∼ is an equivalence relation on G.

(b) Describe the elements in the equivalence class containing a ∈ G.
(These equivalence classes are called double cosets)

6. Let H be a subgroup of a group G such that g −1 hg ∈ H for all g ∈ G and all h ∈ H.

Show that every left coset gH is the same as the right coset Hg.

7. Show that if H and K are normal subgroups of a group G such that H ∩ K = {e},
then hk = kh for all h ∈ H and k ∈ K.
[Hint : Consider hkh−1 k −1 = (hkh−1 ) k −1 = h (kh−1 k −1 )].

8. Let G and H be groups. Show that if f : G −→ H is a group homomorphism and

S C H, then T = {x ∈ G | f (x) ∈ S} is a normal subgroup of G.

9. Show that if H and N are subgroups of a group G, and N is normal in G, then

H ∩ N is normal in H.

10. Find the order of the given factor group.

(a) Z6 / h3i
(b) (Z4 × Z12 ) / (h2i × h2i)
(c) (Z4 × Z2 ) / h(2, 1)i

11. Find the order of

30
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups
(a) 5 + h4i in Z12 / h4i
(b) 26 + h12i in Z60 / h12i
(c) (2, 1) + h(1, 1)i in (Z3 × Z6 ) / h(1, 1)i

31
Benedict Vasco Normenyo
Chapter 3. Cosets and Normal Subgroups

32
Benedict Vasco Normenyo
Chapter 4
Homomorphisms

Definition 4.1. (Homomorphism) Let (G, ⊗), (G0 , ⊕) be groups and let φ : G −→ G0
be a function. Then φ is called a homomorphism if ∀ a, b ∈ G, we have

φ(a ⊗ b) = φ(a) ⊕ φ(b).

Example 4.1. Let G = (R, +) and H = (R+ , ·). Let φ : G −→ H be defined by

φ(x) = ex . Show that φ is a homomorphism.

SOLUTION : Let G = (R, +) and H = (R+ , ·). Let φ : G −→ H be defined by φ(x) = ex .

Now, let x, y ∈ G. Then we obtain

φ(x + y) = ex+y = ex ey = φ(x) · φ(y).

Hence φ is a homomorphism.

Example 4.2. Let F be the additive group of all functions mapping R into R, let R be
the additive group of real numbers and let c ∈ R. Define the function φc : F −→ R by
φc (f ) = f (c) for all f ∈ F . Show that φc is a homomorphism.

SOLUTION : Now let g, h ∈ F .

φc (g + h) = (g + h)(c)
= g(c) + h(c)
= φc (g) + φc (h)

Therefore φc is a homomorphism.

Definition 4.2. (Isomorphism) Let φ : G −→ H be a homomorphism, where G, H are

groups. Then φ is called an isomorphism if it is a one-to-one and onto function. G and
H are said to be isomorphic, denoted by G ∼
= H.

Definition 4.3. A 1-1 homomorphism is called a monomorphism and an onto

homomorphism is called an epimorphism. A homomorphism from a group G to itself
is called an endomorphism. An isomorphism from a group G onto itself is called an
Automorphism.

33
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
Theorem 4.1. Let G, H, and K be groups and let φ : G −→ H and ψ : H −→ K be
functions.
(i) If φ and ψ are homomorphisms, then ψ ◦ φ : G −→ K is a homomorphism.
(ii) If φ and ψ are isomorphisms, then ψ ◦ φ is also an isomorphism.
(iii) If φ is an isomorphism, then φ−1 : H −→ G is an isomorphism.

PROOF : Let G, H, and K be groups and let φ : G −→ H and ψ : H −→ K be functions.

(i) Suppose that φ and ψ are homomorphisms. Let a, b ∈ G.

(ψ ◦ φ)(ab) = ψ[φ(ab)]
= ψ[φ(a)φ(b)]
= ψ[φ(a)]ψ[φ(b)]
= (ψ ◦ φ)(a)(ψ ◦ φ)(b)
Therefore, ψ ◦ φ is a homomorphism.
(ii) Suppose that φ and ψ are isomorphisms. From (i), ψ ◦ φ is a homomorphism. Let
k ∈ K. Then ∃ h ∈ H such that k = ψ(h), since ψ is onto. Also, ∃ g ∈ G such that
h = φ(g), since φ is onto. Now, (ψ ◦ φ)(g) = ψ[φ(g)] = ψ(h) = k. That is, for all k ∈ K,
there exists g ∈ G such that (ψ ◦ φ)(g) = k. Hence ψ ◦ φ is onto.
Now, let x, y ∈ G such that (ψ ◦ φ)(x) = (ψ ◦ φ)(y). Then,

ψ[φ(x)] = ψ[φ(y)]
φ(x) = φ(y), since ψ is one-to-one
x = y, since φ is one-to-one.
Hence, ψ ◦ φ is one-to-one. Therefore, ψ ◦ φ is an isomorphism.
(iii) Suppose that φ is an isomorphism. We know that if φ is 1-1 and onto, then so is
φ−1 . Now,

φ[φ−1 (xy)] = xy
= φ[φ−1 (x)]φ[φ−1 (y)]
= φ[φ−1 (x)φ−1 (y)], since φ is a homomorphism .
Thus φ[φ−1 (xy)] = φ[φ−1 (x)φ−1 (y)]. Since φ is 1-1, φ−1 (xy) = φ−1 (x)φ−1 (y). So φ−1 is a
homomorphism. Therefore, φ−1 is an isomorphism.

Definition 4.4. Let φ be a mapping of a set X into a set Y and let A ⊆ X and B ⊆ Y .
The image of A in Y under φ is the set φ[A] = {φ(a)|a ∈ A}. The set φ[X] is the range
of φ. The inverse image φ−1 [B] of B in X is {x ∈ X | φ(x) ∈ B}.

Theorem 4.2. Let φ be a homomorphism of a group G into a group G0 .

(1) If e is the identity element in G, then φ(e) is the identity element e0 in G0 . That is,
φ(e) = e0 .
(2) ∀ a ∈ G and ∀ n ∈ Z, φ(an ) = [φ(a)]n . In particular, φ(a−1 ) = [φ(a)]−1 .
(3) If H is a subgroup of G, then φ[H] is a subgroup of G0
(4) If K 0 is a subgroup of G0 , then φ−1 [K 0 ] is a subgroup of G.

PROOF : Let G and G0 be groups. Let φ : G → G0 be a homomorphism.

34
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
(1) Let e ∈ G and e0 ∈ G0 be the identity elements. Now, φ(a) = φ(ae) = φ(a)φ(e) where
a ∈ G. Thus, we have that

(φ(a))−1 φ(a) = (φ(a))−1 φ(a)φ(e)

e0 = e0 φ(e)

Hence φ(e) = e0 .
(2) Let P (n) : φ(an ) = [φ(a)]n , ∀ a ∈ G and ∀ n ∈ Z. Let n = 0. Then φ(a0 ) = [φ(a)]0 ,
since from (i) φ(e) = e0 . Let n = 1. Then clearly, φ(a1 ) = [φ(a)]1 is true. Now let k ≥ 2
and suppose that P (k) is true, that is, φ(ak ) = [φ(a)]k ∀ a ∈ G.

φ(ak+1 ) = φ(ak a)
= φ(ak )φ(a)
= [φ(a)]k φ(a), by the inductive hypothesis
= [φ(a)]k+1

Thus P (k + 1) is true. Hence P (n) is true for all n ≥ 0. Now, let n = −m, with m > 0.
Since an am = e we get φ(an )φ(am ) = e0 , which implies that φ(an ) = [φ(am )]−1 . Thus,

φ(an ) = [(φ(a))m ]−1 , since m > 0,

= [φ(a)]−m
= [φ(a)]n

Therefore, the statement P (n) is true for all n ∈ Z. In particular, φ(a−1 ) = [φ(a)]−1
whenever n = 1.
(3) Let H be a subgroup of G. Since e ∈ H, φ(e) = e0 ∈ φ[H]. So φ[H] is not empty. Let
x, y ∈ φ[H]. Then x = φ(h1 ) and y = φ(h2 ) for some h1 , h2 ∈ H. Now,

xy = φ(h1 )φ(h2 )
= φ(h1 h2 ), since φ is a homomorphism
= φ(h), where h = h1 h2 ∈ H

Hence xy ∈ φ(H). We also have that

x−1 = [φ(h1 )]−1 = φ(h−1

1 ).

Since h−1
1 ∈ H, we see that x
−1
∈ φ[H]. Therefore, φ[H] ≤ G0 .
(4) Let K 0 be a subgroup of G0 . Since e ∈ G and φ(e) = e0 ∈ K 0 , we have that e ∈ φ−1 [K 0 ].
Thus, φ−1 [K 0 ] is not empty. Let a, b ∈ φ−1 [K 0 ]. Then a, b ∈ G such that φ(a), φ(b) ∈ K 0 .
Since K 0 is a subgroup of G0 , φ(a)φ(b) = φ(ab) ∈ K 0 . Hence ab ∈ φ−1 [K 0 ]. Also,
φ(a) ∈ K 0 implies that φ(a)−1 ∈ K 0 , K 0 being a subgroup. Thus φ(a−1 ) ∈ K 0 , meaning
a−1 ∈ φ−1 (K 0 ). Therefore, φ−1 [K 0 ] is a subgroup of G.
Definition 4.5. Let φ : G −→ G0 be a homomorphism of groups. The subgroup
φ−1 [{e0 }] = {x ∈ G|φ(x) = e0 } is the kernel of φ, denoted by Ker(φ).
0
Theorem 4.3. Let φ : G −→ G be a group homomorphism and let H = Ker(φ).
Let a ∈ G. Then the set φ−1 [{φ(a)}] = {x ∈ G|φ(x) = φ(a)} is the left coset aH of H
and is also the right coset Ha of H.

35
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
0
PROOF : Let φ : G −→ G be a group homomorphism and let H = Ker(φ). Let a ∈ G.
Now let x ∈ φ−1 [{φ(a)}]. Then φ(x) = φ(a), which leads to
0
φ(a)−1 φ(x) = e
0
φ(a−1 )φ(x) = e
0
φ(a−1 x) = e .

This means that a−1 x ∈ H, and so a−1 x = h, for some h ∈ H. Thus x = ah ∈ aH, which
tells us that φ−1 [{φ(a)}] ⊆ aH. Next, let y ∈ aH. Then y = ah for some h ∈ H. So,
0
φ(y) = φ(ah) = φ(a)φ(h) = φ(a)e = φ(a),

meaning that y ∈ {x ∈ G|φ(x) = φ(a)}. We see that aH ⊆ φ−1 [{φ(a)}]. Hence

φ−1 [{φ(a)}] = aH. Similarly, we can show that φ−1 [{φ(a)}] = Ha. [This is left as
an exercise.]
0
Corollary 4.1. If φ : G −→ G is a group homomorphism, then Ker(φ) is a normal
subgroup of G.

Theorem 4.4. A group homomorphism φ : G −→ G0 is a 1-1 map iff Ker(φ)= {e}.

0
PROOF : Let φ : G → G be a homomorphism. Suppose that Ker(φ) = {e}. Let a ∈ G.
Applying Theorem 4.3, we find that {x ∈ G|φ(x) = φ(a)} = a{e} = {a}. Hence φ is
1 − 1. Conversely, suppose that φ is 1-1. Then we get

Ker(φ) = {x ∈ G | φ(x) = e0 }
= {x ∈ G | φ(x) = φ(e)}
= {x ∈ G | x = e} , since φ is one − to − one
= {e}

0
Theorem 4.5. Let φ : G −→ G be a group homomorphism with kernel H. Then the
left cosets of H form a factor group, G/H where (aH)(bH) = (ab)H. Also, the map
µ : G/H → φ[G] defined by µ(aH) = φ(a) is an isomorphism.
0
PROOF : Let φ : G −→ G be a group homomorphism with kernel H. From Corollary
4.1, H / G. By Corollary 3.2, the left cosets of H form a factor group G/H where
(aH)(bH) = (ab)H. Define µ : G/H → φ[G] by µ(aH) = φ(a). We have that

µ[(aH)(bH)] = µ[(ab)H]
= φ(ab)
= φ(a)φ(b), since φ is a homomorphism,
= µ(aH)µ(bH),

which implies that µ is a homomorphism. For all φ(a) ∈ φ[G], there exists aH ∈ G/H
such that µ(aH) = φ(a). As a result, we find that µ is onto.

36
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
Next, we have that
0
Ker(µ) = {aH ∈ G/H | µ(aH) = e ∈ φ[G]}
0
= {aH ∈ G/H | φ(a) = e }
= {aH ∈ G/H | a ∈ H}, since Ker(φ) = H
= {H},

which means that µ is 1 − 1. Therefore µ is an isomorphism.

Theorem 4.6. Let H be a normal subgroup of G. Then γ : G −→ G/H given by
γ(x) = xH is a homomorphism with kernel H.
PROOF : Let H be a normal subgroup of G. Let γ : G −→ G/H be given by γ(x) = xH.
Let x, y ∈ G. Then
γ(xy) = (xy)H = (xH)(yH) = γ(x)γ(y).
Hence γ is a homomorphism. We also have that

Ker(γ) = {x ∈ G|γ(x) = H}
= {x ∈ G|xH = H}
= {x ∈ G|x ∈ H}
= {x ∈ H}
= H

0
Theorem 4.7. (The Fundamental Homomorphism Theorem) Let φ : G −→ G
be a group homomorphism with kernel H. Then φ[G] is a group and µ : G/H −→ φ[G]
given by µ(gH) = φ(g) is an isomorphism. If γ : G −→ G/H is the homomorphism given
by γ(g) = gH, then φ(g) = (µ ◦ γ)(g) ∀ g ∈ G.

φ
G φ[G]

γ µ

G/H

0
Definition 4.6. (Inner Automorphism) The automorphism ig : G −→ G , where
ig (x) = gxg −1 for all x ∈ G is the inner automorphism of G by g. Performing ig on x is
called conjugation of x by g.

Remark 4.1. We have that

gH = Hg ⇐⇒ gHg −1 = H ⇐⇒ ig [H] = H.

Hence the normal subgroup of a group G are those that are invariant under all inner
automorphisms.

37
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
Definition 4.7. Let H < G and let N (H) = {g ∈ G|gHg −1 = H}. Then N (H) is called
the normalizer of H.

Definition 4.8. (Simple Group) A group is simple if it is nontrivial and has no proper
nontrivial normal subgroups.

Remark 4.2. Definition 4.8 says that a group G 6= {e} is simple if and only if the only
normal subgroups of G are {e} and G.

Definition 4.9. (Maximal Normal Subgroup) A maximal normal subgroup of a

group G is a normal subgroup M 6= G such that there is no proper normal subgroup N
of G properly containing M .

Theorem 4.8. M is a maximal normal subgroups of G if and only if G/M is simple.

PROOF : Let M be a maximal normal subgroup of G. Suppose that G/M is not simple.
Then there exists a nontrivial proper normal subgroup of G/M , say K (That is, K 6=
{M }, K 6= G/M and K / G/M ). From the homomorphism γ : G −→ G/M , γ −1 [K] is
a proper normal subgroup of G properly containing M (Recall that M is the kernel of
the homomorphism γ. We also make use of the result in question 6 of Exercise 4). This
is a contradiction since M is maximal. Hence G/M is simple. Conversely, suppose that
G/M is simple. Suppose that M is not a maximal normal subgroup of G. Then there
exists a proper nontrivial normal subgroup N of G properly containing M . It follows
that γ[N ] 6= G/M, γ[N ] 6= {M }, and so γ[N ] is a nontrivial proper normal subgroup of
G/M . This is a contradiction since G/M is simple. Hence M is maximal.

38
Benedict Vasco Normenyo
Chapter 4. Homomorphisms
Exercise Four

1. Let Sn be the symmetric group on n letters and let φ : Sn −→ Z2 be defined by


0 if σ is an even permutation
φ(σ) =
1 if σ is an odd permutation

Show that φ is a homomorphism. [Hint : Consider four different cases]

2. Let G and G0 be groups. Let φ : G −→ G0 be defined by φ(g) = e0 for all g ∈ G,

where e0 is the identity element in G0 . Show that φ is a homomorphism. (This is
called the trivial homomorphism)
0
3. Let φ : G −→ G be a group homomorphism and let H = Ker(φ). Let a ∈ G. Show
that the set
φ−1 [{φ(a)}] = {x ∈ G|φ(x) = φ(a)}
is the right coset Ha of H.

4. Let G be a group. Show that

(a) the commutator subgroup C of G is a normal subgroup of G.

(b) if N / G, then G/N is Abelian if and only if C ≤ N .

5. Let G be a group. Show that the map

0
ig : G −→ G defined by ig (x) = gxg −1 ∀ x ∈ G

is an automorphism.
0
6. Let φ : G −→ G be a group homomorphism.

(a) Show that if N / G, then φ[N ] / φ[G].

0 0
(b) Show that if N / φ[G], then φ−1 [N ] / G.

7. Consider the group (R, +) and the subgroup

H = {2nπ | n ∈ Z}.

Let G denote the group of matrices of the form

 
cos θ − sin θ
sin θ cos θ

where θ is any real number and the group operation is matrix multiplication. Prove
that R\H is isomorphic to G by first giving and explicit formula for a map φ :
R\H −→ G, and then checking that your map is well-defined, bijective and respects
the group operation.

39
Benedict Vasco Normenyo
Chapter 4. Homomorphisms

40
Benedict Vasco Normenyo
Chapter 5
Direct Products

Definition 5.1. The Cartesian product of sets S1 , S2 , ..., Sn is the set of all ordered n-
Q ai ∈ Si for i = 1, 2, ..., n. The Cartesian product is denoted
tuples (a1 , a2 , ..., an ), where
by S1 × S2 × · · · × Sn or ni=1 Si .

Theorem
Qn 5.1. Let G1 , G2 , ..., Gn be groups. For (a1 , a2 , ..., an ) and (b1 , b2 , ..., Qbnn ) in
i=1 Gi , define (a1 , ..., an )(b1 , ..., bn ) to be the element (a1 b1 , ..., an bn ). Then, i=1 Gi
is a group under this binary operation. [The proof is left as exercise.]

Definition 5.2. (External Direct Product) Let G1 , G2 , ..., Gn be groups. For

(a1 , a2 , ..., an ), (b1 , b2 , ..., bn ) ∈ G1 × G2 × ... × Gn ,

let
(a1 , ..., an )(b1 , ..., bn ) = (a1 b1 , ..., an bn ).
The group G1 ×G2 ×...×Gn is called the external direct product of the groups G1 , G2 , ..., Gn .

Definition 5.3. (Internal Direct Product) A group G is the internal direct product
of its subgroups G1 × G2 × ... × Gn if the function

φ : G1 × G2 × ... × Gn −→ G

defined by letting
φ(g1 , g2 , ..., gn ) = g1 g2 · · · gn
is an isomorphism. We write G = G1  G2  ...  Gn .

Theorem 5.2. Let G = G1 × G2 × ... × Gn .

(i) If gi ∈ Gi for 1 ≤ i ≤ n and each gi has finite order, then o((g1 , g2 , ..., gn )) is the
least common multiple of o(g1 ), o(g2 ), ..., o(gn ).

(ii) If each Gi is a cyclic group of finite order, then G is cyclic if and only if |Gi | and
|Gj | are relatively prime for i 6= j.
PROOF : Let G = G1 × G2 × · · · × Gn .
(i) Let gi ∈ Gi for 1 ≤ i ≤ n. Let o(gi ) = ki for 1 ≤ i ≤ n. Let m ∈ Z+ . Then,

(g1 , g2 , ..., gn )m = (g1m , g2m , ..., gnm ).

41
Benedict Vasco Normenyo
Chapter 5. Direct Products
Suppose that
(g1 , g2 , ..., gn )m = (e1 , e2 , ..., en ),
where ei is the identity element in Gi . Then

g1m = e1 , g2m = e2 , ..., gnm = en .

Applying part 2 of Theorem 1.1 leads us to

k1 |m, k2 |m, ..., kn |m,

which means that m is a common multiple of k1 , k2 , ..., kn . Conversely, suppose that m

is a common multiple of k1 , k2 , ..., kn . Then, m = t1 k1 , m = t2 k2 , ..., m = tn kn for some
ti ∈ Z+ , 1 ≤ i ≤ n. It follows that

(g1 , g2 , ..., gn )m = (g1t1 k1 , g2t2 k2 , ..., gntn kn )

= (et11 , et22 , ..., etnn )
= (e1 , e2 , ..., en ).

Therefore (g1 , g2 , ..., gn )m = (e1 , e2 , ..., en ) if and only if m is a common multiple of

k1 , k2 , ..., kn . Thus,

o((g1 , g2 , ..., gn )) = lcm(k1 , k2 , ..., kn )

= lcm(o(g1 ), o(g2 ), ..., o(gn ))

(ii) Let Gi be a cyclic group such that |Gi | is finite for 1 ≤ i ≤ n. Now, let G be cyclic
with G = hgi, where g = (g1 , g2 , ..., gn ). Then, Gi = hgi i, since gi 6= ei ∈ Gi and Gi is
cyclic. Thus, |Gi | = o(gi ). By part (i),

o(g) = lcm(|G1 |, |G2 |, ..., |Gn |)

Since g generates G, we get

o(g) = |G| = |G1 ||G2 | · · · |Gn |.

This leads to
lcm(|G1 |, |G2 |, ..., |Gn |) = |G1 ||G2 | · · · |Gn |,
which means that gcd(|Gi |, |Gj |) = 1 for i 6= j. Conversely, suppose that |Gi | and |Gj |
are relatively prime for i 6= j. Then,

lcm(|G1 |, |G2 |, ..., |Gn |) = |G1 ||G2 | · · · |Gn |.

Let Gi = hgi i. Then, o(gi ) = |Gi |. It follows that

o((g1 , g2 , ..., gn )) = lcm(o(g1 ), o(g2 ), ..., o(gn ))

= lcm(|G1 |, |G2 |, ..., |Gn |)
= |G1 ||G2 | · · · |Gn |
= |G|.

42
Benedict Vasco Normenyo
Chapter 5. Direct Products
Thus, g = (g1 , g2 , ..., gn ) generates G, and hence, G is cyclic.

Example 5.1. Consider Z2 × Z3 = {(0, 0), (0, 1), (0, 2), (1, 0), (1, 1), (1, 2)}. It can be
seen that |Z2 × Z3 | = 2 · 3 = 6 and gcd(2, 3) = 1

1(1, 1) = (1, 1)
2(1, 1) = (1, 1) + (1, 1) = (0, 2)
3(1, 1) = 2(1, 1) + (1, 1) = (0, 2) + (1, 1) = (1, 0)
4(1, 1) = (1, 0) + (1, 1) = (0, 1)
5(1, 1) = (0, 1) + (1, 1) = (1, 2)
6(1, 1) = (1, 2) + (1, 1) = (0, 0)

Thus, (1, 1) is a generator of Z2 × Z3 , and so Z2 × Z3 is cyclic.

Example 5.2. Consider Z2 × Z2 = {(0, 0), (0, 1), (1, 0), (1, 1)}. We can see that
|Z2 × Z2 | = 2 · 2 = 4 and gcd(2, 2) = 2 6= 1. As shown below, each non-identity element
has order 2. Consequently, there is no element of order 4 to generate the group. Hence,
Z2 × Z2 is not cyclic.

(0, 1) + (0, 1) = (0, 0)

(1, 0) + (1, 0) = (0, 0)
(1, 1) + (1, 1) = (0, 0)

Example 5.3. Consider G = Z12 × Z60 × Z24 . Is G cyclic? What is the order of (8,4,10)
in G?
SOLUTION : We have that gcd(12, 24) = 12 6= 1. Hence, G is not cyclic. We now employ
the part 3 of Theorem 1.1. We obtain the following.

8 ∈ Z12 = h1i, 8 = 8(1), gcd(8, 12) = 4, o(8) = 12/4 = 3

4 ∈ Z60 = h1i, 4 = 4(1), gcd(4, 60) = 4, o(4) = 60/4 = 15
10 ∈ Z24 = h1i, 10 = 10(1), gcd(10, 24) = 2, o(10) = 24/2 = 12

We have that lcm(3, 15, 12) = 60, and so o(8, 4, 10) in G is 60.

43
Benedict Vasco Normenyo
Chapter 5. Direct Products
Exercise Five

) in ni=1 Gi , define
Q
1. Let G1 , G2 , ..., Gn be groups. For (a1 , a2 , ..., an ) and (b1 , b2 , ..., bnQ
(a1 , ..., an )(b1 , ..., bn ) to be the element (a1 b1 , ..., an bn ). Then, ni=1 Gi is a group
under this binary operation.

2. Prove that the group Zm × Zn is cyclic and isomorphic to Zmn if and only if m and
n are relatively prime. [Hint : Follow the proof of part (ii) of Theorem 5.2]

3. Let H and K be subgroups of a group G. Let the map

f : H × K −→ G

be defined by
f (h, k) = hk,
for all h ∈ H and for all k ∈ K. Show that

(a) f [H × K] = HK.
(b) f is one-to-one if and only if H ∩ K = {e}.
(c) f is a homomorphism of groups if and only if hk = kh for all h ∈ H, k ∈ K.

4. Let H and K be subgroups of a group G. Prove that G = H  K if and only if the

following conditions hold.

(a) HK = G.
(b) H ∩ K = {e}.
(c) H and K are normal subgroups of G.

[Hint : You may use the results in Question 3.]

44
Benedict Vasco Normenyo
Chapter 6
Action of a Group on a Set

Definition 6.1. (Group Action on a Set) Let G be a group and X a set. Then G is
said to act on X if there is a mapping ∗ : G × X −→ X such that

(1) e ∗ x = x, for all x ∈ X,

(2) (g1 g2 ) ∗ x = g1 ∗ (g2 ∗ x), for all x ∈ X and for all g1 , g2 ∈ G.

The mapping ∗ is called an action of G on X and X is called a G-set.

Example 6.1. Let G be a group and H / G. Define

∗ : G × G/H −→ G/H

by
a ∗ xH = axa−1 H,
for all a ∈ G and for all xH ∈ G/H. Show that G/H is a G-set.

SOLUTION : Let x ∈ G. Then we have that

e ∗ xH = exe−1 H = xH.

Now let a, b ∈ G. Then we obtain

(ab) ∗ xH = abx(ab)−1 H
= abxb−1 a−1 H
= a ∗ bxb−1 H
= a ∗ (b ∗ xH)

Therefore ∗ is an action of G on G/H and so G/H is a G-set

Example 6.2. Let G be a group. Define

∗ : G × G −→ G

by
a ∗ x = axa−1 ,
for all a, x ∈ G. Show that ∗ is an action of G on itself.

45
Benedict Vasco Normenyo
Chapter 6. Action of a Group on a Set
SOLUTION : Let x ∈ G. Then we have e ∗ x = exe−1 = x. Let a, b ∈ G.

(ab) ∗ x = abx(ab)−1
= abxb−1 a−1
= a ∗ bxb−1
= a ∗ (b ∗ x)

Hence G is a G-set. This action of G on itself is called conjugation.

Example 6.3. Let G be the group D4 = {ρ0 , ρ1 , ρ2 , ρ3 , µ1 , µ2 , δ1 , δ2 } of symmetries of the

square, where
π
ρi corresponds to rotating the square counterclockwise through i radian,
2
µi corresponds to flipping on the axis mi and
δi corresponds to flipping on the diagonal di .

P3 S3
4 • 3

S4 d1 d2

P4 • •C • P2
m2

m1 S2

1 • 2
S1 P1

We label the vertices 1, 2, 3, 4; the sides S1 , S2 , S3 , S4 ; the diagonals d1 and d2 ; vertical

and horizontal axes m1 and m2 ; the center C and midpoint Pi of the sides Si .
Let X = {1, 2, 3, 4, S1 , S2 , S3 , S4 , m1 , m2 , d1 , d2 , C, P1 , P2 , P3 , P4 }. Then X is a D4 -set and
the table below describes the action of D4 on X.

1 2 3 4 s1 s2 s3 s4 m1 m2 d1 d2 C P1 P2 P3 P4
ρ0 1 2 3 4 s1 s2 s3 s4 m1 m2 d1 d2 C P1 P2 P3 P4
ρ1 2 3 4 1 s2 s3 s4 s1 m2 m1 d2 d1 C P2 P3 P4 P1
ρ2 3 4 1 2 s3 s4 s1 s2 m1 m2 d1 d2 C P3 P4 P1 P2
ρ3 4 1 2 3 s4 s1 s2 s3 m2 m1 d2 d1 C P4 P1 P2 P3
µ1 2 1 4 3 s1 s4 s3 s2 m1 m2 d2 d1 C P1 P4 P3 P2
µ2 4 3 2 1 s3 s2 s1 s4 m1 m2 d2 d1 C P3 P2 P1 P4
δ1 3 2 1 4 s2 s1 s4 s3 m2 m1 d1 d2 C P2 P1 P4 P3
δ2 1 4 3 2 s4 s3 s2 s1 m2 m1 d1 d2 C P4 P3 P2 P1

Theorem 6.1. Let X be a G-set. For each g ∈ G, the function σg : X −→ X defined by

σg (x) = gx for x ∈ X is a permutation of X. The map φ : G −→ SX defined by φ(g) = σg is a
homomorphism with the property that φ(g)(x) = gx.

PROOF : Let X be a G-set. For each g ∈ G, define σg : X −→ X by σg (x) = gx ∀x ∈ X.

46
Benedict Vasco Normenyo
Chapter 6. Action of a Group on a Set
Suppose that σg (x1 ) = σg (x2 ) for x1 , x2 ∈ X. Then we obtain

gx1 = gx2
−1
g (gx1 ) = g −1 (gx2 )
(g −1 g)x1 = (g −1 g)x2 , using condition 2 in Definition 6.1
ex1 = ex2
x1 = x2 , using condition 2 in Definition 6.1

Hence σg is one-to-one. Let x ∈ X(codomain). Choose g −1 x ∈ X(domain). Then

σg (g −1 x) = g(g −1 x) = (gg −1 )x = ex = x.

Hence σg is onto. We conclude that σg is a permutation . Now, define φ : G −→ SX by

φ(g) = σg Let g1 , g2 ∈ G. We get the following.

φ(g1 g2 )(x) = σg1 g2 (x) = (g1 g2 )x = g1 (g2 x)

= g1 σg2 (x) = σg1 (σg2 (x))
= (σg1 ◦ σg2 )(x) = (σg1 σg2 )(x)
= (φ(g1 )φ(g2 ))(x)

Hence φ(g1 g2 ) = φ(g1 )φ(g2 ) and so φ is a homomorphism. Clearly, φ(g)(x) = σg (x) = gx.

Definition 6.2. Let X be a G-set. If the identity element of G is the only element that leaves
every x ∈ X fixed , then G is said to act faithfully on X.

Notation : Let X be a G-set. We let Xg = {x ∈ X | gx = x} and Gx {g ∈ G | gx = x}.

Example 6.4. Consider the D4 -set in Example 6.3. Compute Xρ0 , Xρ1 , Xµ1 and with G = D4 ,
compute G1 , GS3 , Gd1

SOLUTION : We get Xρ0 = X, Xρ1 = {C}, Xµ1 = {S1 , S3 , m1 , m2 , C, P1 , P3 }, G1 = {ρ0 , δ2 },

GS3 = {ρ0 , µ1 }, Gd1 = {ρ0 , ρ2 , δ1 , δ2 }

Theorem 6.2. Let X be a G-set. Then Gx is a subgroup of G for each x ∈ X

PROOF : Let X be a G-set. Let x ∈ X and g1 , g2 ∈ Gx . Since ex = x, we have that e ∈ Gx ,

and so Gx is not empty. We also have that

(g1 g2 )x = g1 (g2 x) = g1 x = x,

so g1 g2 ∈ Gx . Let g ∈ Gx then gx = x. It follows that

g −1 x = g −1 (gx) = (g −1 g)x = ex = x,

so g −1 ∈ Gx . Thus Gx is a subgroup of G.

Definition 6.3. (Isotropy Subgroup) Let X be a G-set and let x ∈ X. The subgroup Gx is
the isotropy (stabilizer) subgroup of x in G.

Theorem 6.3. Let X be a G-set. For all x1 , x2 ∈ X, let x1 ∼ x2 if and only if there exists
g ∈ G such that gx1 = x2 . Then ∼ is an equivalence on X. [The proof is left as exercise. ]

47
Benedict Vasco Normenyo
Chapter 6. Action of a Group on a Set
Definition 6.4. (Orbit) Let X be a G-set. Each cell in the partition of the equivalence relation
in Theorem 6.3 is an orbit in X under G. If x ∈ X, the cell containing x is the orbit of x denoted
by Gx.

We observe that

Gx = {y ∈ X | y ∼ x}
= {y ∈ X | y = gx for some g ∈ G}
= {gx | g ∈ G}

Theorem 6.4. (The Stabiliser-Orbit Theorem) Let X be a G-set and let x ∈ X. Then
|Gx| = (G : Gx ). If |G| is finite, then |Gx| is a divisor of |G|.

PROOF : Let X be a G-set and let x ∈ X. Let x1 ∈ Gx. Then there exists g1 ∈ G such that
g1 x = x1 . Define the map
ψ : Gx −→ G/Gx
by
ψ(x1 ) = g1 Gx .
Let us suppose that g10 x = x1 for some g 0 ∈ G. Then we have
0
g1 x = g1 x
0
g1 (g1 x) = g1−1 (g1 x)
−1
0
x = (g1−1 g1 )x.
0 0 0
We deduce that g1−1 g1 ∈ Gx , so g1 ∈ g1 Gx . Hence g1 Gx = g1 Gx , and the map ψ is well-defined.
Now, let x1 , x2 ∈ Gx. Then there exists g1 , g2 ∈ G such that x1 = g1 x and x2 = g2 x. Suppose
that ψ(x1 ) = ψ(x2 ). Then g1 Gx = g2 Gx . Thus g2 ∈ g1 Gx , so that g2 = g1 g for some g ∈ Gx .
We have
x2 = g2 x = (g1 g)x = g1 (gx) = g1 x = x1 .
Hence ψ is one-to-one. Let g1 Gx ∈ G/Gx . Choose x1 ∈ Gx such that g1 x = x1 . Then
ψ(x1 ) = g1 Gx , hence ψ is onto. Therefore |Gx| = (G : Gx ).
Now suppose that G is finite. Then |G| = |Gx |(G : Gx ) = |Gx ||Gx|. Therefore |Gx| divides |G|.

We want to develop an equation that counts the number of element in a finite G-set. We
proceed as follows Let X be a finite G-set. Suppose that there are r orbits in X under G and
let {x1 , x2 , . . . , xr } contain one element from each orbit in X. Then
r
X
|X| = |Gxi | (6.1)
i=1

Let XG = {x ∈ X | gx = x for all g ∈ G}. Then XG is the union of the one-element orbits in
X. Suppose there are s one-element orbit, where 0 ≤ s ≤ r. Then |XG | = s and Equation 6.1
becomes
r
X
|X| = |XG | + |Gxi | (6.2)
i=s+1

Now, let us consider the case where X = G and the action of G on G is by conjugation. Then

48
Benedict Vasco Normenyo
Chapter 6. Action of a Group on a Set
we get
XG = {x ∈ G | gxg −1 = x for all g ∈ G}
= {x ∈ G | gx = xg for all g ∈ G}
= Z(G),
where Z(G) is the center of G. Let c = |Z(G)| and ni = |Gxi |. That is, ni is the number of
elements in the ith orbit of G under conjugation by itself. Then Equation 6.2 becomes
r
X
|G| = c + ni . (6.3)
i=c+1

Definition 6.5. (The Class Equation) Equation 6.3 is called the class equation of G. Each
orbit in G under conjugation by G is called a conjugate class in G.

Theorem 6.5. Let p be a prime. Let G be a group of order pn with n ∈ Z+ , and let X be a
finite G-set. Then |X| ≡ |XG | (mod p).

PROOF : By Theorem 6.4, |Gxi | divides |G|. Hence p divides |Gxi | for s + 1 ≤ i ≤ r. So p
divides |X| − |XG | by Equation 6.2, which implies that |X| ≡ |XG | (mod p).

49
Benedict Vasco Normenyo
Chapter 6. Action of a Group on a Set
Exercise Six

1. Let X be a G-set. For all x1 , x2 ∈ X, let x1 ∼ x2 if and only if there exists g ∈ G such
that gx1 = x2 . Show that ∼ is an equivalence on X.

50
Benedict Vasco Normenyo