You are on page 1of 174

OXFORD IB DIPLOM A PRO GRAM M E

M ATH E M ATI CS H I GH E R LE VE L :
SETS, RELATION S
AN D GROU PS
C O U R S E C O M PA N I O N

Josip Harcet
Lorraine Heinrichs
Palmira Mariz Seiler
Marlene Torres-Skoumal
3
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the Universitys objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With ofces in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan South Korea Poland Portugal
Singapore Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
 Oxford University Press 2015
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First published 2015
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
ISBN 978-0-19-830486-9
10 9 8 7 6 5 4 3 2 1
Printed in Great Britain by Bell and Bain Ltd, Glasgow
Paper used in the production of this book is a natural, recyclable product made from wood grown in sustainable forests.
The manufacturing process conforms to the environmental regulations of the country of origin.
Acknowledgements
The publisher would like to thank the following for permission to reproduce photographs:
p3: Trajano Paiva/Alamy; p11: Diego Cervo/Shutterstock; p12l-r: Eric Pradel; p14t: Eric Pradel; p14b: OUP; p47: Private
Collection/Archives Charmet/Bridgeman Images; p88: snig/Shutterstock; p126: Science Photo Library.
Course Companion denition
The IB Diploma Programme Course Companions are mindedness, the IB learner prole and the IB Diploma
resource materials designed to support students Programme core requirements, theory o knowledge,
throughout their two-year Diploma Programme course the extended essay, and creativity, activity, service
o study in a particular subject. They will help students (CAS).
gain an understanding o what is expected rom the Each book can be used in conjunction with other
study o an IB Diploma Programme subject while materials and indeed, students o the IB are required
presenting content in a way that illustrates the purpose and encouraged to draw conclusions rom a variety o
and aims o the IB. They reect the philosophy and resources. Suggestions or additional and urther
approach o the IB and encourage a deep understanding reading are given in each book and suggestions or how
o each subject by making connections to wider issues to extend research are provided.
and providing opportunities or critical thinking.
In addition, the Course Companions provide advice and
The books mirror the IB philosophy o viewing the guidance on the specic course assessment requirements
curriculum in terms o a whole-course approach; the and on academic honesty protocol. They are distinctive
use o a wide range o resources, international and authoritative without being prescriptive.

IB mission statement
The International Baccalaureate aims to develop programmes o international education and rigorous
inquiring, knowledgable and caring young people who assessment.
help to create a better and more peaceul world through These programmes encourage students across the
intercultural understanding and respect. world to become active, compassionate, and lielong
To this end the IB works with schools, governments learners who understand that other people, with their
and international organizations to develop challenging diferences, can also be right.

The IB learner Prole


The aim o all IB programmes is to develop They take responsibility or their own actions and the
internationally minded people who, recognizing their consequences that accompany them.
common humanity and shared guardianship o the Open-minded They understand and appreciate their
planet, help to create a better and more peaceul world. own cultures and personal histories, and are open to
IB learners strive to be: the perspectives, values, and traditions o other
Inquirers They develop their natural curiosity. They individuals and communities. They are accustomed to
acquire the skills necessary to conduct inquiry and seeking and evaluating a range o points o view, and
research and show independence in learning. They are willing to grow rom the experience.
actively enjoy learning and this love o learning will be
Caring They show empathy, compassion, and respect
sustained throughout their lives.
towards the needs and eelings o others. They have a
Knowledgable They explore concepts, ideas, and issues
personal commitment to service, and act to make a
that have local and global signicance. In so doing, they
positive diference to the lives o others and to the
acquire in-depth knowledge and develop understanding
environment.
across a broad and balanced range o disciplines.
Risk-takers They approach unamiliar situations and
Thinkers They exercise initiative in applying thinking
uncertainty with courage and orethought, and have
skills critically and creatively to recognize and approach
the independence o spirit to explore new roles, ideas,
complex problems, and make reasoned, ethical
and strategies. They are brave and articulate in
decisions.
deending their belies.
Communicators They understand and express ideas
and inormation condently and creatively in more Balanced They understand the importance o
than one language and in a variety o modes o intellectual, physical, and emotional balance to achieve
communication. They work efectively and willingly in personal well-being or themselves and others.
collaboration with others. Refective They give thoughtul consideration to their
Principled They act with integrity and honesty, with a own learning and experience. They are able to assess and
strong sense o airness, justice, and respect or the understand their strengths and limitations in order to
dignity o the individual, groups, and communities. support their learning and personal development.
iii
A note on academic honesty
It is o vital importance to acknowledge and What constitutes misconduct?
appropriately credit the owners o inormation Misconduct is behaviour that results in, or may
when that inormation is used in your work. Ater result in, you or any student gaining an unair
all, owners o ideas (intellectual property) have advantage in one or more assessment component.
property rights. To have an authentic piece o Misconduct includes plagiarism and collusion.
work, it must be based on your individual and
original ideas with the work o others ully Plagiarism is dened as the representation o the
acknowledged. Thereore, all assignments, written ideas or work o another person as your own.
or oral, completed or assessment must use your The ollowing are some o the ways to avoid
own language and expression. Where sources are plagiarism:
used or reerred to, whether in the orm o direct  Words and ideas o another person used to
quotation or paraphrase, such sources must be support ones arguments must be
appropriately acknowledged. acknowledged.
How do I acknowledge the  Passages that are quoted verbatim must be
work of others? enclosed within quotation marks and
acknowledged.
The way that you acknowledge that you have
 CD-ROMs, email messages, web sites on the
used the ideas o other people is through the use
Internet, and any other electronic media must
o ootnotes and bibliographies.
be treated in the same way as books and
Footnotes (placed at the bottom o a page) or journals.
endnotes (placed at the end o a document) are to  The sources o all photographs, maps,
be provided when you quote or paraphrase rom illustrations, computer programs, data, graphs,
another document, or closely summarize the audio-visual, and similar material must be
inormation provided in another document. You acknowledged i they are not your own work.
do not need to provide a ootnote or inormation
 Words o art, whether music, lm, dance,
that is part o a body o knowledge. That is,
theatre arts, or visual arts, and where the
denitions do not need to be ootnoted as they
creative use o a part o a work takes place,
are part o the assumed knowledge.
must be acknowledged.
Bibliographies should include a ormal list o Collusion is dened as supporting misconduct by
the resources that you used in your work. The another student. This includes:
listing should include all resources, including
books, magazines, newspaper articles, Internet-
 allowing your work to be copied or submitted
based resources, CDs and works o art. Formal or assessment by another student
means that you should use one o the several  duplicating work or diferent assessment
accepted orms o presentation. You must provide components and/or diploma requirements.
ull inormation as to how a reader or viewer Other forms of misconduct include any action
o your work can nd the same inormation. that gives you an unair advantage or afects the
A bibliography is compulsory in the extended results o another student. Examples include,
essay. taking unauthorized material into an examination
room, misconduct during an examination, and
alsiying a CAS record.

iv
About the book
The new syllabus or Mathematics Higher Level Questions are designed to increase in difculty,
Option: Sets is thoroughly covered in this book. strengthen analytical skills and build condence
Each chapter is divided into lesson-size sections through understanding.
with the ollowing eatures: Where appropriate the solutions to examples are
given in the style o a graphics display calculator.
Did you know? History Mathematics education is a growing, ever
changing entity. The contextual, technology
Extension Advice integrated approach enables students to become
adaptable, lielong learners.
Note: US spelling has been used, with IB style or
The Course Companion will guide you through
mathematical terms.
the latest curriculum with ull coverage o all
topics and the new internal assessment. The
emphasis is placed on the development and
improved understanding o mathematical
concepts and their real lie application as well as
prociency in problem solving and critical
thinking. The Course Companion denotes
questions that would be suitable or examination
practice and those where a GDC may be used.

About the authors


Lorraine Heinrichs has been teaching Marlene Torres-Skoumal has taught IB
mathematics or 30 years and IB mathematics or mathematics or over 30 years. During this time,
the past 1 6 years at Bonn International School. she has enjoyed various roles with the IB,
She has been the IB DP coordinator since 2002. including deputy chie examiner or HL,
During this time she has also been senior senior moderator or Internal Assessment,
moderator or HL Internal Assessment and calculator orum moderator, workshop leader,
workshop leader o the IB; she was also a and a member o several curriculum review
member o the curriculum review team. teams.
Palmira Mariz Seiler has been teaching Josip Harcet has been involved with and teaching
mathematics or over 25 years. She joined the IB the IB programme since 1 992. He has served as a
community in 2001 as a teacher at the Vienna curriculum review member, deputy chie
International School and since then has also examiner or Further Mathematics, assistant IA
worked as Internal Assessment moderator in examiner and senior examiner or Mathematics
curriculum review working groups and as a HL as well as a workshop leader since 1 998.
workshop leader and deputy chie examiner or
HL mathematics. Currently she teaches at
Colegio Anglo Colombiano in Bogota, Colombia.

v
Contents

Chapter 1 The development of Set Theory 2


Introduction The language o sets 3
1 .1 Set denitions and operations 4
Well-dened sets, equal sets and set diference 5
1 .2 Partitions and Venn diagrams 12
1 .3 Venn diagrams and set properties 14
Set properties 16
1 .4 The Cartesian product o two sets 21
1 .5 Relations 23
Equivalence relations 25
Modular Congruence 27
1 .6 Equivalence classes and partitions 32
Review exercise 42

Chapter 2 Extension of the concept of function 46


Introduction Evolution o the unction concept 47
2.1 Functions as relations 48
Equality o unctions 50
Composition o unctions 59
Inverse unctions 61
2.2 Properties o unctions 66
Identity unctions 70
2.3 Binary operations 72
Properties o binary operations 76
The identity element e 78
The inverse o an element 79
The cancellation laws 81
Review exercise 83

Chapter 3 The Universal Theory of Everything


in Mathematics 86
Introduction Group Theory 88
3.1 Groups 89
Innite groups 90
Finite groups 94
Groups o integers modulo n 98
Symmetry groups 1 00
3.2 Properties and theorems o groups and subgroups 1 05
Right and let cancellation laws or groups 1 05
Subgroups 1 08
3.3 Cyclic groups 114
Review exercise 119

vi
Chapter 4 The classifcation o groups 124
Introduction Group structures 1 26
4.1 Permutation groups 1 26
Permutations and cycle form 1 30
Properties of cycle form 1 32
4.2 Cosets and Lagranges theorem 1 35
4.3 Homomorphisms 1 39
The kernel of a homomorphism 1 42
4.4 Isomorphisms 1 44
Review exercise 1 53

Answers 156

Index 165

vii
The
1 development
of Set Theory
CHAPTER OBJECTIVES:
8.1 Finite and infnite sets; subsets; Operations on sets: union, intersection,
complement, set dierence, symmetric dierence; Venn diagrams;
De Morgans laws: distributive, associative and commutative laws or union
and intersection.
8.2 Ordered pairs: the Cartesian product o two sets; relations: equivalence
relations, equivalence classes and partitions.

Before you start


You should know how to: Skills check:
1 Given that ,  are the roots o the 1 a Given that ,  are the roots o the equation
equation z2  4z + 13 = 0, nd the value z 2  4z + 1 = 0,
o (  ) +  (   ), without solving
2 2
the quadratic equation.  1  1
nd the value o     +   .
    
Using Vietes ormulas or sum and
diference o roots:
b I  and  are the roots o 2x2 + 3x + 4 = 0,
 +  = 4,   =  3 show that the roots o the equation
 (  ) +  (   )  
8z2 + 7z + 8 = 0 are and without solving
 
=   2 +   2 either o the two given equations.
=  +   ( +  )
2 2

=  +   (( +  ) 2  2  )
= 4  ( 6  26) =  4

2 The development of Set Theory


The language of sets
In this chapter we will be looking at the basic elements o set theory.
Georg Cantor, 1 9th century German mathematician who is best
known or his creation o the language o sets, explained the notion
o a set as . . . the taking together into a whole o distinct well-defned
objects o our intuition or thought. He went on to study the relation
between sets, and to do this he associated with each set a cardinal
A cardinal number
number which would help him compare sizes, not only o nite sets
is one which
but also innite ones. Stated simply, by comparing diferent innite
denotes quantity or an
sequences Cantor discovered that there are diferent sizes o innity. amount of something.
The innite size o the set o Natural numbers, made up o discrete
elements, is smaller than the innite size o the set o real numbers,
which is continuous. The Natural numbers, Integers and Rational
numbers are all said to be countable, innite and have the same size
(cardinality). He called the size o the countable innite sets 0 whereas
the innity associated with the uncountable real numbers was 1 .
He urther made a conjecture that became known as the Continuum
Hypothesis. In his conjecture Cantor says that there is no set whose size
is between 0 and 1 . Cantor never proved this, and the Continuum
Hypothesis was the rst on the amous David Hilbert list o unsolved
problems at the turn o the 20th century. Kurt Gdel and Paul Cohen
worked extensively on this conjecture between 1 930 and 1 966. Their
work changed the ocus o mathematics in the second hal o the
20th century and opened doors to many other theories.

Chapter 1 3
The Hilbert Hotel: Hotel Infnity is a thought experiment created by the Georg Cantor is known as
German mathematician David Hilbert. When Hotel Infnity frst opened, the ounder o Set Theory.
it advertised itsel as the hotel that always has room or one more guest! His doctoral thesis was
Initially everything was fne, as there were more than enough rooms or titled, In mathematics the
anyone wanting to stay. One week, however, it was an especially busy time, art o asking questions
and an infnite number o people were staying in the hotel, so it was ull! How is more valuable than
does the hotel live up to its promise, that there is always room or one more? solving problems David
At this point, Hilbert asked his riend Cantor or help in solving his hotel Hilbert (18621943)
problem. Cantor said that each guest should move to the room number that said Cantors work was,
was twice the room number they originally occupied. This way all the odd the fnest product o
numbered rooms became available. Cantor however did warn Hilbert that mathematical genius
there were situations when it would be very difcult to fnd a way o allocating and one o the supreme
rooms. For example, suppose an infnite number o buses show up at the achievements o purely
hotel, with an infnite number o people in each bus? You might want to intellectual human
research this intriguing problem, and all its dierent aspects! activity.

1.1 Set defnitions and operations


Much o the rst part o this chapter you will have already encountered,
since sets is the basic language o most o the mathematics you have
studied, and is also included in the Prior Learning o the Higher Level
syllabus.
A set S is a collection o objects, and i x is one o these objects
we say that x is an element o S. We denote this by x  S.
For example, the subjects ofered in the IB diploma orm a set.
The number o elements in a set S is called the cardinality o the set
and we will denote it by n(S). In some books it is denoted by
card (S ) or | S| .
A nite set is one with a nite number o elements, i.e. a nite set is
one whose cardinality is a natural number. I a set has an innite
number o elements then we say that the set is innite.
The set A = {l, 3, 5, 7, 9} is nite whereas the set B = {2, 4, 6, 8, ...}
is innite.
There is exactly one set that has no elements and we call this the
empty set, denoted by  = {} .

4 The development of Set Theory


Set builder notation is a mathematical notation used to describe Set builder notation
sets, whether fnite or infnite. The ollowing examples consists of three parts
illustrate this: within curly brackets: a
variable, a vertical line
A = {l, 3, 5, 7, 9} in set builder notation becomes
(or a colon) and any
A = { x | x = 2 n  1, n   + , n  5 } restrictions on the
variable.
B = {2, 4, 6, 8, ...} in set builder notation becomes
B = { x | x = 2 n, n   + }

You have been using a number o infnite sets in your mathematical


journey so ar. Here is a list o them using the IB symbols or the sets:

The natural numbers  = { 0 , 1, 2, 3  }


The integers  = { 0 ,  1,  2 ,  3 ,  }
The positive integers  + = {1, 2, 3,  }
The negative integers 

= {  1,  2,  3,  }

p 
The rational numbers  =  p, q   , q  0  Note that Q + can also be
q  described as
p  p 
+ =  p, q    
The positive rational numbers + =  p, q   +  q 
q 

The real numbers, denoted by R, are oten represented by a number line.

2 0

The positive real numbers  + = { x | x   , x > 0}


The complex numbers  = { a + ib | a, b   , i = 1 }

Well-dened sets, equal sets and set diference

Defnition
A set S is said to be well-defned i or any given x, we can
determine i x belongs to the set.

For example, P = { n| n   + , n < 5 0, n is a prime number} is a


well-defned set because given any number n   + we can determine
whether n  P or n  P .
S o for the set P , 5  P , 1  P , 5 9  P . Although 59 is a prime number
it is greater than 50 and thereore not in P.

Chapter 1 5
The set T = {x| x  Z + , x is a prime number} is well-defned even
though it is infnite, because we know that any positive integer is
either prime or non-prime.
The set L = {numbers which are lucky} is not well-defned because
we do not know which numbers are lucky and which are not.
The defnition o a lucky number depends on the context.
Given two sets A and B, i every element in B is also an element
o A, we say that B is a subset o A and denote this by B  A.
I all the elements o B are in A and there is at least one element
in A which is not in B then we say that B is a proper subset o A,
denoted by B  A.

Defnitions
I x  B  x  A or all x  B, then B  A.
I x  B  x  A or all x  B, and there is y  A
such that y  B, then B  A.

Axiom
I a set B is a subset o A, and A is also a subset o B, then it
ollows that the two sets are equal. The converse o this is also
true, i.e. i A and B are equal sets then A is a subset o B, and
B is a subset o A.

Using set notation: B  A and A  B  A = B  is the notation used


The empty set  is a subset o any given set. We say that  is a for  if and only if .
trivial subset. Another trivial subset o any given set is the set Whenever we need
itsel. to prove a statement
containing  we need
to prove both ways,
i.e.  and .

6 The development of Set Theory


Denitions
A set containing all the elements under discussion is called the
universal set and is denoted by U.

I set S  U, then the complement o S, denoted by S, consists o


all those elements that are in U but not in S,
i.e. S  = { x  U | x  S } .

The intersection o two sets A and B, denoted by A  B, is made


up o those elements which are in both A and in B,
i.e. A  B = { x | x  A and x  B } .

Since or all x  A  B, x  A it ollows that A  B  A .


Similarly A  B  B .

The union o two sets A and B, denoted by A  B, is made up o


those elements which are either in A, in B, or in both A and B,
i.e. A  B = { x | x  A or x  B } .
I A  B =  then A and B are said to be disjoint sets.

The set consisting o those elements that are in set A but not in
set B is called the set diference B rom A denoted by A \ B,
i.e. A \ B = { x | x  A and x  B } .
In Example 4 you will fnd the proo that A \ B = A  B  .

The symmetric diference o two sets A and B is denoted by A  B


and consists o those elements which are either in A, in B, but not
in both A and B,
i.e. A  B = { x | x  A or x  B , x  A  B } = ( A \ B )  ( B \ A ) .

The ollowing example demonstrates the application o set operations


on two fnite sets.

Chapter 1 7
Example 1

Consider the sets U = { n| n   , n  65 } , A = {2, 4, 6, 8, 1 0, 1 2, 1 4, 1 6} ,


and B = {2, 4, 8, 1 6, 32, 64} .
Find:
a AB
b AB
c A\B
d AB
e (A  B )\(A  B )
Comment upon your results.
a A  B = {2, 4, 6, 8, 1 0, 1 2, 1 4, 1 6, 32, 64} List all elements that are in A or in B.
b A  B = {2, 4, 8, 1 6} List the elements that are in both A and B.
c A \ B = {6, 1 0, 1 2, 1 4} List the elements which are in A but not in B.
d A  B = {6, 1 0, 1 2, 1 4, 32, 64} List the elements that are in A or B, but not in
both A and B.
e (A  B) \ (A  B) = {6, 1 0, 1 2, 1 4, 32, 64} List the elements that are in A  B but not in
From the results of parts d and e, we see that: A  B.
A  B = (A  B) \ (A  B)

In the next example the sets are described using set builder notation.

Example 2

A = { x| x   + , x < 1 0} , B = { y| y   ,| y|  5 } , C = { z| z   , z  1 5} .
List the elements in the following sets:
a AB
b AC
c C\ B
d AB

A = {1 , 2, 3, 4, 5, 6, 7, 8, 9} List the elements of the given sets.


B = {5, 4, 3, 2, 1 , 0, 1 , 2, 3, 4, 5}
C = {0, 1 , 2, 3, 4, . . . , 1 5}

a A  B = {1 , 2, 3, 4, 5} Since all the elements of A are in C this means


b A  C = {0, 1 , 2, 3, ..., 1 5} = C that A  C.
c C \ B = {6, 7, 8, ..., 1 5}
d A  B = {5, 4, 3, 2, 1 , 0, 6, 7, 8, 9}

8 The development of Set Theory


The next example deals with subsets o a fnite set.

Example 3

Given that S = { x| x = 2 n  1, n  +
, n  4 } , list all the proper subsets o S.

S = {l, 3, 5, 7} List all the elements o S.


The proper subsets o S are: List all the proper subsets o S. Note that
{1 } , {3} , {5} , {7} , the empty set and { 1, 3, 5, 7} are not
{1 , 3} , {1 , 5} , {1 , 7} , {3, 5} , {3, 7} , {5, 7} , proper subsets o S.
{l, 3, 5} , {1 , 3, 7} , {1 , 5, 7} , {3, 5, 7}

Investigation
The power set, P(S ), o a fnite set S with n elements is the set
o all subsets o S including the empty set  and S itsel.
a Find the number o sets in the power set o S when n(S ) = 0 to 4.
b Make a conjecture about the number o sets in the power set o S.
c Check that your conjecture works or n(S ) = 5.

One method to show that two sets A and B are equal is called the
containment method, or the double inclusion method.
To show that two sets A and B are equal we need to show both
containment conditions, i.e. A  B and B  A.
The ollowing example illustrates how to use the double inclusion
method to show that two statements are equal.

Example 4

Show that A \ B = A  B 
Let x  A \ B Use the double inclusion method.
 x  A and x  B Working rom let to right.
 x  A and x  B  Denition o set diference.
 x  A  B . Denition o complement.
Thereore A \ B  A  B . Denition o intersection.

Let x  A  B  Working rom right to let.


 x  A and x  B  Denition o intersection.
 x  A and x  B Denition o complement.
 x  A\B Denition o set diference.
 A  B  A \ B
Since A \ B  A  B  and A  B   A \ B,
it ollows that A \ B = A  B .

Chapter 1 9
Example 5 illustrates how to use the double inclusion method
to show that two sets are equal.

Example 5

A = {n| n = 5k + 2, k  Z} and B = {n| n = 5k  3, k  Z}


Show that A = B.
Let x  A Use the double inclusion method.
 x = 5m + 2, m  Z.
Let m = k  1. Since m is an integer, k is also an integer.
Then x = 5(k  1) + 2 = 5k  3.
Thereore A  B.

Let x  B
 x = 5m  3, m  Z.
Let m = k + 1. Since m is an integer, k is also an integer.
Then x = 5(k + 1)  3 = 5k + 2.
Thereore B  A.
Since A  B and B  A it ollows that A = B.

Example 6 proves the conjecture suggested by the investigation on page 9.

Example 6
Prove that the power set o a fnite set S with n elements has exactly 2 n elements.
Method I
By defnition, the power set o S is the set o all
subsets o S including the empty set and S itsel.
We can count these subsets as ollows: The number
n 
o subsets containing no elements is given by  .
0 
The number o subsets containing only one element
n 
is given by  .
1 
The number o subsets containing only 2 elements
n
is given by  , etc.
2
The total number o subsets is thereore given by: Use the binomial expansion of
(1 + x) n with x = 1.
n  n  n  n 
  +   +   + +  =2 .
n

0  1  2   
n

10 The development of Set Theory


Method II
Let P(Sn) denote the power set of a set S with n
elements and let | P(Sn)| denote the order of P(Sn),
i.e. the number of elements in the power set.
Proof by induction:
Pn:| P(Sn)| = 2 n
When n = 0, S0 =  which has only one subset. Write down the statement.
 LHS = | P(S0)| = 1 Prove that the statement is true or n = 0.
RHS = 20 = 1
So P0 is true.
Assume that Pk is true for some k  0, since we have Assume that statement is true or n = k.
started with 0, i.e. | P(Sk)| = 2 k. When we add
another element to S, n = k + 1 . Then Sk+1 consists Show using assumption that the
of all those subsets that do not contain the new statement is true or n = k + 1.
element, i.e. 2k subsets, and all those other subsets
which contain it, i.e. another 2 k possible subsets.
This gives us a total of 2  2 k = 2 k+1 subsets.
Since we proved that P0 is true and we showed that Write fnal statement.
if Pk is true Pk+1 is also true it follows by the
principle of mathematical induction that
Pn:| P(Sn)|  = 2 n for all n  0.

Russells Paradox: The


development o set theory in
the early 20th century was plagued
by some thorny questions, the
most amous o these posed by
the eminent philosopher Bertrand
Russell, and known as Russells
Paradox. The problem he posed
was to fnd the set o all sets that
do not contain themselves as
members. The reason it is a
paradox is easy to see in the
well-known Barbers paradox, which
poses the question: " Suppose
there is one barber in town and he
shaves all the men in town, except or those who shave themselves. Who shaves the barber?
I he shaves himsel, then he contradicts his job description. I he doesnt shave himsel, he goes against
his mandate to shave all those men who do not shave themselves. This paradox arises because Russell
tries to fnd the set containing all sets. Such paradoxes led to a ormal axiomatic system o sets.

Chapter 1 11
Exercise 1A
1 Given that A = {a, b, c, d, e} , B = {a, e, i, o, u} and
C = {b, c, d, f, g} , list the elements o the ollowing:
a A\B b B\ A c AB
d (A  B) \ (A  C ) e A  (B  C ).

2 Use the double inclusion method to prove that:


a A  B= B A b A  B = B  A.

3 Prove that or three non-empty sets A, B and C


(C \ A )  (C \ B ) = C \ (A  B ).
4 Given that A  B and B  C, prove that A  C.

5 Prove that (A  B )\(A  B ) = A  B.

1.2 Partitions and Venn diagrams

The picture on the let above shows a collection o seashells. On the right,
the seashells have been organized by type. All the seashells rom the
let-hand picture are in the right-hand picture but each seashell belongs
to only one subset determined by its type. The seashells have been
partitioned into sets which are disjoint but together make up the whole set.

Defnition

Let A be a non-empty set.


A partition o a set A is another set P made up o non-empty
subsets o A which are disjoint and whose union makes up the

Ai = A
whole set.
means the union

i.e. P = { Ai | Ai  A, if i is not equal to j Ai  A j = , Ai = A} of all A i

For example, one partition o {l, 2, 3} would be P = {{l} , {2, 3} } .


Another partition would be P = {{l, 2} , {3} } .
In act there are only fve partitions o the set {l, 2, 3} , the other
partitions being {{l, 3} , {2} } , {{l} , {2} , {3} } and {{1 ,2,3} } .

12 The development of Set Theory


If A = {all the countries of the world} , one partition would be
P = {all the continents} provided we assume that each country
belongs to only one continent.

Example 7
Let W = {all the countries of the world} .
Determine which of the following subsets of W form a partition:
a A = {countries in Africa} b A = {any country whose name begins with a
B = {countries in N and S America} vowel}
C = {countries in Europe} B = {any country whose name contains the
D = {countries in Asia} letter a}
E = {countries in Australasia} C = {any country whose name starts with a
consonant}
a The given sets form a partition of W. The sets represent all the continents and each
country belongs to one continent only.
b The given sets do not form a partition. The sets in a partition must be disjoint.
Armenia, for example, is in both set A
and set B.

Example 8
Let S = {all subjects that can be chosen for an Extended Essay} .
Consider the sets A = {all subjects in group  } D = {all subjects in group 4}
B = {all subjects in group 2} E = {all subjects in group 5}
C = {all subjects in group 3} F = {all subjects in group 6}
Determine whether the sets A to F partition the set S.
The given sets do not partition S because the The sets in a partition must be disjoint.
subject Environmental Systems and Societies
falls into group 3 and group 4.

Exercise 1B
1 A deck of playing cards contain 52 cards. These are divided into
two red suits (hearts and diamonds) and two black suits (spades
and clubs). Each suit contains 13 cards representing the numbers
1 to 10 plus three picture cards (Jack, Queen and King). The picture
on the next page shows a deck of cards partitioned into 4 suits.
List a further two ways in which you could partition a deck of cards.

Chapter 1 13
2 Let S = {l, 2, 3, . . . , 9} . Determine whether each o the ollowing is
a partition o S.
a P = {{1, 2, 3, 9} , {4, 5, 6} , {7, 8} }

b Q = { { x | x  S , x is even} , { y | y  S , y is a multiple of 3} , {1, 5, 7} }

c B = { { x| x  S , x is a prime number}, {1, 2, 4, 6, 8, 9} }

3 Which o the ollowing collections o subsets are partitions o Z ?


a {{ x| x = 2 n , n   } , { x| x = 2 n + 1, n   }}
b {{ x| x = 4 n , n   } , { x| x = 4 n + 1, n   } , { x| x = 4 n + 2, n   } , { x| x = 4 n + 3, n   }}

c {{ x| x   , x <  5 0} , { x| x   , | x|  5 0} , { x| x   , x > 5 0}}


4 Give examples with the given properties o a partition P on the set R.
a P divides R into a fnite and an infnite set.
b P divides R into two infnite sets.
c P divides R into an infnite number o sets.

1.3 Venn diagrams and set properties

Venn diagrams are named ater the logician and philosopher John Venn.
It may well be that these types o diagram were used earlier than his
time. In act Venn diagrams are very similar to Euler diagrams which were
frst used by Leonhard Euler a century earlier.

Venn diagram stained


glass window rom
Gonville and Caius
College, Cambridge.

14 The development of Set Theory


Venn diagrams are very useul or showing relationships between
diferent sets. A Venn diagram consists o a rectangle representing
the universal set U, and circles inside the rectangle to represent the
sets under consideration. The ollowing Venn diagrams represent the
operations and relationships described above them. You should
remember that a correct Venn diagram provides an illustration o a
statement but does not constitute a ormal proo.

A is the complement o A AB


U U

A'
A A B

Disjoint sets have no intersection


AB A  B=
U U

A B A B

The diference o sets A\ B The symmetric diference A  B


U U

A B A B

As previously seen, one way o showing that two sets are equal is by
the containment or double inclusion method. Example 9 involves
using the double inclusion method, which means establishing that
i x is an element o the set on the LHS, then it is also an element o
the set on the RHS, and vice versa.
Beore you start with the ormal proo it is useul to draw a Venn diagram.
This will help you visualize what you are aiming to prove.

Chapter 1 15
Example 9

Show that ( A  B )  = A   B 

( A  B ) Draw Venn diagrams of both the left hand side


and right hand side of the equation to help
U
illustrate what you are asked to prove.

A B

A  B
U

A B

Let x  ( A  B )  Show both containment conditions.


 x A  B Start by showing that ( A  B )  A  B.
 x  A and B Since A  B is made up of elements that
are in both A and in B it follows that an
 x  A  or x  B  element which is not in this set is either not in
 x  A  B A or not in B or not in both.
Therefore ( A  B )   A   B 

Let x  A   B  Now we must show that A'  B'  ( A  B) '.


 x  A  or x  B 
 x  A and B Since x is missing from A or B or both it cannot
 x A  B be in the intersection.
 x  ( A  B )
Therefore A  B  ( A  B )
Thus we conclude that A   B  = ( A  B ) 

(A  B) = A  B is one of De Morgans Laws. The other one of De Morgans


laws states that (A  B) = A  B. The proof is left as an exercise.

Set properties
Before we move on, we need to prove some properties of sets that will
be used in the rest of the book. The following theorem concerns
properties that may seem trivial. These basic properties will be required
for proofs of less obvious results.

16 The development of Set Theory


Theorem 1
For any non-empty set A  U the ollowing statements hold:
i AA=A ii A   = 
iii A  A = U iv A  A = 
v AA=A vi A  U = U
vii A  U = A viii A   = A

Proofs:
i For all x in A  x  A and x  A, it ollows that A  A = A .
ii By the defnition o intersection we know that A  B  B .
I we let B =  then this becomes A    .
But by defnition   A   , since the empty set is a trivial subset o any set.
It thereore ollows that A   =  (double inclusion).
iii x  A  A 
 x  A  (U \ A ), by defnition o complement
 x  A or x  U \ A, by defnition o union
 x  U, by defnion o the universal set
 A  A  U
By defnition o the universal set
x  U  x  A or x  A
 x  A or x  A  , by defnition o complement
 x  A  A  , by defnition o union
 U  A  A
Since A  A   U and U  A  A  it ollows that A  A  = U .

The proos o the last fve properties are let as exercises.

You proved the next theorem in question 2 o Exercise 1 A.

Theorem 2: Commutative property


For any two sets A and B the ollowing statements are true:
i AB=BA ii AB=BA

We shall now look at a very important property o sets, namely the


associative property or intersection and union. Again this property is very
useul when proving other relations between sets.

Theorem 3: Associative property


For any three non-empty sets A, B and C, the ollowing
statements are true:
i A  ( B  C) = ( A  B)  C
ii A  ( B  C) = ( A  B)  C

Chapter 1 17
Proof of i : This proo is carried out using double inclusion.
i LHS:
x  A  (B  C)
 x  A and x  ( B  C ) , by defnition o intersection
 x  A and x  B and x  C , by defnion o intersection
 ( x  A and x  B ) and x  C
 x  A  B and x  C
 x  ( A  B)  C
 A  ( B  C)  ( A  B)  C
RHS:
x  ( A  B)  C
 x  ( A  B ) and x  C , by defnition o intersection
 x  A and x  B and x  C, by defnition o intersection
 x  A and ( x  B and x  C )
 x  A and x  ( B  C )
 x  A  (B  C)
 ( A  B)  C  A  (B  C)
Since A  ( B  C )  ( A  B )  C and
( A  B )  C  A  ( B  C ),
There are similarities between these
it ollows that A  ( B  C ) = ( A  B )  C.
properties and the associative and
The proo o ii is let as an exercise. distributive properties of addition and
multiplication of real numbers.
Another useul property when establishing urther For all a , b , c   :
relations between sets is the distribution o a+b=b+a
intersection over union and vice versa. Proo o ab = ba
the distributive law is ound in the next theorem. a + (b + c) = (a + b) + c

Theorem 4: Distributive property a ( bc ) = ( a b ) c

a ( b + c ) = a b + bc
For any three non-empty sets A, B and C the
With set operations, both intersection and
ollowing statements are true:
union behave like addition and multiplication.
i A  (B  C) = ( A  B)  ( A  C) For this reason we have two distributive
Intersection is distributive over union. properties, one for union over intersection
and one for intersection over union. This
ii A  (B  C) = ( A  B)  ( A  C)
is important for the study of algebraic
Union is distributive over intersection.
structures where the focus is on the
Proof: (Once more we shall use the double inclusion similarities (and differences) of properties of
method for part i. Part ii is included in the next exercise. ) different operations acting on different sets.

i For all x  A  ( B  C )
 x  A and x  B  C Defnition o intersection
 x  A and ( x  B or x  C ) Defnition o union
 ( x  A and x  B ) or ( x  A and x  C ) Rearranging within context
18 The development of Set Theory
 x  A  B or x  A  C
 x  ( A  B)  ( A  C)
Thereore A  ( B  C )  ( A  B )  ( A  C ).

For all x  ( A  B )  ( A  C ),
x  A  B or x  A  C. Defnition o union
 ( x  A and B ) or ( x  A and C ) Defniton o intersection
 x  A and ( x  B or x  C ) There are dierent set
theories. The one we cover
 x  A and x  B  C in the HL syllabus is Naive Set
 x  A  (B  C) Theory. This set theory is defned
inormally using natural language
Thereore (A  B)  (A  C)  A  (B  C).
and properties o Boolean
Hence by the double inclusion principle Algebra rather than the ormal
A  ( B  C ) = ( A  B)  ( A  C ). axioms o Symbolic Logic.

Sometimes it is easier to prove a statement by using set properties


than by using the double inclusion method. The previous theorems
are essential when proving complex results, especially when the
double inclusion method becomes too cumbersome. This is
illustrated in the next example.
Example 10
Given two non-empty sets A and B, show that:
a A  ( A  B ) = 
b ( A \ B )  ( B \ A ) = ( A  B )  ( A  B )

a A  ( A  B ) = A  ( A   B  ) De Morgans Law
= ( A  A )  B Associative property.
=   B With the addition o the proposed
= statements in Theorem 1, we can state
that: A  A' =  .
b RHS
= ( A  B )  ( A  B ) = ( A  B )  ( A   B  ) De Morgans Law.
= [ ( A  B )  A  ]  [ ( A  B )  B ] Distributive property.
= [(A  A)  (B  A)]  [(A  B)  (B  B)] Distributive property.
= [  (B  A)]  [(A  B)  ] Denition o intersection.
= (B  A)  (A  B) Denition o union.
= ( A  B )  ( B  A ) Commutative property.
= (A \ B)  (B \ A) Alternative orm o symmetric
= LHS diference.

Chapter 1 19
Here is a list of rules you should remember because you will need
to use them for proving more complex properties:
 A  A  = U

A  A = 
AA=A
AA=A
 A  ( A  B) = A
A  ( A  B) = A
 ( A  ) = A
  = U
U = 
 A=A
A  U= A
 A=
A  U= U
 Commutative Laws A  B = B  A
AB=BA
 Distributive Laws A  (B  C) = (A  B)  (A  C)
A  (B  C) = (A  B)  (A  C)
 Associative Laws A  ( B  C) = ( A  B)  C
A  ( B  C) = ( A  B)  C
 De Morgans Laws ( A  B ) = A  B
( A  B ) = A   B 

Exercise 1C
1 Prove that:
a (A  B)  C  A  (B  C )

b A  (B  C)  (A  B)  C

(You may frst want to draw Venn diagrams to help visualize what you are
trying to prove. )
2 Prove that ( A  B )  = A   B  .
(Hint: Use the double inclusion method used in Example 4. )
3 Prove that for all sets A, B and C :
A  (B  C) = ( A  B)  ( A  C) This is the second
part of the distributive
4 Given that A and B are subsets of a universal set U, law, i.e. union is
use De Morgans laws to prove that: distributive over
a ( A   B ) = A  B  b ( A  B )  B = U intersection.
c ( A  B )  B = 

20 The development of Set Theory


5 Use the double inclusion method shown in Example 4 to prove that
A  B = ( A \ B)  (B \ A ).

6 a Use a Venn diagram to illustrate that A  B = A   B .


b Prove this result using the double inclusion method and the
result o question 5.

7 Prove that (( A  C )  ( B  C  ))  = ( A   C )  ( B   C  )  ( A   B  )

8 Use mathematical induction to prove De Morgans laws or n sets, i.e.


a ( A1  A 2  A3   A n )  = A1  A 2  A3   A n

b ( A1  A 2  A3   A n )  = A1  A 2  A3   A n

1.4 The Cartesian product of two sets


In mathematics, a Cartesian product is a method which allows us to
construct a new set o multiple dimensions by combining multiple sets.
For example i we take the Cartesian product o the sets R  R  R
we obtain a three-dimensional set we are amiliar with, namely the
three-dimensional set o ordered triplets which was used when studying
vectors in three dimensions. In general, i we take the Cartesian product
o n sets, we obtain a representation o an n-dimensional space.
Ren Descartes rst came up with this concept when he ormulated
analytic geometry by using a Cartesian plane.

It was thanks to a common housefy that the Cartesian plane came about. Ren
Descartes, a French philosopher and mathematician, was in bed and noticed a fy
on the ceiling o his bedroom. He wondered whether he would be able to describe the
exact position o the fy to someone not in the room. Looking at a corner in the ceiling
he saw three lines and three planes which intersected at the corner. He imagined
dividing the lines into equal segments, calling the corner the origin and giving it the
value (0,0,0) and numbering the segments along each line 1, 2, 3 ...
The position o the fy in the room could then be described by three numbers.
Descartes had created a system to describe 3D space. I he used only one plane,
the ceiling, and two perpendicular lines, then the position o the fy on the ceiling
would be described by just two numbers. This was the birth o the 3D Cartesian
coordinate system as well as the Cartesian plane in 2D.

The ollowing two examples illustrate how new sets are constructed
using the Cartesian product.
I Fabienne has three blouses: plain, fowered and striped, and our pairs o
jeans: blue, red, white and green, then the total number o ways o combining
these would be the Cartesian product o the sets {blouses} and { jeans} .
B = {blouses} = { p, f, s}
J = {jeans} = {b, r, w, g}
B  J = {( p, b), ( p, r), ( p, w), (p, g), ( f, b), ( f, r), ( f, w), (f, g), (s, b), (s, r), (s, w), (s, g)}

Chapter 1 21
Note that in the set denoting the Cartesian product B  J, each pair
is ordered so that the frst item is a blouse and the second is a pair Global Positioning
o jeans. Systems calculate our
longitude and latitude in real
Another example o the Cartesian product would be coordinates time. Dierential GPS is an
used to locate positions on a globe, i.e. Latitude  Longitude. enhancement o some GPS
Valletta, the capital city o the island o Malta, would be located at units that, in addition to orbiting
(35 53 58 N, 1 4 30 52 E). satellites, uses ground stations
on the Earth to calculate a
position more accurate than
Defnition satellite-only GPS. Dierential
The Cartesian product o two non-empty sets A and B denoted by GPS can improve the accuracy
A  B is the set o all ordered pairs (a, b) where a  A and b  B. o readings rom about
50 eet to within 10 eet o the
In set-builder notation, A  B = {(a, b) | a  A, b  B} actual location.

So i A = {l, 3} and B = {2, 4, 6} ,


A  B = {(l, 2), (l, 4), (l, 6), (3, 2), (3, 4), (3, 6)}
B  A = {(2, 1 ), (2, 3), (4, 1 ), (4, 3), (6, 1 ), (6, 3)}
Clearly you can see that A  B  B  A.

With this defnition it becomes evident that the Cartesian product R  R


represents the Euclidean plane, also reerred to as the Cartesian plane.
The Cartesian product Z  Z is made up o the points on this plane whose
coordinates are integers. It is usual to denote the Cartesian product o a set A
with itsel as A 2 . So R  R = R2 and Z  Z = Z 2 . Since the Cartesian product is
a set, the number o ordered pairs in a Cartesian product is its cardinality. The
three dimensional Cartesian coordinate system, also known as Euclidean space,
is represented by R x R x R = R3 .

Example 11
The Cartesian product o two sets A and B consists o six elements.
Three o these are (a, a), (b, b) and (c, a). Find the sets:
i A
ii B
iii A  B

i a, b, c  A Since they are the frst elements in the


a, b  B three ordered pairs given.
A = {a, b, c}

ii B = {a, b} They are the second elements in the


ordered pairs.
iii A  B = {(a, a), (a, b), (b, a), (b, b), (c, a), (c, b)} Since n(A  B) = 6

22 The development of Set Theory


1.5 Relations
You should have noticed rom the previous examples the ollowing points:
1 The Cartesian product o two sets is a set.
2 The elements o the set are ordered pairs.
3 In each ordered pair, the frst element comes rom the frst set and the
second element comes rom the second set.
Now that you understand what a Cartesian product is we can move
on to appreciate how this product allows us to construct other sets.

Defnition Actually a relation


does not have to be
A relation, R, between two non-empty sets A and B is a subset o
governed by a rule.
A  B and is usually governed by a rule connecting the ordered Any random subset
pair in the relation, commonly denoted by aRb. of A  A is a relation
on A whether or not it
For example i A = {1, 2, 3, 4} and B = {1, 2, 3, 4, 5, 6, 7, 8, 9, 1 0} and describes a rule.
we defne the relation aRb  b = a2, then R = {(1 , 1 ), (2, 4), (3, 9)} .

Example 12
Given a set A, prove that a relation R on A is a subset o A  A.
Let ( a, b )  R
 aRb
 a  A and b  A
 ( a, b )  A  A Since R is a relation on A.
Thereore R  A  A.

Defnition
Let R be a relation rom set A to set B. The inverse o relation R,
denoted by R1 , is the set o ordered pairs {(b, a) | (a, b)  R} .

Example 13
b
Given that A = {1 , 2, 3, . . . ,1 0} and R  A  A such that aRb  = 2, fnd R1 .
a
R = {(1 , 2), (2, 4), (3, 6), (4, 8), (5, 1 0)} List all the elements o R.
 R1 = {(2, 1 ), (4, 2), (6, 3), (8, 4), (1 0, 5)} Use the defnition o R1 to list its elements.

In other words, bR1 a  aRb.

Chapter 1 23
Exercise 1D
1 I A = {1, 2, 3} and B = { p, q} , fnd A  B and B  A.
Are the two products equal?

2 A tetrahedral die A and a normal six-sided die B are tossed


simultaneously, thus A = {1, 2, 3, 4} and B = {1, 2, 3, 4, 5, 6} .
a i List the elements o the Cartesian product A  B.
ii Show that A  A  A  B .
b Write down the sets that represent the ollowing relations:
i I a  A and b  B, aRb  a + b is a prime number.
ii I a  A and b  B, aRb  b = a2.
iii I a  A and b  B, aRb  b  a is a prime number.
iv I a, b  A, aRb  a + b  B.

3 Given set A = {a, b} and set B = { p, q} , list all the elements o A  B


and fnd the number o subsets in the power set o A  B.

4 Let A = {a, b} , B = {l, 2, 3} and C = {3, 4} . Find:


a ( A  B)  ( A  C)
b A  (B  C)
c What can you conclude rom the answers to a and b?

5 Let A, B and C be three non-empty sets. Prove that i A  B then A  C  B  C .

6 Let S = {0, 2, 4, 6, 8} . Write out the elements o set R which is a subset o S  S


given by aRb  a  b.

7 Prove that or three non-empty sets A, B and C : ( A  B )  ( A  C ) = A  ( B  C ) .

8 Let A, B, C and D be our non-empty sets such that A  C and B  D .


Show that A  B  C  D .

9 For three non-empty sets A, B and C show that A  (B \ C ) = (A  B ) \ (A  C ).

10 Given that A = R+ and B = {x| x  Z + , x  10} , defne the relation R on A  B


as ollows: aRb  a = 2b. List all the elements that make up the
relation R. Find R1. What is the cardinality o R1?

24 The development of Set Theory


Equivalence relations
Among all the relations that can be established in sets there is a special
class, namely equivalence relations. d

As just discussed, a relation on a set A is a collection o ordered


e c
pairs (a, b) which are governed by this relation. As such,
a relation R on a set A is a subset o A  A.
For example, A = {a, b, c, d, e, f } ,
the sides o a regular hexagon, and Rl is the relation dened by
xR1 y  x is parallel to y, where, x, y  A. f b

Then we can say that segment a is parallel to itsel and also to d.


a
Segment b is parallel to itsel and also to e, etc.
Thus R1 = {(a, a), (a, d ), (b, b), (b, e), (c, c), (c, f ), (d, d ), (d, a), (e, e), (e, b), ( f, f ), ( f, c)}
which is a subset o A  A.
Note that in the hexagon example, or all elements a  A, we have the
ordered pair (a, a) in R. Thereore we say that xR1 x or all x  A, i.e.
the relation is refexive.
Also in this example we notice that i (x, y) is in R1 then ( y, x) is in R1 ,
or example (a, d ) and (d, a).
i.e. xR1 y  yR1 x or all x, y  A. The relation is said to be symmetric.
Now consider a dierent example, the set o all polygons P.
Let R be the relation on P  P dened by xRy  x and y are similar polygons.
R is refexive since any polygon is similar to itsel.
R is symmetric since i polygon x is similar to polygon y,
then y is also similar to x.
Now consider xRy  x is similar to y and yRz  y is similar to z.
Then by properties o similarity it ollows that x is similar to z.
Since xRy and yRz  xRz we say that the relation R is transitive.

Denition
A relation R dened on a set A is said to be an equivalence
relation i the ollowing three conditions are true:
 R is refexive, i.e. aRa or all a  A
 R is symmetric, i.e. aRb  bRa or all a, b  A
 R is transitive, i.e. aRb and bRc  aRc or all a, b, c  A
A
The diagram on the right illustrates an equivalence relation
on the set S = {A, B, C, D} . The arrows indicate the relation B

between individual elements o S, which are represented


by the vertices. Note that although C is related only to itsel,
the relation is still symmetric and transitive. D
C
In simple cases like the next example, it may be useul to draw
a similar diagram.

Chapter 1 25
Example 14
Let A = {l, 2, 3, 4} and R  A  A such that
R = {(1 , 1 ), (1 , 2), (1 , 3), (2, 1 ), (2, 2), (2, 3), (3, 1 ), (3, 2), (3, 3), (4, 4)} .
Is R an equivalence relation?
For all a  A, aRa We see that
R is refexive 1R1, 2R2, 3R3 and 4R4

For all a, b  A, i aRb then bRa We can see that

R is symmetric 1R2 and 2R1, 1R3 and 3R1, 2R3 and 3R2
Also or We see that
a, b, c  A, aRb and bRc  aRc 1R2 and 2R3 and 1R3
R is transitive 1R3 and 3R2 and 1R2
2R3 and 3R1 and 2R1
2R1 and 1R3 and 2R3
3R2 and 2R1 and 3R1
Thus, R is an equivalence relation. 3R1 and 1R2 and 3R2

Example 15
Let A = {l, 2, 3, 4} and Ri  A  A .
Construct the ollowing relations:
a A relation R1 that is refexive and symmetric but not transitive.
b A relation R2 that is refexive and transitive but not symmetric.
c A relation R3 that is symmetric and transitive but not refexive.
The ollowing working shows three examples to illustrate the relations. There are other
examples that you might be able to come up with.
a R1 = {(1 , 1 ), (2, 2), (3, 3), (4, 4), (1 , 2), (2, 1 ), (2, 3), (3, 2)}
Refexive because
1 R1 , 2R2, 3R3 and 4R4.
Symmetric because
1 R2 and 2R1 and 2R3 and 3R2.
a R b means that a is
Not transitive because not related to b.
1 R2 and 2R3 but 1 R 3.
b R2 = {(1 , 1 ), (2, 2), (3, 3), (4, 4), (1 , 2)}
Refexive because
1 R1 , 2R2, 3R3 and 4R4.
Transitive because, or example
1 R1 and 1 R2 and 1 R2.
Not symmetric because 1 R2 but 2 R 1

26 The development of Set Theory


c R = {(1 , 1 ), (l, 2), (l, 3), (2, l), (2, 2), (2, 3), (3, l), (3, 2), (3, 3)}
Symmetric because
1 R2 and 2R1 , 1 R3 and 3R1 , 2R3 and 3R2
Transitive because
1 R3 and 3R2 and 1 R2
1 R2 and 2R3 and 1 R3
2R1 and 1 R3 and 2R3
2R3 and 3R1 and 2R1
3R2 and 2R1 and 3R1
3R1 and 1 R2 and 3R2
Not refexive because 4 R 4

Example 16

A = {3, 4, 5, 9, 1 0, 1 1 , 1 3} and aRb  | a  b| is divisible by 5.


Show that R is an equivalence relation.
R = { (3, 3), (4, 4), (5, 5), (9, 9), (1 0, 1 0), (1 1 , 1 1 ), (1 3, 1 3),
(3, 1 3), (4, 9), (5, 1 0), (1 3, 3), (9, 4), (1 0, 5)}
| a  a| = 0 = 0  5, or all a, thereore R is refexive
| a  b| = | b  a|  aRb  bRa, thereore R is symmetric
| a  b| is divisible by 5  a  b = 5m, m  Z
| b  c| is divisible by 5  b  c = 5n, n  Z
Combining these two we obtain
a  c = 5(m + n)
 | a  c| = 5| m + n|  aRc. Thereore R is transitive.
R satises all three conditions necessary to qualiy as an equivalence relation.

Modular Congruence
The ollowing is a common example o equivalence relations. It generates
all the Z sets that will later be used to dene groups o every single order n.
x, y  Z are said to be congruent modulo n i | x  y| is divisible by n.
We denote this by x  y (mod n).
Consider the ollowing lists o numbers rom 1 to 60:
1 2 3 4 5 6 1 2 3 4 5
7 8 9 10 11 12 6 7 8 9 10
13 14 15 16 17 18 11 12 13 14 15
19 20 21 22 23 24 16 17 18 19 20
25 26 27 28 29 30 21 22 23 24 25
31 32 33 34 35 36 26 27 28 29 30
37 38 39 40 41 42 31 32 33 34 35
43 44 45 46 47 48 36 37 38 39 40
49 50 51 52 53 54 41 42 43 44 45
55 56 57 58 59 60 46 47 48 49 50
51 52 53 54 55
56 57 58 59 60
Chapter 1 27
These tables are another way o writing the congruences
modulo 6 ( (mod 6)) and modulo 5 ( (mod 5)). Note that the frst
column o each table represents x  1 (mod 6) and y  1 (mod 5)
respectively. Another way o expressing these numbers would be to
say that all the numbers in the frst column leave a remainder o
1 when divided by 6 and by 5 respectively. Similarly the second
column o each table represents the numbers that leave a remainder
o 2, and so on. So each column would represent a relation on all
the positive integers i we were to continue building up the tables.
It is also easy to see rom the tables that congruence modulo 6 and
modulo 5 are equivalence relations.
Lets look at the frst table only. Since each column represents x (mod 6)
with x  {0, 1 , 2, 3, 4, 5} with 0 representing 6 since 0(mod 6) represents
all the multiples o 6, we see that x  x (mod 6) since these are the
numbers in the frst row. Any two numbers in the same column are
congruent to each other modulo 6.
For example, 28  4 (mod 6) and 52  4 (mod 6)
 28  52 (mod 6) and 52  28 (mod 6) i.e. symmetric.
We can do this or any pair o numbers in the same column; this
leads to the conclusion that the relation congruence modulo 6
is symmetric.
Similarly i we take any three numbers in a column, we realize that they
59  35 (mod 6 ) 
are all related to each other. e.g.   59  1 1 (mod 6 ).
35  1 1 (mod 6 ) 
Again we can do this or any three numbers in a particular column,
thus we can deduce that congruence modulo 6 is a transitive relation
on the positive integers.
We can now list the frst table as ollows:

1(mod 6) 2(mod 6) 3(mod 6) 4(mod 6) 5(mod 6) 0(mod 6)


1 2 3 4 5 6
7 8 9 10 11 12
13 14 15 16 17 18
19 20 21 22 23 24
25 26 27 28 29 30
31 32 33 34 35 36
37 38 39 40 41 42
43 44 45 46 47 48
49 50 51 52 53 54
55 56 57 58 59 60

28 The development of Set Theory


Each column represents the diferent congruences modulo 6 and the
columns have divided the numbers into distinct, disjoint subsets
(equivalence classes). Although the table represents only the integers
1 to 60, it is clear that we could continue to build up the table endlessly.
All the positive integers could be included in such an endless table, and
they would all be separated into distinct equivalence classes representing
the particular congruence.
We say that  (mod 6) partitions the positive integers into
six equivalence classes.
Repeat the process above with  (mod 5) and show that this is an
equivalence relation.

We are now in a position to explain the properties o modular


arithmetic in a more rigorous manner.

Defnition
a is congruent to b modulo n i n divides (a  b),
i.e. (a  b) = kn, k  Z.

a  b(mod n)  n| (a  b)  (a  b) = kn, k  Z.
So, 1 4  0(mod 7) since 7| l4.
But 1 3  5(mod7) since 7| (1 3  5).

34  6(mod 7) since 7| (34  6).

Note that congruence


Theorem 5
(mod 0) does not exist
The relation R which is dened as : aRb  a  b(mod n), n  Z+ , is since we cannot
an equivalence relation on Z. divide by 0.

Proof:
a  a(mod n) since n| 0 or all a  Z
Thereore R is reexive.
a  b(mod n)  a  b = kn, k  Z
 b  a = kn, k  Z
 n| b  a  b  a (mod n)
Thereore R is symmetric.

Chapter 1 29
a  b(mod n)  a  b = pn, p  Z
b  c(mod n)  b  c = qn, q  Z
Adding  a  c = n(p + q), p + q  Z
 a  c (mod n)
Thereore R is transitive. Q.E.D.

Example 17
For each given set S and associated relation R, determine whether or not
R is an equivalence relation.
a S is the set o all people in Asia,
aRb  a and b have the same parents.
b S is the set o all people in Australia,
aRb  a and b live within 100 km o each other.
c S is the set o straight lines in a plane,
aRb  a is parallel to b.
a It is clear that aRa  R is refexive.
aRb  bRa since both have the same parents. R is symmetric
aRb  a and b have the same parents.
bRc  b and c have the same parents.
It ollows that a, b and c have the same parents so
aRc and R is transitive.
R is an equivalence relation.
b Clearly aRa. R is refexive.
aRb  a and b live within  00 km o each other  bRa. R is symmetric.
Let b live 90 km due east o a and c 80 km due east o b.
aRb and bRc but a is not related to c because
c lives  70 km due east o a. R is not transitive.
Thereore it is not an equivalence relation.
c By denition o parallel lines in a plane aRa.
Similarly aRb  bRa. So R is symmetric.
aRb  a is parallel to b
bRc  b is parallel to c
By denition o parallel lines, aRc which means
R is transitive, so R is an equivalence relation.

30 The development of Set Theory


Example 18

Let the relation R on N be dened as xRy  2x  y = 5n, n  Z.


Determine i the relation is:
a refexive
b symmetric
c transitive

a When x = 1 , 2  1  1  5n Proof by counter-example


Thereore x R x Substitute x = 1 into 2x  x.
R is not refexive. (We could also have chosen any other
non-zero natural number for x. )
b Let xRy  2x  y = 5n, n  Z Proof by counter-example
 2y  x = 4x  1 0n  x = 3x 1 0n Substitute y = 2x  5n into 2y  x
 3x  1 0n  5k, k  Z
Thereore y R x
R is not symmetric.
c 8R1 1 Proof by counter-example
1 1 R2 16  11 = 5
8R2 22  2 = 5  4
R is not transitive. 16  2 = 14
and 14 is not a multiple of 5.

A counter-example is a valid method to show that a property does not hold;


in fact it is the most common method to disprove  for all statements.

Example 19
Let S = {0, 2 } and the relation R = {(0, 0), (0, 2 ), (2 , 0), (2 , 2 )} .
Determine i R is an equivalence relation.

Draw a diagram to illustrate the


0 2 relation.
R is refexive because 0R0 and 2 R2 
R is symmetric because 0R2  and 2 R0.
R is also transitive because:
0R2 , 2 R2  and 0R2 
0R2 , 2 R0 and 0R0
R is an equivalence relation o S.

Note that when a relation includes all the elements of S  S the relation is
an equivalence relation.

Chapter 1 31
Exercise 1E
In questions  to 5 determine whether or not the given relation is
an equivalence relation on the dened set.
1 For a, b  Z, aRb  | a| = | b| .

2 For m, n  Z + , mRn  m and n have the same number o digits.

3 For x, y  R, xRy | x  y|  3.

4 For x, y  R, xRy  x + y  Z.

5 For p, q  Q, pRq  p  q  Z.

6 Let S = { fi (x)| fi (x) = mi x + ci , where mi , ci  R} . The relation


R is dened on S such that fi (x)Rfj (x)  mi = mj.
Show that R is an equivalence relation on S.

7 Let S = { fi (x)| fi (x) = mi x + ci , where, mi , ci  R} . The relation


R is dened on S such that fi (x)Rfj (x)  mi mj = 1.
Show that R is symmetric but not refexive or transitive.

8 The relation R is dened on Z such that mRn  m2  n2(mod 4).


Show that R is an equivalence relation.

9 The relation R is dened on R  R such that


(a, b)R(c, d )  a2 + b2 = c2  d 2.
Determine whether or not R is an equivalence relation.

p 
10 Let S =  : p , q  Z , q  0  . The relation R is dened on S such that
q 
a c
R  ad = bc . Determine whether or not R is an equivalence relation.
b d

1.6 Equivalence classes and partitions


Reer back to Example  4 where we had A = {l, 2, 3, 4} and
R  A  A such that R = {( ,  ), ( , 2), ( , 3), (2,  ), (2, 2), (2, 3),
(3,  ), (3, 2), (3, 3), (4, 4)} .
We create the set o those elements related to  , i.e.{l, 2, 3} .
Similarly the set o those elements related to 2, i.e. {l, 2, 3} .
And the set o elements related to 3 would also be {l, 2, 3} .

32 The development of Set Theory


Then the set o elements related to 4 is {4} .
We can also illustrate this by sketching a diagram.

4
3

Note that the equivalence relation R has separated A into two distinct
subsets, {l, 2, 3} and {4} . We call these the equivalence classes o the
elements o A under the relation R. The relation has induced a partition
o the set A into two disjoint subsets.
Now consider the relation x  y (mod 6) on Z + . We saw on page 28
that this relation distributes the positive integers into 6 distinct sets
o integers as ollows:
This diagram shows the
[1 ] = x  l(mod6)  x  {l, 7, 1 3, 1 9, . . .} division of a set into 6
[2] = x  2(mod6)  x  {2, 8, 1 4, 20, . . .} partitions by a given relation
[3] = x  3(mod6)  x  {3, 9, 1 5, 21 , . . .}
[4] = x  4(mod6)  x  {4, 1 0, 1 6, 22, . . .}
[5] = x  5(mod6)  x  {5, 1 1 , 1 7, 23, . . .}
[0] = x  0(mod6)  x  {6, 1 2, 1 8, 24, . . .}

Notice that the equivalence classes orm a partition o the set A.


We say that congruence modulo 6 divides Z+ into 6 distinct In some text books
sets denoted by Z6. The relation has broken up the infnite set Z into equivalence classes
a set o six infnite sets, each one called an equivalence class. are denoted by x or x .

These equivalence classes constitute a partition o the original set.


Hence we say that an equivalence relation induces a partition o the set.

Defnition
An equivalence class [x ] under an equivalence relation R on a set
A is the set o all elements related to x in A,
i.e. [x] = {a| a  A, aRx} .

Chapter 1 33
Lets reer back to Example 1 6 where A = {3, 4, 5, 9, 1 0, 1 1 , 1 3}
and aRb | a  b| is divisible by 5.
[3] is an equivalence
R = {(3, 3), (4, 4), (5, 5), (9, 9), (1 0, 1 0), (1 1 , 1 1 ), (1 3, 1 3), (3, 1 3),
class consisting of
(4, 9), (5, 1 0), (1 3, 3), (9, 4), (1 0, 5)} {3, 13} because 3R3
The equivalence classes induced by this relation are and 3R13. No other
[3] = {3, 1 3} , [4] = {4, 9} , [5] = {5, 1 0} and [1 1 ] = {1 1 } element in A is related
to 3.
Example 20
Let S = {1 , 2, 3} .
The relation R = {(1 , 1 ), (1 , 2), (2, 1 ), (2, 2), (3, 3)} .Show that R is an
equivalence relation and nd the partition o the set S induced by R.
It is easy to check that R is refexive, symmetric and
transitive. So R is an equivalence relation on S.
We can illustrate the relation on a diagram.

1
A diagram helps us visualize
2 the equivalence classes.

Under the relation R, [1 ] = {1 , 2} and [2] = {1 , 2}


and [3] = {3} .
Since [1 ] = [2], {[1 ], [3]} or {[2] , [3]} are partitions o S.

Example 21

T = {triangles} and R is dened on T such that or a, b  T, aRb  a is similar to b. Determine


whether R is an equivalence relation and explain the equivalence classes o T under R.

Any triangle is similar to itsel. Check the properties for an equivalence relation.
aRa  R is refexive
Triangles whose angle measures are the same Check the properties for an equivalence relation.
are similar to each other.
aRb  bRa  R is symmetric
Similar triangles are triangles o the same shape. Check the properties for an equivalence relation.
aRb  a is similar to b 
 a is similar to c
bRc  b is similar to c 
aRc  R is transitive
Hence, R induces the partition o T into those
triangles which are similar to each other.

34 The development of Set Theory


Note that in all the examples seen so ar the equivalence classes
ormed by a particular relation are disjoint. In the ollowing
theorem we will prove that an equivalence relation generates a
collection o disjoint subsets whose union is the set itsel.
In other words, we will now ormally prove that an equivalence
relation R on a set A induces a partition o A.

Theorem 6
Equivalence classes ormed by an equivalence relation on a set
A are disjoint, and their union is A.

Proof:
First we need to prove that [xi ] are disjoint or dierent
values o i.
Assume that two equivalence classes [xi ] and [xj ] are not disjoint.
Then there must be some a  A such that a  [xi ] and a  [xj ].
By denition o equivalence classes this means that aRx where x  [xi]
and aRy where y  [xj ].

Because o the symmetric and transitive properties, this can be true


only i [xi ]  [xj ] and [xj ]  [xi ]  [xi ] = [xj ].

That is, i aRx then xRa since R is symmetric, and xRa together with
aRy implies that xRy because R is transitive. Hence [xi] = [xj ].
Thereore equivalence classes are disjoint.

Now we need to prove that the equivalence classes are exhaustive,


The most trivial
i.e. all o set A is partitioned by the set o equivalence classes. case of the partition
Since R is an equivalence relation we know that it is refexive: would be when each
equivalence class has
xRx or all x  R only one element.

So at least one element x  [x], which means that there is no


element in A which does not belong to an equivalence class.
This means that the equivalence classes [xi ] partition the set A.
We say that an equivalence relation induces a partition o a set. Q.E.D.
In the ollowing example you will see how an equivalence relation
on R2 divides the plane into equivalence classes that can be
described geometrically.

Chapter 1 35
Example 22
A relation R is dened on R  R as ollows: (a, b) R (c, d )  2a  b = 2c  d
a Show that R is an equivalence relation.
b Find the equivalence classes and explain them geometrically.
a (a, b)R(a, b) Show that the properties o
2a  b = 2a  b or all a, b  R equivalence relations are
Thereore R is refexive. satisfed.
(a, b)R(c, d )  2a  b = 2c  d
 2c  d = 2a  b
 (c, d )R(a, b)
Thereore R is symmetric.
(a, b)R(c, d )  2a  b = 2c  d
(c, d )R( p, q)  2c  d = 2p  q
 2a  b = 2p  q
 (a, b)R( p, q)
Thereore R is transitive, hence R is an equivalence relation.
b Let (x, y)  [(a, b)] One o these lines would be
2x  y = k where k = 2a  b the line
 y = 2x  k y = 2x  (x, y)  [(1, 2)]
This represents the set o lines parallel to y = 2x.

In the ollowing example we will look at a relation that categorizes


the integers into odd and even numbers.
Example 23
The relation R is dened on Z such that aRb  a + b is even.
a Show that R is an equivalence relation.
b Find the partitions o Z under R.
a aRa Show that the properties o
a + a = 2a  aRa  Z equivalence relations are satisfed.
Thereore R is refexive.
aRb  a + b is even
 b + a is even  bRa Addition is commutative in Z.
Thereore R is symmetric.
aRb  a + b = 2p p Z
bRc  b + c = 2q q Z
 a + c = 2( p + q  b)  aRc I a is odd then c must be odd and i
Thereore R is transitive. a is even then c must also be even.
b Let x  [a]
 xRa  x + a = 2n Under R
Thereore R partitions Z into two equivalence classes Z = { Z1  Z 2} since
[1 ] and [2] which represent the odd and even Z1 = 1(mod 2) = { odd numbers}
numbers respectively. Z 2 = 0(mod 2) = { even numbers}

36 The development of Set Theory


The next relation organizes ordered pairs o integers along lines passing
through the origin.

Example 24

The relation R is dened on Z +  Z + such that (a, b)R(c, d )  ad = cb.


Show that this is an equivalence relation and give a geometric description
o the equivalence classes.
(a, b)R(a, b) Show that the properties of an equivalence
ab = ab relation hold.
Thereore R is refexive

(a, b)R(c, d )  ad = cb
 cb = ad
 (c, d )R(a, b)
Thereore R is symmetric

(a, b)R(c, d )  ad = cb  adq = cbq


(c, d )R( p, q)  cq = dp  cqb = dpb
 adq = cbq = dpb  aq = pb
 (a, b)R( p, q)
Thereore the relation is transitive.

Let (x, y)  [(a, b)] This is illustrated on the


 xb = ay diagram below:
b
 y= x
a 4

The equivalence class [(a, b)] represents 3

ordered pairs o positive integers which lie 2


on the straight lines passing through 1
b
the origin with gradient . 4 3 2 1
0
1 2 3 4 5 6
a 1
2
3
f 4

Chapter 1 37
Example 25

The relation R is dened on S = {x| x  Z + , x  1 5} by


aRb  a(a 1 )  b(b  1 )(mod 7).
a Show that R is an equivalence relation.
b Show that the equivalence R can be written in the orm
(a  b)(a + b  1 )  0(mod 7).
c Hence, or otherwise, determine the equivalence classes.

a a(a  1 )  b(b  1 )(mod 7) We need to confrm the properties o an


 a(a  1 ) b(b  1 ) = 7n, n  Z equivalence relation.
Refexive:
aRa  a(a  1 )  a(a  1 )(mod 7)
a(a  1 )  a(a  1 ) = 0n = 0
Symmetric:
aRb  a(a  1 )  b(b  1 ) = 7n
 b(b  1 )  a(a  1 ) = 7(n)  bRa
Transitive:
aRb  a(a  1 )  b(b  1 ) = 7n
bRc  b(b  1 )  c(c  1 ) = 7m Add the two equations.
 a(a  1 )  c(c  1 ) = 7(n + m)
 aRc
b a(a  1 )  b(b  1 ) = 7n
 a2  a  b2 + b = 7n Expand.
 (a  b)(a + b  1 ) = 7n  0(mod 7) Rearrange and actorize.

c  (a  b)(a + b  1 ) = 7n.
a  b = 7n or a + b  1 = 7n, n  Z Since the product is divisible by 7
Thereore the equivalence classes are: b = a  7n or b = 7n  a + 1
[l] = {l, 7, 8, 1 4, 1 5} substitute a = 1 and n = 0, 1, 2
[2] = {2, 6, 9, 1 3} substitute a = 2 and n = 0, 1, 2
[3] = {3, 5, 1 0, 1 2} substitute a = 3 and n = 0, 1, 2
[4] = {4, 1 1 } substitute a = 4 and n = 0, 1, 2

The next example illustrates how the innite set Z2 is partitioned into
six equivalence classes.

38 The development of Set Theory


Example 26

The relation R is dened on Z  Z such that (a, b)R(c, d ) i and only i


a  c is divisible by 2 and b  d is divisible by 3.
a Show that R is an equivalence relation.
b Find the equivalence class or (1 , 3).
c Write down the ve remaining equivalence classes.

a Refexive: (x, y)R(x, y) We need to confrm the properties o an


since x  x = 0 and y  y = 0 equivalence relation.
which are both divisible by 2 and 3
so R is refexive.
Symmetric:
(x, y)R(a, b)
 x  a = 2m, m  Z
 a  x = 2m
y  b = 3n, n  Z
 b  y = 3n so R is symmetric.
Transitive:
(x, y)R(a, b) and (a, b)R(c, d )
x  a = 2p
  x  c = 2( p + q )
a  c = 2p 

y  b = 3m 
  y  c = 3( m + n )
b  c = 3n 
(x, y)R(c, d ) so R is transitive.
Thereore R is an equivalence relation.
b (x, y)R(1 , 3)
Let x  1 = 2m  x = 2m + 1
y  3 = 3n  y = 3n + 3 = 3n Since n is any integer we can write 3n.
So
[(1 , 3)] = {(x, y)| x = 2m + 1 , y = 3n,
m, n elements o Z}
c The other equivalence classes will be
{(x, y)| x = 2m, y = 3n} i.e. [(2,3)]
{(x, y)| x = 2m, y = 3n + l} i.e. [(2,1 )]
{(x, y)| x = 2m, y = 3n + 2} i.e. [(2,2)]
{(x, y)| x = 2m + 1 , y = 3n + l} i.e. [(1 ,1 )]
{(x, y)| x = 2m + 1 , y = 3n + 2} i.e. [(1 ,2)]

Chapter 1 39
Example 27
The relation R is dened on cubic polynomials P o the orm
Pn (z) = z3 + az2 + bz where a, b  R, z  C.
The relation R is dened by P1 RP2 i and only i the
sum o the three zeros o P1 is equal to the sum o the three zeros o P2.
a Show that R is an equivalence relation.
b Determine the equivalence class containing z3  2z2 + 8z.

a Let the zeros o Pn (z) be n,  n, n

Since Pn(z) = z (z2 + az + b)


We know that n = 0 or all n
So sum o roots becomes n + n = a Using Vietes theorem about sum and product
o roots.
Refexive:
Pn(z )RPn (z )
The sum o the zeros o Pn (z) is equal to
the sum o the zeros o Pn (z).

Symmetric:
P1 (z)RP2(z)   1 + 1 = 2 + 2 = a Coefcient o z 2 is the same in both cubic
 P2 RP1 polynomials.

Transitive:
P1 (z)RP2(z)   1 + 1 =  2 + 2 = a Coefcient o z 2 is the same in all three cubic
P2RP3   2 + 2 = 3 + 3 = a polynimials.

 1 + 1 =  3 + 3
 Pl(z)RP3 (z) Using Vietes theorem.

Thereore R is an equivalence
relation.

b The equivalence class One o the roots is zero and the sum o roots
containing z 3  2z 2 + 8z consists must be two. The product o the two remaining
o cubic polynomials o the zeros could be any number.
orm z 3  2z 2 + bz

40 The development of Set Theory


Exercise 1F
1 Consider the set o words:
W = {set, table, chair, car, tennis, bike, stairs, sea, wave, sun} .
In a and b, show that R is an equivalence relation
and list the equivalence classes induced by each relation on W.
a R is the relation on W, has the same number o letters.
b R is the relation on W, starts with the same letter
o the alphabet.

2 a Let L = {li| li is a line segment o length | li| } . Let R be a relation


on L such that li R lj  | li| = | lj|. Show that this is an equivalence
relation on L and describe the partition induced by R.
b Let P = {polygons} and R be a relation on P such that
aRb  a has the same number o sides as b . Show that R is
an equivalence relation and describe the partitions induced by R.

3 Let P = { f (x)| f(x) = ax2 + bx + c, with a, b, c  R} . The relation


R on P is such that f (x)Rg(x)  f(0) = g(0). Show that R is an
equivalence relation and describe the partition induced by R on P.

4 Let S = {(x, y)| x, y  R} . Let R be a relation on S such that


aRb  a2 + b 2 = r2 where r  R+ .
Show that R is an equivalence relation and give a geometric
meaning o the partitions o R  R under this relation.

5 Let R be a relation on Z + such that aRb  a + 2b is divisible by 3.


Show that R is an equivalence relation and list the equivalence
classes o Z+ under this relation.

6 Let R be a relation defned on Z+ such that aRb  a2 = b 2 (mod 3).


Show that R is an equivalence relation and list the equivalence
classes o Z+ under this relation.

7 Show that the relation R defned on R2 such that (a, b)R(c, d )  a = c


is an equivalence relation and give a geometrical description o the
equivalence class [(a, b)].

8 Show that the relation R defned on (Z+ ) 2 such that (a, b)R(c, d )  ad = cb
is an equivalence relation. Describe the equivalence class [(l, 2)].
Hence or otherwise describe the partition induced by R on (Z + ) 2.

9 Let R be a relation defned on {R2 \ (0, 0)} such that


(a, b)R(c, d )  ab = cd. Show that R is an equivalence relation on
{R2 \ (0, 0)} . Describe the equivalence class [(l, l)]. Hence or otherwise
describe the partition induced by R.

Chapter 1 41
10 Consider the relation R on Q such that xRy  x  y  Z.
a Show that this is an equivalence relation.
b Determine the equivalence class [0] or this relation.

Determine the equivalence class   or this relation.


3
c
4
d Describe the partition induced by R on the rational numbers.

Review exercise
EXAM-STYLE QUESTIONS
1 A, B and C are subsets o the universal set U.
a Use Venn diagrams to illustrate
i A\B = A  (U \ B)
ii (A\B )  (B\ A) = (A  B )\(A  B )
b Use double inclusion to prove that A\B = A  (U \ B ).
c Use De Morgans laws to prove that (A\B )  (B\A)
= (A  B )\(A  B ).

2 A, B and C are subsets o the universal set U. Use Venn diagrams


to illustrate the distributive laws. Use these properties and
De Morgans laws to show that (A  B )  C = (A  C) (B  C)

3 The relation R on C\{0} is defned as: z R z2  arg z = arg z2 or z , z2  C\{0} .


a Show that R is an equivalence relation on C.
b Describe the equivalence classes under the relation R.

4 Sets A, B, C, D and E are subsets o Z:


A = {n | 0 < n < 20, n is a prime number}
B = {n | | n  2 |  l}
C = {n | n2  3n  4 < 0}
D = {n | n5 =  6n}
E = {n | (n   ) 2  4}
a List the elements o each o these sets.
b Determine, giving reasons, which o the ollowing statements
are true and which are alse.
i n(A) = n(D) + n(E )
ii n(D  A) = 1
iii B  E
iv (D \ B )  A = 
v C E = 

42 The development of Set Theory


5 Let R be a relation on Z such that aRb  5ab  0.
a Determine whether R is
i refexive
ii symmetric
iii transitive.
b Write down whether or not R is an equivalence relation and give
a reason or your answer.

6 The relation R is dened on the set N such that or


a, b  N, aRb  a3  b3 (mod 5).
a Show that R is an equivalence relation.
b Denote the equivalence class containing n by Cn.
i Find C0.
ii List the rst six elements o C1 .
iii Prove that Cn = C n+5 or all n  N.

7 P is the set o polynomials o the orm P (z) = z2 + bz + c


where b, c  R and z  C.
a The relation S on the set P is such that P1RP2  the sum o the
zeros o P1 is equal to the sum o the zeros o P2.
i Show that S is an equivalence relation.
ii Determine the equivalence class containing the polynomial
P = z2  3z + 4.
b The relation R on the set P is such that P1 RP2  the product
o the zeros o P1 is equal to the product o the zeros o P2.
i Show that R is an equivalence relation.
ii Determine the equivalence class containing the polynomial
P = z2  3z + 4.

8 The relation R is dened on Z + such that aRb  5 a  5 b (mod 8).


a Show that R is an equivalence relation.
b Identiy the two equivalence classes ormed by this relation.
c Find the value o 5 355 (mod 8).

9 The relation R is dened on Z  Z such that (a, b)R(c, d )


i and only i a = c and b  d is divisible by 5.
a Prove that R is an equivalence relation.
b Describe the equivalence classes induced by R.

Chapter 1 43
10 The relation S is defned on quadratic polynomials P o the orm:
P (z) = z2 + az + b, where a, b  R, z  C.
The relation S is defned by P1 SP2 i and only i P1 and P2 have
at least one zero in common. Determine whether or not S is transitive.

11 The points in a plane or space are given. AB is a directed line


segment where A is the starting point and B is the terminal point.
AB R CD i line segments [AD] and [BC] have a common midpoint.
a Show that R is an equivalence relation.
b Give a geometrical description o the partition o all the directed line
segments in a plane or space.

Chapter  summary
A set S is a collection o objects. I x is one o these objects we say x  S.
The number o elements in a set is called the cardinality o the set.

The empty set denoted by  = { } .


B  A and A  B  A = B . The converse o this is also true,
i.e. i A and B are equal sets then A is a subset o B and B is a subset o A.
I set S  U, then the complement o S is denoted by S  where S  = { x  U | x  S}.
The intersection o two sets A and B is denoted by A  B where
A  B = { x | x  A and x  B} .
The union o two sets A and B is denoted by A  B where A  B = { x | x  A or x  B} .
I A  B =  then A and B are said to be disjoint sets.
The set consisting o those elements that are in set A but not in set B is called
the set diference B rom A denoted by A \ B = { x| x  A and x  B } = A  B  .
The symmetric diference o two sets A and B is denoted by A  B and consists
o those elements which are either in A or in B but not in both A and B.
A  B = ( A  B ) \ ( A  B ) = ( A \ B )  ( B \ A ).

The power set o a fnite set S with n elements is the set o all subsets o
S including the empty set  and S itsel. The total number o distinct subsets
o a fnite set S with n elements is 2n. n(P (S )) = 2 n

44 The development of Set Theory


Commutative Laws
AB=BA
AB=BA
Distributive laws
A  ( B  C ) = ( A  B )  ( A  C ), i.e. intersection is distributive over union
A  ( B  C ) = ( A  B )  ( A  C ), i.e. union is distributive over intersection
Associative laws
A  (B  C) = ( A  B)  C
A  (B  C) = ( A  B)  C
De Morgans Laws
(A  B) = A  B 
(A  B) = A  B 
Cartesian product
A  B = { ( a, b ) : a  A, b  B}
A relation R defned on a set A is said to be an equivalence relation i the
ollowing three conditions are true:
 R is refexive, i.e. aRa or all a  A
 R is symmetric, i.e. aRb  bRa or all a, b  A
 R is transitive, i.e. aRb and bRc  aRc or all a, b, c  A
a is congruent to b modulo n i n divides (a  b), i.e. a  b = kn, k  Z.
Notation: a  b (mod n )  n a  b
An equivalence class [ x ] under an equivalence relation R on a set A is
the set o all elements related to x in A, i.e. [ x ] = { a | a  A , aRx }.
A partition o a set A consists o another set P made up o non-empty
subsets o A which are disjoint and whose union makes up the whole set.
Equivalence classes are mutually exclusive and the set A is partitioned
into equivalence classes by an equivalence relation R on A.

Chapter 1 45
Extension of
2 the concept
of function
CHAPTER OBJECTIVES:
8.3 Functions: injections, surjections, bijections; composition of functions and
inverse functions.
8.4 Binary operations and operation tables (Cayley tables).
8.5 Binary operations: associative, distributive and commutative properties.
8.6 The identity element e.
The inverse a 1 of an element a.
Proof that left-cancellation and right-cancellation by an element a hold, provided
that a has an inverse.
Proofs of the uniqueness of the identity and inverse elements.

Before you start


You should know how to: Skills check:
1 Find the intervals for which the function 1 When a certain drug is administered,
10x the concentration of medication in the
f( x ) = , x  2 is increasing or
x2 bloodstream t hours after the drug is
decreasing. Find the derivative: 4t
administered is given by: A ( t ) =
( x  2 )(1 0 )  1 0 x 20 2
3 t + 27
f ( x ) = 2
=
( x  2) ( x  2)2 a Over which interval of time is
Since f (x) < 0 for all values of x in the the concentration of medication
domain, it follows that f (x) is a strictly increasing?
decreasing function. b Over which interval is the
concentration decreasing?
2 Find the inverse of the function 2 Find the inverse function for each of the
x +1 following:
f( x ) = , where x  R, x  1
x 1 x + 3
and state its domain. a f: x  , x  -2
x + 2
Interchange y and x, and make x the b f : x  2x
subject of the formula: c f : x  e x  2e  x
y +1
x =  x ( y  1) = y + 1
y 1
 xy  y = x + 1
1 + x
 y=
x 1
x +1
f 1 ( x ) = where x  R, x  1
x 1

46 Extension of the concept of function


Evolution of the function concept
So ar you have studied unctions as ormulas defned on real
number sets where every ordered pair (x, y)  R2 represented a
dependent variable y  R which was a unction o x  R, the
independent variable. You learned that, or any given unction, there
is a rule that determines the unique value o y or any value o x and
this could be illustrated by a graph o this unction, e.g. the ordered
pair (1 , 5) would be a point on the graph o  (x) = 2x2 + 3.
The term unction frst appeared in a letter written by Leibniz in The name Nicolas
1 673. He used it to describe quantities related to curves. In 1 755 Bourbaki does not
represent just a single
Euler introduced a more general concept when he wrote When
mathematician. A small
certain quantities depend on others in such a way that they undergo a change
group of French
when the latter change, then the frst are called unctions o the second.
mathematicians used this
In the 1 9th century more emphasis was placed on rigour in name as a pseudonym in
mathematics. The notion o unction continued to evolve with the the mid 1930s. The
development o Set Theory by Cantor. Cauchy was the frst to group was originally
consider the act that a unction may have a restricted domain. formed to write rigorous
textbooks based on Set
This eventually led to the defnition o unctions by Dedekind in
Theory initiated by Cantor.
1 888 that said a unction is a single-valued relation between two
However their work
non-empty sets. However the most accurate defnition o a unction
included studies of many
was given by Nicolas Bourbaki in 1 939 which described a unction branches of mathematics
as a possibly infnite set o ordered pairs (x, y) in which each x is including Topology.
paired with only one y.

Chapter 2 47
2.1 Functions as relations
A relation that associates each element in a non-empty set S with a
unique element in a non-empty set T is called a unction rom S to T.
A unction rom S to T is a subset M o S  T such that or every s  S
there is a unique t  T such that (s, t)  M.
For example: S = {2, 3, 4} , T = {1 , 2, 3, ... , 9} and M = {(2, 5), (3, 7), (4, 9)} .
We can represent this pictorially as shown here.
S T
1
2 2
3
4
3 5
6
7
4 8
9

2 is mapped onto 5 so we say that 5 is the image o 2 under this unction.


The unction has a rule that enables us to fnd the image o every element
o S under f. In this case the rule is f (s) = 2s + 1 .
We denote this unction by f : S  T such that s  2s + 1 or all s  S.
The set S is called the domain and T, the target set, is called the
co-domain. The set f (S ) = {t| t  T, t = f (s) or some s  S} is called
the range. It is the set containing all the images o S under the
unction f. In the above example the range is the set {5, 7, 9} .
Example 1
Determine which o the ollowing relations are unctions, and state the domain
and the range or those which are unctions.
a b c
1 0 9 3 7 0
1
1
0 1 1 0 3
2
1
1 2 0 3 1 3

a This is a unction. Each element on the domain has a unique


Domain {1 , 0, 1 } , Range {0, 1 } image in the co-domain.
b This is not a unction.
In this relation 9 and 1 are mapped to 3
and 1 respectively, hence 9 and 1 are not
mapped to unique elements.
c This is a unction.
Domain {1 , 3, 7} , Range {0, 1 , 3} Each element is mapped to a unique image.

48 Extension of the concept of function


Example 2
Determine which o these relations are unctions:
a R on Z such that aRb  a2 = b2
b R on R+ such that aRb  a2 = b2  1

a This is not a unction.


1 R1 since 1 2 = 1 2
1 R (1 ) since 1 2 = (1 ) 2
Thereore 1 is mapped to two distinct elements
so it is not a unction.
b Suppose this is not a unction. Then:
aRb  a2 = b2  1 Subtraction o the two equations yields:
aRc  a2 = c2  1 0 = b2  c2
 b2 = c2
b=c Since R is defned on R+
Thereore R is a unction. Proved by contradiction.

There are two rules governing unctions as ollows:


  (s) must speciy an element o T or every s  S
 i s = s and both s, s  S then  (s) =  (s)
Although the above rules may look trivial, they actually have strong
implications. They make sure that there are no contradictory or
ambiguous connotations. Lets take the example o a unction defned
on the rational numbers as ollows:
p q
Let  :    such that f   = .
q p
Because the domain o this unction is , every element o  must have
an image in . However by the defnition o this unction it is clear that
0 does not have an image because division by zero is undefned. In other
words the frst rule ensures that we do not have any singularities.

You have met and discussed singularities in the core book, when
discussing limits and graphs o unctions. At a singularity the mathematical
x2  1
unction is not defned or is not well-behaved , e.g. f( x) = , has a
x +1
singularitiy at x = 1 and the graph is a straight line with a hole at x = 1.
1
Similarly the unction f( x) = is not defned when x = 0 and so this is a
x
vertical asymptote. The unction also has a singularity at x = 0. The unction
f(x) = | x  1 | is continuous but it is not dierentiable at x = 1. Once more there
is a singularity at x = 1. The unction is said to be not well-behaved at x = 1.

The second rule ensures that the unction is well-defned. In other words,
it ensures that each element o the domain has only one image in the co-domain.
This is illustrated by the ollowing example in which the rule is violated.

Chapter 2 49
p 1
Suppose that f :    such that f   = .
q q
But f   = and f  
2 1 10 1
= which violates the second rule
 
5 5   25
25
governing unctions.

Equality of functions
Two unctions f : S  T and g : P  Q are equal i and only i S = P,
f (S ) = g (P ) and f (a) = g (a) or all values o a  S.
We can illustrate this property by looking at some unctions.
x
Consider the unctions f: R \ {0}  R \ {0} such that f ( x ) = ,
x2
1
and g :  \{ 0}   \{ 0} such that g( x ) = .
x
These two unctions are equal because they both have the same domain
which excludes x = 0 and or all values o x in the domain, f (x) = g (x).
Now let us dene another two unctions as ollows:
f (x) = x with domain x  R
g (x) = arccos(cos x) with domain x  R
I we compare these two unctions we see that f (2 ) = 2 , but
g (2) = arccos(cos2 ) = arccos(1 ) = 0. The image o 2  under f is
diferent to the image o 2 under g, so the unctions are not equal.
These two unctions are equal only i we restrict both domains to
For a unction to be
x  R, 0  x < 2 .
injective we can state
that i a is not equal to
Defnition b in the domain, then
f(a) is not equal to f(b)
A unction is said to be injective i it preserves distinctness. In other in the co-domain.
words, every element o the co-domain is mapped to by no more The contrapositive
than one element in the domain. A unction f : S  T or which statement o this is
each element o the range, f (S), corresponds to exactly one element that i f(a) = f(b)
o the domain, S, is said to be injective. In other words, i f (a) = then a = b.
Contrapositive
f (b)  a = b or a, b  S. A logically equivalent statement would be:
statements are useul
a  b  f (a)  f (b). when it is difcult
to examine all the
dierent elements o
We say that an injective unction is one in which each element o
the domain to check
f (S), the range o the unction, is the image o only one element o S, or unique images in
the domain o the unction. An injective unction is thereore a the co-domain. Then
one-to-one unction. it is much easier to
check by using the
logically equivalent
contrapositive
statement.

50 Extension of the concept of function


1 2 1 1 1 2
1
1 0 0
0 0
1
0
1 2
4 1 1 1 3
2 4
2

One-to-many : not a unction Many-to-one : is a unction One-to-one : injective unction

The diagrams above illustrate three types o mapping.


 The rst mapping is not a unction because elements 1 and 4 both
have two images in the co-domain.
 The second mapping is a unction because each element in the
domain is mapped onto a unique element in the co-domain.
It is not injective because 1  co-domain is the image o 1 and
1 in the domain. Alternatively we can say that two diferent
elements 1 and 1 rom the domain don't have two diferent
images in the co-domain but have the same image o 1 . The
unction doesn't preserve distinctness.
 The third mapping is a unction because each element in the domain is mapped
onto a unique element in the co-domain. It is also injective because each
element in the range is the image o only one element in the domain.

Defnition
I every element in the co-domain o a unction is the image o at
least one element in the domain we say that the unction is a
surjection, i.e. or all b in the co-domain there exists an a in the
domain such that f (a) = b. A surjection is also called an onto unction.
None o the previous three examples represent surjections. The two
examples below illustrate surjections.

2 1 2

1 0 0 0

0 1 2
1
1 2 4

Many-to-one : surjection One-to-one correspondence : injection and surjection

To check or injection we look at the elements in the domain and


check that diferent elements have diferent images in the co-domain.
We can establish a relationship between the cardinality o sets S and
T as ollows: n (S )  n (T). For a surjection we look at the elements in
the co-domain and check that they are all images. So or a surjection,
n (T)  n (S ). I a unction is both surjective and injective then
n (S ) = n (T). This is the case in the second mapping above.
Chapter 2 51
Defnition
A function is a bijection if it is an injection and a surjection.
A bijection is also called a one-to-one correspondence.

In the core book we studied functions of real variables, i.e. when the domain
was a subset of R. It is good to remember that a graph that did not pass the
vertical line test did not represent a function. This is a one-to-many relation.
If a function passed the horizontal line test then the function is an injective
function. Graphically speaking, if a horizontal line drawn in any region of the
co-domain crosses the graph exactly once then the function is a bijection.
We also say that the function is surjective if any horizontal line drawn in any
region of the co-domain crosses the graph at least once.

Example 3
A = {1 , 2, 3, 4} and B = {5, 6, 7, 8, 9} . Given that  : A  B such that
 (x) = x + 4, determine whether  is an injection, a surjection or both.
 (a) =  (b) Assume two elements in the co-domain are equal.
a+4=b+4 Prove that they are the images o the same element
a=b in the domain.
So  is an injection.
Let y  B  y = x + 4, x  A Given any element in the co-domain try to nd an
 x= y4 element in the domain which maps onto it. It is
y= 9  x= 5 5A sufcient to nd one such element or which the
So  is not a surjection. statement is not true.

Example 4

Given  : Z  Z such that  (x) = x + 4, determine whether  is an injection,


a surjection or both.
 (a) =  (b) Use the contrapositive statement o injective
a+4=b+4 unctions.
a=b
So  is an injection. Check that each element in the co-domain is the
Let y  Z  y = x + 4, x  Z  x = y  4 image o an element in the domain.
since for all y  Z, y  4  Z,  is a
surjection.

52 Extension of the concept of function


Examples 3 and 4 involve discrete unctions. In Example 3 the unctions were
mappings between fnite sets, and in Example 4 they were mappings between
infnite sets. We now look at examples with unctions as mappings between
infnite continuous sets.

Example 5

The unction  is defned by  : R+  R+ where  (x) = e cos2x + 1 .


a Find the exact range, A, o .
b i Explain why is not an injection.
ii Giving a reason, state whether or not is a surjection.

a The range o cos2x is the interval [1 , 1 ], Find the minimum and maximum values that
so A = [e1 + 1 , e + 1 ] cos2x can take to nd the range o .
b i Method I
For to be an injection or one-to-one
unction, x  y   (x)   (y)
 (0) =  (n) = e + 1 , n   We know rom the core syllabus that  is a
Thereore is not an injection. periodic unction. Use a counter example to show
that  is not injective.
Method II
 (x) = e cos2x + 1
  (x) = (2sin 2x)e cos2x Since the unction is continuous and
  (x) < 0 or 0 < x <
 diferentiable over the whole domain we can use
2 the derivative.

and  (x) > 0 or < x < 
2
Since  (x) is not strictly increasing or
decreasing over the whole domain it
is not injective.
ii The co-domain o is R+ but the Use the result o part a to show that  is not a
range o is A = [e1 + 1 , e + 1 ] surjection.
i.e. or y  R+ , y  A there is no x  R+
such that (x) = y. Thereore is not
surjective.

Note that the derivative test shown in the second method can be used
only for functions that are continuous and differentiable on the given
domain. It is not valid for functions that are discrete, like the ones in
Examples 3 and 4.

Chapter 2 53
Example 6
Consider the following functions:
f: ]2, +[ R+ where f (x) = (x  2)(x + 1 )
g : R  R  R  R where g (x, y) = (cos(x  y), x  y)
a Show that fis bijective.
b Determine, with reasons, whether
i g is injective
ii g is surjective.

a Method I
Injective: f (a) = f (b)
 (a  2)(a + 1 ) = (b  2)(b + 1 )
 a 2  a  b2 + b = 0
Use quadratic formula to solve for a:
Solve for a.
1 1  4( b  b 2 )
a= 2
2 x =| x| and since b > 2 we have
(1  | 1  2 b| ) | 1  2b| = 1 + 2b. The second solution is
= discarded because it is out of the domain.
2
(1  1 + 2 b )
=
2
a=b
So f (x) is an injection.
Surjective: Let f (x) = y Show that for all possible values of y in
 y = x2  x  2  x2  x  2  y = 0 the co-domain, there is a value of x in the
domain.
1+ 9 + 4y
 x =
2
For all y  R+ , 9 + 4 y > 3, so x > 2.
Therefore for all y  R+ there is x  ]2, +[ such
that f (x) = y.
Therefore fis a surjection.
Since fis injective and surjective it is a bijection.
Method II
Sketch the graph of f :
4
3
2
1

3 2 1 0 1 2 3
1
2
Use the graph of f with the horizontal
3
line test.

54 Extension of the concept of function


The graph o fpasses the horizontal line test,
thereore f is injective.
From the graph it is clear that the range o f is
equal to the co-domain so f is surjective. Compare range and co-domain on graph.
Since fis both an injection and a surjection
it is a bijection.
5  
b i g (2 , ) = g  ,  Find a counterexample.
 4 4
Thereore it is not injective.
ii The range o g is contained in 1  cos(x  y)  1
[1 , 1 ]  R  R  R so g is not surjective.

In all the examples above the domains are real numbers or ordered pairs.
However by our defnitions o unctions we may have domains or co-domains
that are not subsets o R or R2. The ollowing example illustrates this.

Example 7

Let P = { p ( x )| p ( x ) = a n x n + an 1 x n 1 + a n  2 x n  2 + . . . + a1 x + a0 , ai   , n   }
and f : P  P such that f ( pn) = pn. Determine whether f is an injection,
a surjection or both.
f ( pi) = f( pj)  pi = pj Use the contrapositive statement of injective
This is true even when functions.
pi = anx n + an1 x n1 + ...+ a1 x + a0
pj = anx n + an1 x n1 + ...+ a1 x + b0, a0  b0
So fis not injective.
Let pi  P such that f ( pi) = p
 p = pi
Then pi =  pdx and there are infnitely Since  pdx = q + c where q is a polynomial of
degree n + 1 and c can take any real value.
many pi  P that satisy this condition.
Thereore fis surjective.

In Chapter 1 you learned that the cardinality o a set S, denoted by n (S ), is the


number o elements in the set S. In the ollowing theorem we are going to use
the cardinality o fnite sets to obtain two results or injection and surjection
o unctions with fnite domains and co-domains.

Theorem 1
Let f : S  T where S and T are fnite sets.
Then a ( )
f is inj ective  n f ( S ) = n ( S )

b f is surj ective  n ( f ( S ) ) = n (T )

Chapter 2 55
Proof:
a ( )
f is injective  n f ( S ) = n ( S )
 For any unction we know that n ( f (S))  n(S ) since we cannot
have more images than we have elements in the domain.
Lets assume that n ( f (S )) < n(S ). Then there must be at least one
pair o diferent elements in S that have the same image, which is
in contradiction with the act that f is injective. Thereore
n(f (S )) = n(S ).
 I S = {x1 , x2, ..., xn} then { f (x)} = { f (x1 ), f (x2), ..., f (xn)} .
I f (xi) = f (xj) or some i  j then n( f (S ))  n  1 , which is a
contradiction. Thereore, f is injective. Q.E.D.
b f is surjective  n ( f ( S ) ) = n (T ) .
 Suppose f is surjective.
Then each y  T is the image o an element x  S.
Thereore T  f ( S ) .
But by denition o range and co-domain, f ( S )  T .
Thereore, by double inclusion, T = f ( S )  n (T ) = n ( f ( S ) ).

 Suppose that n ( f ( S ) ) = n (T ) .
We know that f ( S )  T .
But since both f (S) and T have the same number o elements it ollows
that f ( S ) = T  y = f ( x ) or each y  T  surjection. Q.E.D.

In the core syllabus you studied unctions on real numbers.


Consider the unction f :    such that f (x) = x. We can easily
see that this is a bijection.
f ( x1 ) = f ( x2 )  x1 = x2 so it is an injection.
For every x  R, there is an x  R, such that f (x) = x, thereore it
is a surjection. This unction is called the identity function
because it assigns every element to itsel.

Example 8 is another illustration o how the derivative can be


used to check injectivity; parts a and c are a little more challenging.

56 Extension of the concept of function


Example 8
Determine which of the following functions are injective given that in
each case f : R  R.
a f (x) = 3x2 + 7x  2
b f (x) = x5
3x 1
c f( x ) = e 
e2 x

a f (x) = 3x2 + 7x  2 is continuous over R. Check for continuity and use


f (x) = 6x + 7 derivative.
7
 f  ( x )  0 when x  
6
7
and f  ( x ) < 0 when x < 
6
Since f (x) has a turning point, i.e. f (x) is not strictly
increasing or decreasing, it is not injective.
b f (x) = x5 is continuous over R.
f (x) = 5x 4  0 for all x  R.
Hence f(x) is increasing for all x  R which means
that it is injective.
1
c f ( x ) = e3 x  is continuous over R.
e2 x
f  ( x ) = 3 e 3 x + 2 e 2 x > 0 for all x  .
Hence since f (x) is increasing for all x  R,
it must be injective.

Exercise 2A
1 A and B are two non-empty sets, X, Y  A, and f :A  B.
Show that:
a f (X  Y ) = f (X )  f (Y )
b f (X  Y )  f (X )  f (Y )

2 Determine which of these mappings are functions:


m  1
a f :    such that f   =
 n  mn
m  1 1
b f:    =
such that f  +
 n m n
1
c f :    such that f ( m ) =
m

Chapter 2 57
3 Let G denote the set o citizens o Germany. Determine which o
the ollowing statements correctly speciy a unction.
a f : G  G, f (x) is the mother o x
b g : G  G, g (x) is the daughter o x
c h : G  G, h (x) the sister o x

4 The unction f : R+  R+  R+  R+ is defned by


f (x, y) = (x + 2y, x ). Show that fis a bijection.
y
 n

5 P is the set o all polynomials: P =   ai x i| n   , ai    .
 i=0 
Let g :P  P, g ( p) = x 2 p. Determine whether g is
a surjective b injective.

6 Determine which o the ollowing unctions are


a injective b surjective.
i f:  [ 0,  [ , f (x) = e x ii f : [0, 1]  R, f (x) = tan x
 n + 1 if n is odd
iii f :   , f ( n ) = 
 n  1 if n is even
 x
7 Let f : (  + ) 2  (  + ) 2 , such that f ( x , y ) =  xy 2 ,  .
 y
Show that fis a bijection.

8 Determine which o the ollowing unctions, f : +   +  + , are


a injective b surjective
nm ( m + 1 )
i f (n, m) = nm ii f ( n, m ) =
2
iii f (n, m ) = 3 n + 7 m

1  2e  x
9 The unction f : R  R is defned by f ( x ) = .
1 + 2e  x
a Find the range o f.
b Sketch the graph o f.
c Prove that f is an injection.

10 Consider the unctions fand g, defned by


f :    where f (n) = 5n + 4
g : R  R  R  R where g (x, y) = (x + 2y, 3x  5y)
Determine whether:
a the unction fis surjective
b the unction fis injective
c the unction g is surjective
d the unction g is injective.

58 Extension of the concept of function


Composition of functions
If the co-domain of a function g is equal to the domain of a second
The composite
function f, the two functions can be combined. The composition of function (f o g)(x) is
the functions g and f is denoted by f o g. The diagram below also denoted
illustrates this. by f ( g (x))

g f

A B B C

1 a a 

2 b b 

3 c c 

g f

A B C

1 a 

2 b 

3 c 

f g

Note that g (3) = a and f (a) =   ( fo g)(3) = .


We can show that given two functions fand g such that the domain of
f is the co-domain of g, their composition h = f o g is also a function.

Theorem 2
If g : A  B, f: B  C are functions, then f  g : A  C is also a function.

Proof:
Since g is a function we know that for every a  A there is an element
b  B such that g (a) = b.
Since f is a function and b is in the domain of f, we know that there is
an element c  C such that f (b) = c.
Combining the two we obtain that for every a  A there is a c  C such
that ( f  g ) ( a ) = f ( g ( a )) = f (b ) = c , making it a function. Q.E.D.

Chapter 2 59
Example 9

Given that  : R  R such that  (x) = ex and g : R  R such that g (x) = x 2.


a Find
i ( f  g) ( x ) ii ( g  f ) ( x )
b Comment about your results to a i and a ii .
c Check each of the composite functions for injective and surjective properties.
2
a i (  g)(x) =  ( g(x)) =  (x 2) = ex Remember the correct order when working
out composite unctions.
ii ( g  f ) ( x ) = g ( f ( x )) = g (e x ) = (e x ) 2 = e 2 x
2
b e x  e 2x for every x element of R. Composition
of functions is not commutative.
c Method I
Using the result in a i
( f  g ) ( x1 ) = ( f  g )( x2 )
2 2
 ex = ex
1 2

 x1 2 = x2 2
 x1 =  x2
So (  g)(x) is not an injection.
Method II
2
( f  g )  ( x ) = 2 xe x When the unction is continuous and
 ( f  g )  ( x )  0 when x  0 diferentiable it is easier to check by taking
and ( f  g ) ( x ) < 0 when x < 0 the derivative.
Hence (  g)(x) is not continuously increasing or
decreasing since there is a turning point so it is
not injective.
2
Let y  R such that (  g)(x) = y  y = e x > 0
Then for all y  0 there is no x  R such that
(  g)(x) = y
Therefore (  g)(x) is not a surjection.
Using the result in a ii :
( g  f )( x ) = e 2 x
 ( g  f )  ( x ) = 2e 2 x > 0

 g   is a strictly increasing function.


Therefore ( g   )(x) is an injection.

Let y  R such that ( g   )(x) = y  y = e 2 x > 0


Then for all y  0 there is no x  R such that
( g   )(x) = y
Therefore ( g   )(x) is not a surjection.

60 Extension of the concept of function


Example 9 illustrates that composition o unctions is not always
commutative, i.e. ( f g)(x)  ( g  f )(x) or all f (x), g (x).

Inverse functions
Let f : S  T be a bijection. Since it is a surjection, each element in T is
the image o some element in S. But fis also an injection, so every element
in T is the image o a unique element in S. We can thereore defne a new
unction rom T to S that reverses the mapping rom S to T as ollows:
Defnition
Let f : S  T be a bijection rom S to T. The inverse unction o f,
denoted by f 1 : T  S, is a unction such that f f1 = I = f1  f
where I is the identity unction.

Note that f has to be a bijection. I f is not injective then there is some


element in T that is the image o more than one element in S. Let us
say that f (xi) = yi = f (xj ). In this case we cannot assign a unique element
in S such that f 1 ( yi ) = x since yi is the image o two elements in S.
I fis not surjective then there is some element yi  T or which there
is no element in S such that f (x) = yi.

Theorem 3
a A unction f : A  B is bijective  it has an inverse.
b A unction f : A  B is bijective  its inverse is also a bijection.

Proof:
a  : Let f : A  B be a bijection. Then f is injective and f is surjective.
Since f is injective, f (a) = f (b)  a = b or all a, b  A.
Since f is surjective, or every y  B, there is an a  A such that f (a) = y.
Taking these together we have:
fis bijective  or every y  B there is a unique a  A such that f (a) = y.
I we defne a mapping g : B  A, such that g( f (a)) = a or all f (a)  B,
this is a well-defned unction because every element in B may be written
in the orm f (a) and its image a under g is a unique element o A.
Hence we have ( g  f ) (a) = a  g is a let inverse o f.
Also or all f (a)  B, f ( g( f (a)) = f (a)  f g ( f (a)) = f (a).
Thereore g is also a right inverse o f.
Since g is a let and right inverse o f, we can say that f has an inverse.
 : Let g be the inverse o f and let us suppose that f is not injective.
 there are a, b  A such that a  b but f (a) = f (b).
 a = g ( f (a))
= g ( f (b)), since f (a) = f (b)
=b

Chapter 2 61
This is a contradiction since we started by saying that a  b.
Thereore fmust be injective.
Let us now suppose that f is not a surjection.
Then there must be some element y  B that is not the image o any a  A
i.e. f (a)  y or all a  A.
On the other hand f ( g (y)) = y by defnition o inverse.
It ollows that there must be an a  A whose image under f is y in B.
This is a contradiction, thereore fis a surjection.
Since fis both an injection and a surjection it ollows that f is a bijection.
b To show that f 1 : B  A is a bijection we need to show that it is
an injection and a surjection.
Let x1 , x2  B such that f 1 ( x1 ) = f 1 ( x2 ).
Then by the defnition o inverse we know that
x1 = ( f  f  1 )( x1 ) = f ( f  1 ( x1 ) ) = f ( f 1 ( x 2 )) = ( f  f 1 )( x 2 ) = x 2
Thereore f1 : B  A is injective.
Since f : A  B is a surjection we know that or each y  B there is
x  A such that f (x) = y
 f1  f (x) = x = f1 ( y)
Thereore since or all x  A there is y  B such that f1 ( y) = x it
ollows that f1 : B  A is a surjection.
Since f1 : B  A is injective and surjective, it ollows that it is a bijection.

Example 10
Determine whether f is an injection, and/or a surjection and fnd the inverse
unction f 1 where applicable:
a f : R  R+  {0} and f (x) = x 2
b f : R+  R+ and f (x) = x 2 + 1
c f : R  R and f (x) = x 3 + 1

a Method I
Let f (x1 ) = f (x2)
 x1 2 = x2 2
 x1 =  x2
Thereore fis not an injection.
Method II
f( x ) = x 2  f( x ) = 2 x
 f  ( x ) < 0 when x < 0
f  ( x ) = 0 when x = 0
and f  ( x ) > 0 when x > 0
 is not continuously increasing or decreasing over R.

62 Extension of the concept of function


Thereore is not an injection.
For every y  R+ there is x  R such that x2 = y.  (x) > 0 or all x  R+.
Thereore is a surjection.
Since is not a bijection it does not have an inverse.
b Let  (x1 ) =  (x2) Use contrapositive statement or
2
 x1 + 1 = x2 + 12 injection.
 x1 = x2
Thereore is an injection.
Let y   + , y < 1 Use defnition o surjection.
There is no x  R + such that  (x) = y.
Thereore is not a surjection.
Since is not a bijection it does not have an inverse.
c Let  (x1 ) =  (x2)
 x1 3 + 1 = x2 3 + 1
 x1 = x2
Thereore is an injection.
For every y  R, 3 y  1    f ( 3 y  1 ) = y
So is a surjection.
The inverse unction is given by f 1 ( x ) = 3 x  1

As mentioned previously, unctions are not restricted to having the


domain or co-domain as subsets o R. In Example 7 the domain and
co-domain were the set o polynomial unctions with real coefcients.
In the next examples you will see work on unctions that have a
Cartesian product as domain and co-domain.

Example 11

Given  : R  R  R  R such that  (x, y) = ( y  2x, x + y),


a show that  is a bijection
b nd  1 (x, y).

a Let  (x, y) =  (a, b) First show that  is an injection.


 ( y  2x, x + y) = (b  2a, a + b)
 y  2x = b  2a Equate corresponding elements.
and y + x = b + a
 3x = 3a Subtract the second equation
 x= a rom the frst.
Since y + x = b + a then y = b.
Thereore is injective. Show that (x, y) = (a, b)

Chapter 2 63
Let (a, b)  R  R. Now show that  is a surjection.
I  (x, y) = (a, b), ( y  2x, y + x) = (a, b)
 y  2x = a and y + x = b Equate corresponding elements.
 3x = a  b Subtract second equation rom
ba the frst.
 x= 
3
y+ x= b
y= bx Substitute or x
ba
 y =b
3
2b + a
 y= 
3
Thereore (x, y)  R  R.
So  is a surjection, and thereore  is a bijection.
b Let (a, b) =  (x, y)
 (a, b) = ( y  2x, y + x)
 a = y  2x ba 2b + a
 x= , y= Replace a and b by x and y
b=y + x  3 3
respectively to write the inverse
 y  x 2y + x  unction.
Thereore f 1 ( x , y ) =  ,  You need to write the inverse
 3 3 
unction using 1 (x, y).

Example 12
g :  2   2 , g ( x, y ) = ( x  y, 2 x + y )
Given that
h :  2   2 , h ( x, y ) = ( xy, 2 x  y )
a Show that g has an inverse and fnd it.
b Determine whether ( g o h) is a bijection.
a For g to have an inverse it has to be a bijection.
Let g (x, y) = g (a, b)
(x  y, 2x + y) = (a  b, 2a + b)
 x y= a b Equate elements o ordered pairs.
and 2x + y = 2a + b Add the equations.
 3x = 3a  x = a Show that (x, y) = (a, b).
x y= a b y= b
Thereore g is an injection.
Let (a, b)  R2
I  (x, y) = (a, b) then
(x  y, 2x + y) = (a, b)

64 Extension of the concept of function



and 2x + y = b }
x  y = a  x = a + b  R and y = b  2a  R
3 3
Equate elements o ordered pairs
and add to solve or x and y.
Therefore g is a surjection.
Since g is bijective it has an inverse.
Let (a, b) = g (x, y)
Then (a, b) = (x  y, 2x + y)
 a=xy  a+b b  2a
 x = , y= Equate elements o ordered pairs
and b = 2 x + y  3 3
and add to solve or x and y.
 x + y y  2x 
g  1 ( x, y ) =  ,  Again we need to write it out
 3 3 
using inverse notation.

b ( g  h ) ( x, y ) = g ( h ( x, y )) = g ( xy, 2 x  y )
= ( xy  2 x + y, 2 xy + 2 x  y )
Let ( g o h)(x, y) = ( g o h)(a, b)
 (xy  2x + y, 2xy + 2x  y) = (ab  2a + b, 2ab + 2a  b)
 xy  2 x + y = ab  2 a + b 
 xy = ab Equate elements o ordered pairs
2 xy + 2 x  y = 2 ab + 2 a  b 
and add to write x in terms o a,
ab b and y
xy = ab  x = , y 0
y
2ab
 ab  + y = ab  2a + b Substitute or x in the rst
y
equation.
  2 ab + y 2 =  2 ay + by
 y 2 + ( 2 a  b ) y  2 ab = 0

( 2a  b )  ( 2 a  b ) 2 + 8 ab
 y= Solve the quadratic equation or y.
2
( 2a  b )  4 a 2  4 ab + b 2 + 8 ab
 y=
2
( 2a  b )  ( 2 a + b ) 2
 y=
2
y=b or y = 2a
b
When y = b, x = a and when y =  2 a , x =
2
We have (g o h)(x, y) = (g o h)(a, b)
 ( g o h)(2, 6) = ( g o h)(3, 4) It is sucient to nd two diferent
elements that have the same
( g o h) is not injective therefore it is not a bijection.
image.
e. g. a = 2, b = 6

Chapter 2 65
2.2 Properties of functions
We will now consider some properties o unctions by looking at the
three unctions:
f : R  R, f (x) = 3x + 2
2
g :    + , g( x ) = e x

1
h :  +   + , h( x ) =
x
First we fnd f ( g  h)(x) as ollows
1   1 
f  ( g  h ) ( x ) = f  g( h ( x )) = f  g   = f g 
 
x   x 

( ) = f (e
2
1 
 
= f e x  x 2 ) = 3e x
2
+ 2

Now let us compute ( f g)  h (x).


2 2
( f  g )( x ) = f ( g ( x )) = f (e x ) = 3e x + 2
2
1 
1    2
 ( f  g )  h ( x ) = ( f  g )   = 3e  x  + 2 = 3e x + 2
 
x

Thereore or the given unctions we have shown that


(( f g)  h)(x) = ( f ( g  h))(x).
In the next theorem we will prove that this result is true or any three
well-defned unctions.

Theorem 4
Composition o unctions is associative; in other words,
given three unctions f : C  D, g : B  C and h : A  B, it ollows
that (( f g)  h)(x) = ( f ( g  h))(x).

Proof:
LHS = (( f g)  h)(x) = ( f g) (h (x)) = f ( g (h (x))
RHS = ( f ( g  h))(x) = f (( g  h)(x)) = f ( g (h (x))
The domains and co-domains o (( f g)  h) and ( f ( g  h)) are the
same and since (( f g)  h)(x) = ( f ( g  h))(x) or all x  A, it ollows
that (( f g)  h) = ( f ( g  h)). Q. E. D.

66 Extension of the concept of function


Investigation
Beore we look at the next properties you should justiy whether the ollowing
statements are true or alse. A ormal proo is not necessary at this stage and
you may use diagrams to help you decide on an answer.
f : A  B and g : B  C
a i Given that both f and g are injective unctions then g  f is also injective.
ii Given that both f and g are injective unctions then f g is also injective.
iii I g  f is injective and g is also injective then f is injective.
iv I g  f is injective and f is also injective then g is injective.

b i Given that both f and g are surjective unctions then g  f is also surjective.
ii Given that both f and g are surjective unctions then f g is also surjective.
c I f is injective and g is surjective then:
i g  f is injective
ii g  f is surjective.

d I f is surjective and g is injective then


i g  f is injective
ii g  f is surjective.

Theorem 5
a I f: A  B and g : B  C are injective unctions then
g  f: A  C is also injective.
b I f: A  B and g : B  C are surjective unctions then
g  f: A  C is also surjective.
c I f: A  B and g : B  C are bijections then
g  f: A  C is also a bijection.

Proof:
a Let ( g  f )(x1 ) = ( g  f )(x2)
 g ( f (x1 )) = g( f (x2))
 f (x1 ) = f (x2) since g is injective
 x1 = x2 since f is injective
Thereore ( g  f ) is also injective.
b Let q  C, then, since g is surjective, there is some y  B such that g ( y) = q.
For this y there is some x  A such that f (x) = y since fis surjective.
So q = g ( y) = g ( f (x)) = ( g  f )(x)
Thereore ( g  f )is also surjective.
c Since it was shown in a and b that ( g  f ) is both injective and surjective,
then by defnition o bijection it ollows that i fand g are both bijective
then ( g  f ) is a bijection. Q.E.D.

Chapter 2 67
Properties of composite functions
 Associative ( ( f  g )  h ) ( x ) = ( f  ( g  h ) ) ( x )
 I f : S  T and g : T  S are injections, then
( f  g ) ( x ) and ( g  f ) ( x ) are injective.
 I f : S  T and g : T  S are surjections, then
( f  g ) ( x ) and ( g  f ) ( x ) are surjective.

Exercise 2B
1 A and B are two non-empty sets, and A, B  R. The unctions f and g are
defned as ollows: f : A  B  B  A, f (a, b) = (b, a) and g : B  A  B,
g (b, a) = b. Find g o f.

2 Explain why the unction f : R  R, f (x) = 2x  x2 is neither injective


nor surjective.

3 Given that f : R  R+ , f (x) = e2x and g : R+  R, g (x) = lnx, fnd:


a i ( f o g) (x)
ii ( g o f ) (x)
b Check each o the composite unctions in a or injection
and surjection.

4 Two unctions f and g are defned as ollows:


f : Z  Z  Z, f (n) = (n  1, 1) and
g : Z  Z  Z, g (m, n) = m + n
a Show that fis a bijection and fnd its inverse.
b Show that g is not a bijection, but a surjection.
c Find fo g and g o f.

5 Consider the two unctions f : R  R  R  R such that


f (x, y) = (xy, x + y).
a Determine whether or not fis a bijection.
b Find ( fo f )(x, y).

1
6 Let f : R \ {0, 1}  R \ {0, 1} such that f ( x ) = and
x
g : R \ {0, 1}  R \ {0, 1} such that g (x) = 1  x
a Show that f and g are both bijections.
b Find fo g and g o f.
c Show that ( fo g) o ( g o f )(x) = ( g o f ) o ( fo g)(x).
d What can you say about fand g ?
e What can you say about fo g and g o f ?

68 Extension of the concept of function


The unction f : (R+ ) 2  (R+ ) 2 is defned by f ( x, y ) =  , x 2 y .
y
7
x 
a Show that fis a bijection.
b Find the inverse f1 .

8 The unction f : [0,  [  [1,  [ is defned by f (x) = 4e2x  3.


a Find f (x) and hence show that f is a bijection.
b Find an expression or f1 (x).

9 The unction f : R  R is defned by


x
 for x  e
f( x ) =  e
 ln x for x > e
a Sketch the graph o f.
b By reerring to your graph, show that f is a bijection.
c Find f1 .

10 Three unctions mapping Z  Z  Z are defned by


f (m, n) = m  n   , f2(m, n) = | n| and f3 (m, n) = m2  n2.
Two unctions mapping Z  Z  Z are defned by
g ( p) = (2p +  , p) and g2( p) = (| p| , p).
a Find the range o
i f1 o g1
ii f3 o g2
b Find all the solutions o f1 o g2 ( p) = f2 o g1( p).
c Find all the solutions o f3 (m, n) = k in each o the cases
k = 1 and k = 2.

11 Consider the unctions


n
 if n is even
f : Z  Z where f ( n ) =  2
 n +1 if n is o dd

g : Z  Z where g (n) =  6  n
h : Z  Z where h (n) = n(mod 8)
k : Z  Z where k (n) = | n  8|
Find:
a ( h o g )(n) b ( k o f )(n) c ( fo g )(n)
d ( fo h o g )(n) e ( k o h o g )(n) f ( k o f o g )(n)
2
12 Given that f : [1, [  R, f (x) = ln(2x  1), g :    + , g ( x ) = e x and
h : R  R, h (x) = 2x, fnd the ollowing unctions:
a ( g o f )(x) b ( fo g)(x) c (h o f )(x)
d ( g o h o f )(x) e (h o g o f )(x)

Chapter 2 69
Identity functions
In this section we will be focusing on identity functions. An identity
function is one whose output is the same as the original input.

Defnition
The identity unction for a set S is a bijection IS : S  S such that
IS (x) = x for all x  S.

Theorem 6
Let f : S  S be any function.
Then (IS o f )(x) = ( fo IS )(x) = f (x) for all x  S.

Proof:
Let x  S.
Then (IS o f )(x) = IS( f (x)) = f (x) and ( fo IS)(x) = f (IS(x)) = f (x).
Therefore (IS o f )(x) = ( fo IS )(x) = f (x). Q.E.D.

Theorem 7
For a bijection f : S  T such that f (x) = y, x  S and y  T, the
inverse function f1 : T  S is such that ( f 1 o f )(x) = IS and
( fo f1 )( y) = IT.

Proof:
Notice that one composition
( f1 o f )(x) = f 1 ( f (x)) = f 1 ( y) = x = IS gives an identity on the
( fo f1 )( y) = f ( f1 ( y)) = f (x) = y = IT Q.E.D. domain and the other one
gives an identity on the
co-domain.

Example 13
Show that if f : B  C and g : A  B are both bijections then ( fo g) 1 (x) = ( g1 o f1 )(x).
( fo g) o ( g1 o f1 )(x)
= ( fo ( g o g1 ) o f1 )(x) Composition of functions is
= ( fo (IB o f1 ))(x) associative.
= ( fo f1 )(x) = IC
Similarly
( g1 o f 1 ) o ( fo g)(x)
= ( g1 ( f 1 o f ) o g)(x)
= ( g1 (IB o g))(x)
= ( g1 o g)(x) = IA

70 Extension of the concept of function


Example 14
4 x | x|  x + | x|
Given that f :  \{ 0}   + such that f ( x ) = ,
2x
show that f (x) is a bijection.
4 x2  x + x Show that f is both injective and
When x > 0, | x| = x and f( x ) = = 2x
2x surjective.
So f (x1 ) = f (x2)  x1 = x2.
4 x 2  2 x
When x < 0, | x| = x and f ( x ) = =  2x  1
2x
and f (x1 ) = f (x2)  x1 = x2.
Moreover, if x < 0 then f (x) is even, while if x > 0
then f (x) is odd. So if x1 > 0 and x2 < 0 then
f (x1 )  f (x2). Therefore f (x) is an injection.

f (Z+ ) = {2, 4, 6, . . .}
f (Z ) = {1 , 3, 5, . . .}
f (Z) = {2, 4, 6, . . .}  {1 , 3, 5, . . .} = Z
Therefore f (x) is a surjection.
Since the function is both an injection and a surjection
it follows that it is a bijection.

Although R and R+ have the same cardinality and are both infnite they are
uncountable, unlike ,  and Z, which are countably infnite.
Cantor came up with an ingenious yet very simple method to show 0 1/1 1/1 2/1 2/1 3/1 3/1 ...
that the rational numbers are countable. A set is said to be 1/2 1/2 2/2 2/2 3/2 3/2 ...
countable i a one-to-one correspondence can be ound between
1/3 1/3 2/3 2/3 3/3 3/3 ...
the elements o the set and the set o positive integers. Cantor
1/4 1/4 2/4 2/4 3/4 3/4 ...
constructed a table that enables all the rational numbers, both
1/5 1/5 2/5 2/5 3/5 3/5 ...
positive and negative, to be included and hence allows a one-to-one
1/6 1/6 2/6 2/6 3/6 3/6 ...
correspondence to be ound. The table is on the right, with
...

...

...

...

...

...

the lines showing the order o pairing up the ractions with the
positive integers.

Having understood what we mean by a bijection it should be clear that


i we have a bijection f : S  T where S and T are fnite sets it ollows that
n (S ) = n (T ).
What i g : R  R+ such that g(x) = 2 x? You have shown in Exercise 2B that
this is a bijection. In this case we say that the two sets R and R+ have the
same cardinality because there is a bijection g : R  R

Chapter 2 71
Exercise 2C
1 For each o the ollowing questions nd ( fo g)(x) and ( g o f )(x)
and determine whether fand g are mutual inverses.
a f :   , f ( x ) = 1  3 x
x
g :   , g( x ) = 1 
3
1
b f :  \ {0}   \ {  4} , f ( x ) =  4
x
1
g :  \ { 4}   \ {0} , g( x ) =
x + 4

c f :    , f ( x ) = 3 kx  1 , k   +

1 3
g :   , g( x ) = ( x + 1)
k

2 Show that the ollowing unctions are bijective and describe the
respective inverse unctions:
a f : R+  R, f (x) = lnx
 x if x is rational
b f :   , f ( x ) = 
  x if x is irrational
If a function is
3 a Given that f : R  R, f (x) = ex and invertible it means
g:[0, ]  [1, +1], g (x) = cos x nd: that it has an inverse.
i ( fo g) 1(x)
ii ( g 1 o f1)(x)
b Prove that or invertible unctions f and g, ( fo g) 1 = g1 o f1.

2.3 Binary operations


You are amiliar with the operations o addition and multiplication
o numbers, the dot product o two vectors, the union and intersection
o sets and earlier in this chapter we looked at the composition o
unctions. All o these are operations. Other operations you are
amiliar with include:
n! actorial
| z| modulus
A the complement o set A
There is a diference between n! and the product o two numbers.
In order to nd n! we need to know only the value o n. So when
n = 4, 4! = 24. We call these unary operations (operations that have
only one input). However, in order to perorm multiplication we
need two numbers. We need two sets to nd a union or intersection
but we only need set A to identiy its complement.

72 Extension of the concept of function


Defnition In this book we mostly
use * to denote binary
A binary operation  on a non-empty set S is a rule or combining
operations. However,
any two elements x, y  S to give a unique element c o a set. This
we sometimes also
is denoted by x  y = c. use other notations,
or example:
Division on R is not a binary operation because x  0 is not defned.
o, , #, , 
However division on  \{ 0} is a binary operation.
Multiplication on Z produces another integer. The dot product o two
vectors is not a closed binary operation since it produces a scalar quantity.
The vector product however is a closed binary operation since it produces
another vector.
Consider a set S with binary operation . We say that S is closed
under  i or every x, y  S, x  y  S.
The set o vectors, V, is closed under the vector product because or
all (a, b)  V, a  b = c  V.
The set o real numbers, R, is closed under multiplication because
or all x, y  R, xy  R.
When testing or closure on fnite sets it is useul to illustrate the
operation using a Cayley table. This is a square grid which shows
all the possible elements obtained by the binary operation. We can
represent the operation  on the set A = {a, b, c , d } as ollows:

AA a b c d Arthur Cayley (18211895) was


the frst mathematician to defne
a a a a b a c a d the concept o a group (which
b b a b b b c b d you will frst study in Chapter 3)
as a set together with a binary
c c a c b c c c d
operation that satisfes certain
d d a d b d c d d conditions.

Note that order is important when flling out a Cayley table. The element in
the third row and second column above is c  b and not b  c.

Consider the binary operation multiplication on the set S = {1 , 0, 1 } .


The operation table is shown below.
S S 1 0 1 We can see rom the Cayley table that every
1 1 0 1 product is a member o S.
0 0 0 0 We can thereore say that S is closed under
1 1 0 1 multiplication.

Chapter 2 73
Example 15
Determine which o the ollowing operations are binary operations on the given sets and
or those which are, state whether or not they are closed.
a Addition on the set S = {1 , 0, 1 }
b Multiplication on the set o complex numbers 
c Addition on the set A = { x| x = 2 n , n   }
+

d Multiplication on the set B = { x| x = 2 n + 1, n   }


+

e Division on the set o rational numbers 

a
+ 1 0 1
1 2 1 0
0 1 0 1
1 0 1 2
It is a binary operation since each addition gives a unique Check whether it is a
element. Not closed, since 2, 2  S binary operation and or
b It is a binary operation which is closed. closure.
(a + ib)(c + id ) = (ac  bd ) + i(bc + ad )  
c It is a binary operation which is closed on the set o positive
even integers. 2m + 2n = 2(m + n), which belongs to A since
(m + n) is an element o Z + .
d It is a binary operation which is closed on the positive odd integers.
(2n + 1 )(2m + 1 ) = 2(2mn + m + n) + 1 = 2k + 1  B
e It is not a binary operation since 0   , and division by 0 is
not defned

Example 16
The operations  and o on the set S = {1 , 2, 3} are defned as ollows:
| a  2 b| + a
a  b = a b  b a and a  b =
2
Draw a Cayley table or each operation and determine whether the set is closed
under these operations.

 1 2 3
Fill out the table by working out the
1 0 1 2
operation, e.g. 2  1 = 21  12 = 1.
2 1 0 1
3 2 1 0
The set is not closed under  since 2 and 1 are not in S.
o 0 1 2 3
Work out the operations to fll out
0 0 1 2 3
the table, e. g.
1 1 1 2 3
 3  4 + 3
2 2 1 2 3 3 2 = = 2.
2
3 3 2 2 3
The set is closed under o.

74 Extension of the concept of function


Exercise 2D
1 Determine which o the ollowing operations are binary operations
on the given sets and or those which are, state i they are closed.
a  on S = {0, 1, 2, 3} , where a  b = a + b
b  on Z + , where a  b = the smaller o a or b, or the common value i a = b
c  on Z + , where a  b = (ab + 1)
d o on Z + , where a o b = b a

2 Let S = {z| z = a + bi, where a, b   , b  0, i = 1 }. Show that


the ollowing operations are binary operations on S and determine
whether or not they are closed.
a addition b multiplication c division

3 Determine whether or not the ollowing sets are closed under


a addition
b multiplication
i {
A = m| m = 2 n , n   + } ii B = { m| m = 2 n  1, n   + }

4 The operations  and o on the set S = {0, 1, 2, 3} are defned


as ollows: a  b = a + b (mod4) and a o b = ab (mod4).
Draw a Cayley table or each operation and determine whether
or not the set is closed under these operations.
5 Let X = { f | f : R  R, fis a unction} . Show that the
ollowing operations are binary operations on X and determine
whether or not they are closed.
a addition o unctions b subtraction o unctions
c composition o unctions

6 Let S = {1, 1, i, i } where i = 1. Draw a Cayley table to


show that S is closed under multiplication.
7 The binary operation  is defned or a, b  Z + by a  b = 2a + b + ab.
Show that  is a binary operation and determine whether or not Z+ is
closed under .
8 The operations  and o on the set S = {1, 2, 3} are defned as ollows:
a!b!
a  b = ab  b a + ab and a  b =
ab
Draw a Cayley table or each operation and determine
whether or not the set is closed under these operations.
9 Let S = { n 2 | n   + } . Determine whether or not S is closed under
a addition b multiplication.
10 Let S = {1, 2}. The binary operation  is defned on S as ollows.
For a, b  S, a  b = 3ab and the binary operation o is defned on
S  S is defned as (x1, y1) o (x2, y2) = (x1  x2, y1  y2).
a Write the elements o S  S.
b Construct the Cayley table or the operation  on S. Is S closed under ?
c Construct the Cayley table or the operation o on S  S.
Chapter 2 75
Properties of binary operations
Defnition
A binary operation  on a non-empty set S is said to be
associative i or all a, b, c  S, a  (b  c) = (a  b)  c.
The operation addition on R is associative but subtraction is not
since (6  2)  3 = 1 and 6  (2  3) = 7.
Also the operation multiplication on R is associative but division
on R \ {0} is not associative because 8  (1 2  3)  (8  1 2)  3.

Defnition
A binary operation  on a non-empty set S is said to be
commutative i or all a, b  S, a  b = b  a.
Addition and multiplication are commutative on R but the operation
division on R \ {0} is not commutative because it is not the case that
a  b = b  a, or all a, b  R\{0}.
Also the operation subtraction on R is not commutative since it is
not the case that a  b  b  a, or all a, b  R\{0}.
Example 17
The binary operation o on  is defned as ollows z o w = | z + w| . Determine whether o is:
a commutative b associative.

a z = a + ib , w = c + id Check whether
zo w=wo z
 | z + w| = ( a + c ) 2 + (b + d ) 2
= | w + z|
z o w = | z + w|
w o z = | w + z| = | z + w|
The operation is commutative.
b Method I
(1) o ((1) o 1) = (1) o 0 = 1, but ((1) o (1) o 1 = 2 o 1 = 3
Method II
| z +| w + v| | = a + ( (c + e )2 + (d + f )2 ) + ib Check whether
z o (w o v) = (z o w) o v
|| z + w| + v| = ( )
( a + c ) 2 + ( b + d ) 2 + e + if

| z + | w + v||  || z + w| + v|
z o (w o v) = z o (| w + v| )  (| z + w| ) o v
The operation is not associative.

Defnition
Given two binary operations  and o on a set S,  is said to be
distributive over o i a  (b  c ) = ( a  b )  ( a  c ) and
( a  b )  c = ( a  c )  (b  c ) or all a, b, c  S .

76 Extension of the concept of function


The ollowing example illustrates this property.
Example 8
Given the operations  and o on Z, such that a  b = 3 ab and a  b = a + 3b ,
determine whether:
a  is distributive over o b o is distributive over 

a a  (b  c ) = a  (b + 3 c ) = 3 a (b + 3 c ) = 3 ab + 9 ac Check if  is distributive over o


(a  b) o (a  c) = (3ab) o (3ac) = 3ab + 3(3ac)
= 3ab + 9ac
Thereore  is distributive over o
b a  (b  c ) = a  (3bc ) = a + 9 bc
Check if is o is distributive over 
(a o b)  (a o c) = (a + 3b)  (a + 3c)
= 3(a + 3b) (a + 3c)
a o (b  c)  (a o b)  (a o c)
For example, 5 o (5  5) = 5 o 75 = 230, but
(5 o 5)  (5 o 5) =  200.
Thereore o is not distributive over .

Exercise 2E
1 For the binary operations defned below determine whether  is
i commutative ii associative
a  is defned on Z by a  b = a  b
b  is defned on  by a  b = 2 ab + 1
a b
c  is defned on  by a  b = 2 3

2  is a commutative and associative binary operation on a set S.


Show that ( a  b )  ( c  d ) = ( ( d  c )  a )  b .
3 Let f1 , f2 , f3 and f4 be unctions defned on R \ {0} such that f (x) = x,
1 1
f2 ( x ) = , f3 (x) = x and f4 ( x ) =  . The binary operation o on
x x
S = {f , f2, f3 , f4} is defned as the composition o unctions.
Draw a Cayley table to illustrate this operation. Determine whether
a S is closed under composition o unctions
b the operation composition o unctions is commutative in S.

4 The binary operation  is defned on R such that or all


a, b   , a  b = a + 2b  1. Determine whether the binary
operation  is
a commutative b associative.

5 The operation  is defned on R \ {1} by a  b = ab  a  b + 2 or


all a, b  R \ {1} . Show that:
a R \ {1} is closed under the operation 
b the operation  is commutative
c the operation  is associative.

Chapter 2 77
The identity element e
Defnition
Let  be a binary operation on S. I there is an element e  S such
that or every element x  S , e  x = x = x  e, then we say that e
is the identity element o S under the operation .

Theorem 8
The identity element o a binary operation  on S is unique.

Proo:
Let e, f  S, e  f such that or every element x  S :
e  x = x = x  e and f  x = x = x  f
e  x = x = x  e  e  f = f = f  e (replacing x by  since f  S)
f x = x = x f f e = e =e (replacing x by e since e  S)
Combining the two we obtain the result that e = and so the identity is unique. Q.E.D.

In general we could say that an element e  S is the let identity i


or every x  S, e  x = x, and  S is the right identity i or every
x  S, x = x  . We can show that e = as ollows. (i.e. i there is a
let identity and there is also a right identity, then they are equal.)
Since e is a let identity we know that e  x = x.
But since  S we can replace x by  to obtain e  = .
But is a right identity so e  = e.
Thereore e = .

Example 19
ab
Let the binary operation  be defned on set o numbers S such that or a, b  S, a  b = .
2
Determine whether or not an identity exists and i it does, fnd it.
Suppose an identity exists, i.e. e  b = b or b  S .
eb First we fnd the let identity.
e  b = b and e  b =
2
eb
Thereore b = e=2
2
be
b  e = b and b  e = Now we fnd the right identity.
2
be
b =  e = 2
2
Since the let identity is equal to the right identity the
identity exists and e = 2.

78 Extension of the concept of function


Example 0

Let the binary operation  be defned on  + such that or a, b  S, a  b = 2 a + 3b .


Determine whether or not an identity exists and i it does, fnd it.
Suppose an identity exists, i.e. e  b = b or b  S
e  b = b and e  b = 2e + 3b First look for the left identity.
 b = 2e + 3b
 e = b
We can show that the right identity is not equal to
the let identity as ollows:
b  e = b and b  e = 2b + 3e
 b = 2b + 3e
b
e=
3
Since the let identity is not equal to the right identity and
neither let identity nor right identity are elements o  + it
ollows that the operation does not have an identity in S.

Exercise 2F
In Questions 1 to 5 below, determine whether the binary operation  is:
a commutative
b associative.
Determine whether or not an identity element exists and i it does, fnd it.
1 The binary operation  is defned on Q such that or all a, b  Q,
a  b = a + b  ab.

2 The binary operation  is defned on N  N such that or all


(m, n), ( p, q)  N  N, (m, n)  ( p, q) = (mp, nq).

3 The binary operation  is defned on N  N such that or all


(m, n), ( p, q)  N  N, (m, n)  ( p, q) = (m + p, n + q).

4 The binary operation  is defned on Q  Q such that or all


(a, b), (c, d )  Q  Q, (a, b)  (c, d ) = (ac, ad + b).

5 The binary operation  is defned on  +   + such that or all


(m, n), ( p, q)  +   + , (m, n)  ( p, q) = (mq + np, nq).

The inverse of an element


Defnition
Let  be a binary operation on S with identity e. Then or all
x  S i there exists an element y  S such that x  y = e = y  x
then we call y the inverse o x, and we write y = x 1 .

Chapter 2 79
It is easy to understand this concept with some operations that you are
amiliar with. The identity o addition in R is 0 because x + 0 = x = 0 + x
or all x  R. Since x  R and x + (x) = (x) + x = 0 we conclude
that or addition in R, e = 0 and x 1 = x.
Note that x1 here is
Similarly or multiplication in R\{0} , the identity e = 1 the notation or inverse,
1 not the reciprocal
since 1  x = x = x  1 and the inverse is given by x 1 = since
x notation or numbers,
1  1 
x  =    x = 1. which happens to be
x x the same.

Theorem 9
For an associative binary operation  in S with identity e, the
inverse is unique.

Proof:
You should remember
Let a, b be inverses o x. that you cannot discuss
an inverse without
a= e a by defnition o identity
frst establishing that
= (b  x)  a since b is an inverse o x an identity exists. We
also need to assume
= b  (x  a) by associativity
the associativity
=be since a is an inverse o x property but not
commutativity.
=b by defnition o identity Q.E.D.

Example 21

For multiplication in  \ {0} determine whether or not the identity element exists
and i it does fnd the inverse o z   \ {0} .

We know that or z   \ {0} , 1  z = z  1 = z  e = 1 + 0i. Establish whether there is


1 1 an identity.
For z   \ {0} , z  =  z = 1.
z z
z = a + bi
1 1 a  bi z
 = = 2 2
= Find the inverse.
z a + bi a + b zz 
z
Thereore the inverse z 1 = .
zz 

80 Extension of the concept of function


Example 22

Let  be a binary operation on   (\{0} ) such that (a, b)  (c, d ) = (a + c, bd ).


a Show that the operation is
i associative ii commutative.
b Show that the identity exists and fnd the inverse (a, b) 1 under  in   (\{0} ).

a i Associativity: When checking the


(a, b)  ((c, d )  (m, n)) = ((a, b)  (c, d ))  (m, n) properties you should
LHS frst defne them and
= ( a, b )  ( ( c , d )  ( m , n ) ) then prove that the
= (a, b)  (c + m, dn) RHS = LHS.
= (a + (c + m), b(dn))
= ((a + c) + m, ( bd )n)
= ((a + c), bd )  (m, n)
= ((a, b)  (c, d ))  (m, n) = RHS
Thereore  is associative in   (\{0} ).
ii Commutativity:
(a, b)  (c, d ) = (a + c, bd )
= (c + a, db)
= (c, d )  (a, b)
Thereore  is commutative in   (\{0} ).
b Since we have shown that  is commutative there is no need to
fnd both let and right identities since they will be equal.
Let (x, y)    (\{0} ) such that (x, y)  (a, b) = (a, b)
 (x + a, yb) = (a, b)  x + a = a  x = 0 and yb = b  y = 
So the identity is (0,  ).
Let (m, n)    (\{0} ) such that (m, n)  (a, b) = (0,  ).
(m + a, nb) = (0,  )
1
 m =  a, n =
b
 1
Thereore (a, b)  =  a,  .
 b

The cancellation laws


The cancellation laws are very important in abstract algebra and will
An invertible binary
be used extensively in the next chapters. operation on S is
one where for every
Theorem 10 x element of S, there
exits an inverse also
Let  be an invertible associative binary operation on a non-empty
in S.
set S with identity e. Then the operation satisfes
i the left cancelation law, i.e. a  b = a  c  b = c
ii the right cancelation law, i.e. i a  b = c  b  a = c

Chapter 2 81
Proof:
Let a  b = a  c, where a, b, c  S.
a  (a  b) = a  (a  c), since a has a unique inverse in S
 (a  a)  b = (a  a)  c, since the operation is associative
 e  b = e  c by defnition o the inverse element
 b = c by defnition o the identity element Q.E.D.
The proo o the right cancellation law is let as an exercise.

Example 

Show that both right and let cancellation laws are satisfed or the composition
o bijective unctions.
Let f, g, h be bijections. We need to show that
a i f g = f h then g = h
b i f g = h  g then f= h
Since f is a bijection it is invertible, i.e. there exists
a bijection f  such that f f = I = f   f
f g = f  h
 f  ( f g) = f  ( f h)
 ( f  f )  g = ( f  f )  h Composition of functions is associative.
 I g = I h Inverse property.
 g= h Identity property.
Thereore the let cancellation law holds.
It is let as an exercise to prove part b.

Exercise 2G
1 The binary operation  is defned on R as ollows. For any a, b  R
ab=a+b+1
a Show that  is commutative. b Find the identity element.
c Find the inverse o the element a.

2 Consider the binary operation multiplication on the set C \ {0} .


a Show that multiplication is commutative.
b Show that multiplication is associative.
c Find the identity element under multiplication.
d Find the inverse o the element a + bi under multiplication.

3 Consider the set A = {0, 1, 2, 3} under the binary operation  such that
or a, b  A, a  b = a + b (mod 4). Construct a Cayley table to illustrate
this binary relation and show that the relation is commutative. Identiy the
identity element and hence fnd the inverse o each element in A.

82 Extension of the concept of function


4 For each o the ollowing sets,  represents a closed binary operation
defned on the given set S. Determine whether or not the identity element
exists. I it does, fnd it and the inverse o a  S.
a S = {2, 4, 6, 8} , a  b = ab (mod 10)
ab
b S = Q \ {0} , a  b = c S =  + , a  b = 2 + ab
2

5 Consider the binary operation multiplication on the set S = {2 n| n  } .


a Show that
i S is closed under multiplication
ii multiplication is associative
iii an identity exists
iv every element in S has an inverse in S

a+ b
6 Given the set S = ]1, 1[ and the operation a  b = ,
1 + ab
a show that
i S is closed under 
ii  is associative
iii an identity exists.
b Find the inverse o a  S under .

Review exercise
EXAM-STYLE QUESTIONS
1 S = {1, 2, 3, 4, 5, 6} and the unction f : S  S is defned by
f (x) = 6x(mod 7)
a Prove that fis a bijection.
b Show that fis its own inverse.

2 Defne the operation  on the sets A and B by A  B = AB.


Show algebraically that
a A  A = A b (A  A)  (B  B ) = A  B c (A  B )  (A  B ) = A  B.
 
3 Let f : A  B where A = [0, [   0, , B = [ 0, [  [ 0, 1[ and
 2 
f (x, y) = (x cos y, sin y). Determine whether fis a bijection.
I it is, fnd the inverse unction f 1 .
4 The operation  is defned on    as
(a, b)  (c, d ) = (ac + bd, ad + bc), where a, b, c, d  .
Find the identity element or this operation.
5 Consider three sets S, T and U.
f and g are two mappings such that f : S  T, and g : T  U.
i I g  f is surjective, prove that g is surjective
ii I g  f is injective, prove that f is injective.

Chapter 2 83
1
6 The unction f : R  R is defned by f ( x ) = 3 cos x + .
6
a Determine whether or not the unction is injective or surjective,
giving reasons.
b I the domain is restricted to [0, ], what are the restrictions on the
co-domain that would make f invertible? Find the inverse unction.
7 Let  be the binary operation on the set S = { x  1 < x < 1, x  } defned by
x + y
x y= , for any x , y  S .
1 + xy
a Determine whether or not the operation  is
i commutative ii associative.
b Establish whether or not an identity exists and i so fnd it.

8 The unction f : R  R is defned by f (x) = e 2cosx + 1


a Find the exact range o f.
b i Explain why fis not an injection.
ii Giving a reason, state whether or not fis a surjection.
c A new unction g is now defned as ollows:
g : [0, k]  A where g (x) = e 2cosx + 1 and k  0.
i Find the maximum value o k or which g is an injection.
For this value o k, what values can A take to make g (x) a bijection?
ii Find an expression or g 1 (x).
iii Write down the domain o g 1 .

Chapter 2 summary
A relation that associates each element in a non-empty set S with a unique element in a
non-empty set T is called a unction rom S to T.
We denote this unction by f : S  T.
The set S is called the domain and T, the target set, is called the co-domain.
The set f ( S ) = { t| t  T , t = f ( s )} , subset o T, is called the range.
A unction f : S  T or which each element o the range, f (S ), corresponds to exactly
one element o the domain, S, is said to be an injection, i.e. i f ( a ) = f (b )  a = b
or a, b  S.
I every element in the co-domain o a unction is the image o at least one element in
the domain we say that the unction is a surjection, i.e. or all b in the co-domain there
exists an a in the domain such that f (a) = b.
A unction is a bijection i it is an injection and a surjection.
Given f : S  T where S and T are fnite sets, then:
a f is injective  n( f (S )) = n(S )
b f is surjective  n( f (S )) = n(T)

84 Extension of the concept of function


Composite functions
If g : A  B, f : B  C are functions, then f  g : A  C is also a function.
A function f : A  B is bijective  it has an inverse.
A function f : A  B is bijective  its inverse is also a bijection.
Properties of composite functions:
 Associativity (( f  g )  h ) ( x) = ( f  ( g  h ) ) ( x)
 If f : S  T and g: T  S are injections, then ( f  g ) ( x ) and ( g  f ) ( x ) are injective.
 If f : S  T and g: T  S are surjections, then ( f  g ) ( x ) and ( g  f ) ( x ) are surjective.

The identity function for a set S is a bijection IS : S  S such that IS ( x ) = x for all x  S .
Let f : S  S be any function, then ( IS  f ) ( x ) = ( f  IS ) ( x ) = f ( x ) for all x  S .

For a bijection f : S  T such that f ( x ) = y , x  S and y  T , the inverse function


f 1 : T  S is such that ( f 1  f ) ( x ) = IS and ( f  f 1 ) ( y ) = IT
A binary operation  on a non-empty set S is a rule for combining any two elements
x, y  S to give a unique element c. This is denoted by x  y = c. A binary operation on a
non-empty set S is said to be closed if for all a, b  S, a  b  S.
A binary operation on a non-empty set S is closed if for all a, b  S, A  B  S.
A binary operation  on a non-empty set S is said to be associative if for all a, b, c  S ,
a  (b  c) = (a  b)  c.
A binary operation  on a non-empty set S is said to be commutative if for all a, b  S ,
a  b = b  a.
If  is a binary operation on S and there is an element e  S such that for every element
x  S , e  x = x = x  e, then we say that e is the identity element of S under the
operation .
The identity element of a binary operation  on S is unique.
Let  be a binary operation on S with identity e. Then for all x  S if there exists an
element y  S such that x  y = e = y  x then we call y the inverse of x, and we write
y = x 1 .
For a binary operation  in S with identity e, if the inverse exists it is unique, i.e. each x
element of S has a unique inverse.
Let  be an invertible associative binary operation on a non-empty set S with identity e.
The operation is said to satisfy
i the left cancelation law, i.e. a  b = a  c  b = c
ii the right cancelation law, i.e. a  b = c  b  a = c

Chapter 2 85
The Universal
3 Theory of
Everything in
Mathematics
CHAPTER OBJECTIVES:
8.7 The defnition o a group {G, }; the operation table o a group is a
Latin square, but the converse is alse; Abelian groups.
8.8 Examples o groups: , , , and  under addition; integers under addition
modulo n; non-zero integers under multiplication modulo p, where p is prime;
symmetries o plane fgures, including equilateral triangles and rectangles;
invertible unctions under composition o unctions.
8.9 The order o a group; the order o a group element; cyclic groups; generators;
proo that all cyclic groups are Abelian.
8.11 Subgroups; proper subgroups; use and proo o subgroup tests.

Before you start


You should know how to: Skills check:
1
1 Given that f ( x ) = x + 1 and g(x) = e x, 1 Given that fand g are unctions on + such
2
that f (x) = ln(x + 1) and g (x) = x2, fnd the
a Find the inverses o unctions, e.g.
ollowing:
fnd f 1(x) and g1(x). Exchanging x
a ( f  g) ( x )
and f (x), solving or x, and then using
inverse notation, we obtain b ( f  g ) 1 ( x )
f 1(x) = 2(x  1); g1(x) = ln x c ( g  f) (x)
b Find the composition o unctions,
d ( f 1  g 1 ) ( x )
e.g. fnd ( f 1  g 1 ) ( x ) . Substituting
g1(x) or the variable in f 1(x) we
obtain ( f 1  g 1 )( x ) = 2 (ln x  1 )
c Recognize that unction composition
is not commutative, e.g. fnd
( g 1  f 1 ) ( x ) . Substituting f  1
or the variable in g 1 we obtain
( g 1  f 1 ) ( x ) = ln[2( x  1)]
Clearly, by considering the ormulas,
unction composition is not
commutative.

86 The Universal Theory of Everything in Mathematics


2 Check whether the properties o closure, 2 Check whether the properties o closure,
commutativity, associativity, identity and inverse commutativity, associativity, identity and
hold or a set under a given binary operation, e.g. inverse hold or the ollowing sets under
{, } , a  b = 2ab. Determine i any element(s) the given binary operations.
would have to be removed rom  in order or the
properties o identity and inverse to hold under . a (  + ,  ), a  b = a b

Closure, i.e. or all a, b   , a  b   . b (  + ,  ), a  b = 2 ab


Since a  b = 2ab, 2ab  , {, } is closed. c ( ,  ), a  b = ab + 1
Commutativity, i.e. or all a, b  , a  b = b  a.
Since a  b = 2 ab = 2ba = b  a , {, } is
commutative. The geometric nature of Islamic art
incorporates complex symmetries that have
Associativity, i.e. or all
been mathematically analyzed and explored.
a, b, c   , a  (b  c ) = ( a  b )  c. Since
Perhaps the most famous of such art
a  (b  c ) = a  ( 2bc ) = 2 a ( 2bc ) = 4 abc and
forms lies within the Alhambra, a fortress
( a  b )  c = (2 ab )  c = 4 abc , {, } is associative. constructed in Andalusia, Spain, in the 9th
Identity, i.e. or all a   there exists an e   century during the last Islamic sultanate on
such that a  e = a = e  a. the Iberian Peninsula. Some of the geometric
murals in the Alhambra are examples of
We need to fnd an m   such that symmetry groups, which you will learn
a  m = a = m  a. (Note that since we are not about in this chapter, and have some of the
sure that the set has an identity under the binary properties that you have been working with
operation, we do not yet use the symbol e or on the left hand side in the given example.
identity.)
For the right hand identity, a  m = a  2 am = a,
a 1
2am = a  m = = , a  0 . For the let hand
2a 2
1 1
identity, m  a = a  m = , a  0 . Hence, e = .
2 2
Strictly speaking, since {, } is commutative, it is
enough to look or either the right identity or let
identity, since they will be equal.
Inverse, i.e. or all a   there exists an a1  
such that a  a 1 = e = a 1  a . We need to fnd an
n   such that a  n = e = n  a .
(Note again, that since we do not know i each
element has an inverse, we do not yet use the
1
notation or inverse, a1.) Since a  n = 2 an = ,
2
1
n= , a  0. And since (, ) is commutative, we
4a
need only fnd either the right or let inverse.
For  to have an identity and inverse under ,
0 would have to be excluded. Hence, all the
properties hold or {\{0} , } .

Chapter 3 87
Group Theory
The search in Physics or a theory that ully explains and connects all
physical aspects o the universe results rom two major scientifc
paradigms in the last century: General Relativity and Quantum
Mechanics. General theorems providing a mathematical basis or such a
universal theory o everything have been attempted, and, at the time o
writing, a recent work entitled Generalized Mathematical Proo o
Einsteins Theory Using a New Group Theory was reviewed by both the
American Mathematical Society and the European Mathematical Society.
Indeed, increasingly it seems as i Group Theory is the uniying theory
o everything in mathematics, i.e. a branch o mathematics that can
connect all other branches by fnding similarities in their inherent
structures. In essence, Group Theory measures symmetry, the one idea
by which man through the ages has tried to comprehend and create order,
beauty, and perfection  Hermann Weyl.
In 1 824, the Norwegian Mathematician Niels Henrik Abel published
his impossibility theorem, in which he proved there is no general
solution, or ormula, or fnding the solutions o polynomial equations
o degree 5 (quintics) or higher. At about the same time, a brilliant
French teenager, Evariste Galois, explained why this is the case. He not
only resolved one o the great challenges o his day, but more
importantly, he discovered a compelling connection between symmetry,
permutation groups (which you will learn about in Chapter 4), and the
solvability o polynomial equations.
88 The Universal Theory of Everything in Mathematics
Although Galois and Abel laid the oundations o the mathematics
o Group Theory, it is a 20th century emale mathematician,
Emily Noether, who is credited with the title o ather o
Abstract Algebra, mainly through changing the 1 9th century
emphasis o its use rom solving polynomial equations into
creating an abstract axiomatic system.
Today, Group Theory is used in many diferent areas o study, such as
elementary Particle Physics, Music Theory, Crystallography, Chemistry,
Campanology (the study o bells and bell-ringing), and perhaps its most
popular usage in terms o the masses: solving Rubiks Cube!

3.1 Groups
A group consists o a set and a binary operation on that set.
The set with a binary operation has the our properties o closure,
associativity, existence o an identity element, and existence o inverses.

Defnition
The set G with a binary operation  is called a group i the
ollowing our axioms (properties) hold:
1 Closure: or all a, b  G, a  b  G
2 Associativity: or all a, b, c  G, a  (b  c ) = ( a  b )  c
3 Identity: or all a  G , there exists an element e  G such that
a e= a= e a
1
4 Inverse: or each a  G there exists a  G such that
a  a 1 = e = a 1  a

The group G with binary operation  is denoted by {G, } .


Note that the commutative property is not a required group axiom.
It is not necessary
For this reason it is important that both the let and right identity to check for both left
and inverse properties be conrmed. For example, the set  under and right identities if
the binary operation o subtraction has a unique right identity 0, the binary operation is
i.e. a  0 = a or all a  . However, 0  a =  a, hence it has no let commutative.
identity. Thereore {, } is not a group.
I in addition to the our properties above, a set G with binary
operation  is also commutative, then it is said to be an Abelian group.

Defnition
A group {G, } is an Abelian group i G is commutative under ,
i.e. or all a, b  G , a  b = b  a.

Chapter 3 89
It is important to note that the identity element is unique, as are the
inverses, i.e. a group contains only one identity element, and each element
contains a unique inverse. In Chapter 2 you proved these results or binary
operations on a set S using the let and right cancellation laws.
Groups may be fnite or infnite, i.e. consist o fnite or infnite sets.
Set A is fnite, i.e. its cardinality is n  N , i there is a bijection rom set
{0, 1 , 2,  , n} to A. A set is infnite i it is not fnite. I a set G is fnite,
then the group is also fnite, otherwise it is an infnite group.
Infnite groups
You have already been working with many examples o infnite groups,
e.g. the sets , ,  and  under the binary operation o addition.
Since the binary operation o addition is commutative, these our sets
under addition are urthermore Abelian groups.

Example 1
Show that the ollowing infnite sets are groups under the given binary operation.
Determine i any are Abelian groups.
a {  + , }
b {  \{ 0} , }
c The set o all real-valued unctions with domain  under addition.

a Closure: or all a, b  + , ab  + Show that all four of the group


properties hold.
Associativity: or all a, b, c  + , a(bc) = (ab)c
Identity: or all a  + , a  1 = a = 1  a
1 1 1
Inverse: or all a   + ,   + and a  = 1 =  a
a a a
{+ , } is a group since all group properties hold.
Commutativity: or all a, b  + , ab = ba, Determine if the commutative
hence {+ , } is an Abelian group. property holds.

b Closure: or all a, b  \{0} , ab  \{0} Show that all four of the group
Associativity: or all a, b, c  \{0} , a(bc) = (ab)c properties hold.
Identity: or all a  \{0} , a  1 = a = 1  a, 1  \{0}
1
Inverse: or all a   \{ 0} ,   \{ 0} and
a
1 1
a =1 =  a
a a
{\{0} , } is a group since all group properties hold.
Commutativity: or all a, b  \{0} , ab = ba, Determine if the commutative
hence {\{0} , } , is an Abelian group. property holds.

90 The Universal Theory of Everything in Mathematics


c Closure: or all real-valued unctions and g,  + g is a Show that all our o the group
real-valued unction. properties hold.

Associativity: or all real-valued unctions , g and h,


 + ( g + h) = (  + g) + h.
Identity: or all real-valued unctions , g (x) = 0 is a
real valued unction or all real values o x, and
 + g =  = g + .
Inverse: or all real-valued unctions , there exists a
real valued unction  such that
 + ( ) = g = ( ) + , where g (x) = 0, or all x.
The set o all real-valued unctions under addition is a
group, since all group properties hold.
Commutativity: or all real-valued unctions and g, Determine i the commutative
 + g = g + , hence the set o all real-valued unctions property holds.
under addition is an Abelian group.

We now consider other innite sets under a binary operation and


determine i each is a group. For example, the set Z+ under addition
is not a group, since there is no identity or addition in the set Z+ .
Also, the set o all non-negative integers under addition is not a
group, because although it contains the identity element 0, there are
no inverses or the non-zero elements o the set. (It is sufcient to nd
just one element in the set or which an inverse does not exist in order
to show that the set under the binary operation is not a group.)

Example 2
Determine i the ollowing sets are groups under the given binary operation.
a Z + under multiplication
b N under the binary operation  dened as a  b = | a  b|
ab
c Q + under the binary operation # dened as a # b = , a, b   +
2
a Since the identity is 1 , there is no inverse or 2. Identiy a property o groups that is
Indeed, other than 1 , no other elements o the not satisfed.
given set have an inverse.
b I a = 1 , b = 2, and c = 3, then Identiy a property o groups that is
a  ( b  c ) = 1  | 2  3 | = |1  |  1|| = 0 not satisfed.
( a  b )  c = | 1  2|  3 = ||  1 |  3| = 2
Associativity does not hold, so (N, ) is not a
group.

Chapter 3 91
ab Go through all the group axioms to see
c Closure: a # b =   + , so closure holds.
2 if they hold.

Associativity:
bc
a
 bc  abc
a # (b # c ) = a #   = 2 =
2  2 4
ab
c
ab abc
(a # b ) # c = #c = 2 =
2 2 4
Hence associativity holds.
Identity: nd b  Q + such that:
ab
a#b=a = a  b = 2, and
2
ba
b#a=a =ab=2
2
Hence, the identity e = 2; 2  Q + .
Inverse: nd c  Q+ such that

ac 4
a#c=2 =2c= , and
2 a
ca 4
c#a=2 =2c=
2 a
4
Hence, a 1 = ; a 1   + .
a
Since all the group axioms hold, { Q + , #} is a group. Write your conclusion.

Investigation
Consider the diferent number sets and their subsets, e.g. Z, Z+ , and C etc.,
the arithmetic operations +, , and the inverse operations  and . Select a
set and determine the binary operations under which it orms a group.
The ollowing table headings might help organize your work.

Number Set Operation Group (Yes, No) Reason

From the examples and the investigation, you have seen that in order
to show a given set with a binary operation is not a group, it is
sucient to show that any one o the group properties does not hold.

92 The Universal Theory of Everything in Mathematics


In part c o Example 2 you may have noticed that it is important to
ascertain the ollowing when checking to see i the properties hold:
 The identity element must be in the given set, and must commute
with every element in the set.
 The inverse or each element must be in the given set, and must
commute with the original element.

Example 
Show that the set o bijections orms a group under unction composition.
Closure: i and g are bijections such that f : A  B , g : B  C Confrm the group
then g  f : A  C. Hence, the composition o two bijections is properties. This was proven
a bijection and closure holds. in Chapter 2, theorem 5c.
Associativity: i , g and h are bijections, then or all x, The proo that unction
(h  g)  f = (h  g) ( f ( x ) ) composition is associative
is done in Chapter 2,
(
= h g ( f (x)) ) theorem 4.
= h (( g  f) ( x ) )
= h  ( g  f)
Hence, the composition o bijections is also associative.
Identity: the unction e : x  x is a bijection. For all unctions ,
e  f = f = f  e . Hence, e is the identity.
Inverse: every bijection  has an inverse - 1 that is also a This was proven in
bijection such that f  f 1 = e = f 1  f Chapter 2, theorem 3b.
Hence, the set o bijections orms a group under unction
composition. State your conclusion.

You have already seen that unction composition is not usually commutative,
hence the group in Example 3 is not Abelian.

Exercise 3A
1 Show that the set S = {2 n | n  Z} under multiplication orms a group.

2 Show that under addition, the ollowing sets o unctions : R orm a group:
a all continuous unctions
b all diferentiable unctions.

3 Determine i the ollowing sets under the given binary operation orm a group:
a R+ under the operation # dened as a # b = ab
b R\{0} under the operation  dened as a  b = a b
c {3 n
}
n  N under multiplication

d { a + bi a, b  R, }
a + bi = 1 under multiplication.

Chapter 3 93
4 a Show that the set S = R\ {1} under  defned as
a  b = ab + a + b orms a group.
b Determine i {S, } is an Abelian group.
c Find the solution o the equation 2  x = 7 in S.
d Explain why {R, } does not orm a group.

5 Let R = R\{0} and S = R  R , i.e. S is the set o all ordered pairs (a, b)
such that a and b are real numbers, and b is non-zero. Defne  such that
(a1, b1)  (a2, b2) = (a1+b1a2, b1b2).
a Show that { S, } is a group.
b Calculate the ollowing:
i (3, 2)  (1, 1)
ii (1, 2)  (0.125, 1.4)
c Determine i {S, } is Abelian.

6 Show that S = { nm | n   , m   } orms a group under addition.

7 Show that the set o all real-valued unctions o the orm f (x) = ax + b, a  0,
whose domain is R orm a group under the binary operation composition
o unctions. Does it orm an Abelian group?

Finite groups
So ar all our examples have been o infnite groups, i.e. groups where
the set G has an infnite number o elements. We will now consider groups
defned on fnite sets.
Since a group must contain an identity, it must contain at least one
element. The only possible binary operation  on {e} must necessarily
be defned as e  e = e. The identity element is its own inverse, and the
properties o closure and associativity obviously hold. We say that
the order o {{e}, } is  , i.e. the number o elements in the group is  .

Defnition
The order o a group {G, } is the number o elements in the
group. I a group has an infnite number o elements, it is said to
have infnite order, i.e. | G | = .

We will now create a fnite group o order 2. Since one o the elements
must be the identity, we defne the set S as {e, a} , e  a, and the binary
operation . We now set up an operation table or these two elements.
Checking the group properties, we see rom the table below that closure
holds, since there are no extraneous elements. We have said that e is the
identity element, and checking, we see that e  e = e and a  e = e  a = a.
I e is the identity, three out o our our group axioms are satisfed.

94 The Universal Theory of Everything in Mathematics


The one that remains to consider is a  a. For the closure property
to hold, the result can only be e or a. The result cannot be a since
then axiom 4 would not be valid, i.e. a would not have an inverse
in the set S. Hence, in order or {S, } to satisy all the group
properties, we can fll out the table only in the ollowing way.

 e a
e e a
a a e

Note that we can exchange the rows and columns, and obtain
 a e
a e a
e a e
which is actually the same as the frst table. By convention, however,
we put the identity element frst.
Checking or the associative property can be a tedious process or
larger sets, but we will fnd ways to get around this later on. For this
set o order 2, 2 3 or 8 distinct cases would have to be checked, and is
let or the student as an exercise.
Let us do the same with a set o three elements, S = {e, a, b} , under the
binary operation . Since e is the identity, the 1 st row and column mirrors
the initial order o the elements.
 e a b
e e a b
a a
b b
For the 2nd row, 2nd column entry, we can choose either e or b. I we
choose e, we would have to enter b in the 2nd row, 3rd column.
 e a b
e e a b
a a e b
b b
This would mean, however, that we have two let identities or b, namely
e and a, since e  b = b and a  b = b. Hence, we have no choice but to put b
in the 2nd row, 2nd column, and complete the table as below. You should
now justiy the completion o the table using the group axioms.
 e a b
e e a b
a a b e
b b e a

Chapter 3 95
To confrm the associative property, you would have to show that it holds in
3 3 , or 27, distinct cases. To spare you the task o going through this tedious process,
we can confrm or you here that indeed the associative property holds.
With the previous examples o fnite sets o orders 2 and 3 as background,
we will be able to list some necessary conditions that an operation table must
satisy in order to give a group structure on a set.
 In the row and column containing the identity element e, the rows and columns
are mirrored in the same order as they originally appear, i.e. the condition
e  x = x = x  e means that the row and column representing the operations
with e must contain exactly the elements appearing in the same order as
across the top o the table and down the let side o the table.
 Since every element has a unique inverse, the identity element e can
appear only once in each row and column.
 The equations a  x = b and y  a = b must have unique solutions x and y.
(This property will be proved later.) This means that each element o the
group must appear in each row and column only once.
An operation table that has the above properties is called a Latin Square.

Defnition
A Latin Square is a square array o n rows and n columns such
that each element or symbol occurs only once in each row and
each column.

Latin squares frst arose in the 18th century with


card games, such as the problem o arranging the
kings, queens, jacks and aces into a 4 by 4 array such
that each row and column contains one card rom each
o the 4 suits, and one card rom each o the 4 ranks
previously mentioned. In 1779, Euler posed the problem
that he claimed was impossible to solve: o arranging
36 ofcers rom 6 ranks and 6 regiments into a 6 by
6 square so that each row and each column contains
one ofcer rom each rank and one rom each regim ent.
Recently, the development o Latin squares has gained
a major impetus in designing statistical experiments and
also in fnite geometries.

We have shown above that i the elements o a set under a given binary operation
orm a group, we can place the elements in a Latin square. Conversely, i elements
o a set under a given binary operation are placed in a Latin square, the set under
the binary operation will orm a group provided that the group properties hold.
It is thereore not enough to assume that the set under the binary operation is
a group because it can be placed in a Latin square.

96 The Universal Theory of Everything in Mathematics


Although the IB HL syllabus uses Cayley table and operation table
interchangeably, historically a Latin square that orms a group structure is
called a Cayley table, ater the 19th century mathematician Arthur Cayley. He
was instrumental in ounding the British school o pure mathematics and was
also responsible or obtaining admission or women into Cambridge University.
He was one o the frst to realize that many seemingly dierent areas o
mathematics can be brought together under Group Theory. His n-dimensional
geometry has been applied to both relativity theory and quantum mechanics.

Example 4
Construct a Cayley table for the set S = {1 , 1 , i, i } under multiplication, and
show that {S, } is a group.

 1 1 i i Construct an operation table,


1 1 1 i i writing the identity element as the
1 1 1 i i frst element.
i i i 1 1
i i i 1 1
{S, } forms a group if the following properties hold: Confrm the group properties.
Closure: for all a, b  S, a  b  S. From the operation
table it is evident that the set is closed under .
Associativity: for all a, b, c  S, a  (b  c) = (a  b)  c You may assume multiplication o
Multiplication of complex numbers is associative. complex numbers is associative.
Identity: for all a  S, 1  a = a = a  1 , 1  S. State the identity.
Inverse: for all a  S there exists a 1  S such that
a  a 1 = 1 = a 1  a.
a 1 1 i i It is not enough to simply state that
a1 1 1 i i inverses exist. You must also
identiy the inverse o each element.
From the table we see that 1 and 1 are self-inverses
and i and i are mutual inverses. Hence {S, } is a group. State your conclusion.

Is the above group Abelian? We know that the set of complex


numbers under multiplication is commutative, therefore the
commutative property will hold for S, since S  C . A visual
method for determining if it is Abelian is to consider the
symmetry about the main diagonal of its Cayley table,  1 1 i i
i.e. the diagonal from the upper left hand corner to the 1 1 1 i i
lower right hand corner. 1 1 1 i i
Since there is symmetry about the main diagonal of the Cayley i i i 1 1
table, the group is Abelian. i i i 1 1

Chapter 3 97
Groups of integers modulo n
You have worked with integers modulo n, written (mod n), in Modular arithmetic is
Chapter 1 . Two integers, a and b, are said to be congruent (mod n) used in modern day
i a and b have the same remainder on division by n. In other words, banking. Banks require an
a  b (mod n )  a  b = kn, k  Z. IBAN (International Bank
Account Number)
Notation or Modular arithmetic:
identifcation or
 Z n = { 0, 1 , 2,  , n  1 } , n  N , n  2 transerring unds between
 + n denotes addition (mod n), and a + n b is the remainder bank accounts. IBAN
makes use o modulo 97
when a + b is divided by n, i.e. a + n b = a + b (mod n)
to trap user input errors in
  n denotes multiplication (mod n), and a  n b is the bank account numbers.
remainder when a  b is divided by n, i.e. a  n b = ab (mod n)

Example 5

a Construct a Cayley table or { 4, + 4} and confrm that it orms a group.


b Construct an operation table or Z 4 \{ 0} under  4 and show that it does not
orm a group.
c Determine whether or not { 4,  4} is a group.

a Closure: or all a, b  Z 4 , a + 4 b  Z 4 Construct the Cayley table.


Closure is evident rom the table.

+4 0 1 2 3
0 0 1 2 3
1 1 2 3 0
2 2 3 0 1
Confrm the group properties.
3 3 0 1 2

Associativity: or all a , b , c  Z 4 ,


a + 4 (b + 4 c ) = ( a + 4 b ) + 4 c

Addition (mod n) is associative. You may assume addition (mod n) is


associative.
Identity: or all a  Z 4 , State the identity.
0 + 4 a = a = a + 4 0, 0  Z 4

Inverse: or all a  Z 4 there exists a 1  Z 4 such Identiy the inverses o the elements.
that a + 4 a 1 = 0 = a 1 + 4 a
a 0 1 2 3
a1 0 3 2 1
Hence, { Z 4 , + 4 orms a group. State your conclusion.

98 The Universal Theory of Everything in Mathematics


b The operation table for Z 4 \{ 0} under the binary Construct the operation table.
operation  4 is:

4 1 2 3
1 1 2 3
2 2 0 2
3 3 2 1

We see from the table that closure does not hold, Identiy a group axiom that does not
since 0 appears in the table, and 0  Z 4 \{ 0}. hold. (It is sufcient to nd just one
axiom that does not hold. )
Hence, { Z 4 \ {0} , + 4 } does not form a group.
State your conclusion.
c The operation table for { Z 4 ,  4 } is: Construct the operation table.

4 0 1 2 3
0 0 0 0 0
1 0 1 2 3
2 0 2 0 2
3 0 3 2 1

Method 1
The operation table is not a Latin Square, i.e. the Check i the operation table is a Latin
elements 0 and 2 appear more than once in certain Square.
rows and columns.

Method 2
The identity element for all elements is 1 , but 0 and Find one example o a group property
2 have no inverses. that does not hold.
Hence {  4 ,  4 } does not form a group.

Chapter 3 99
Example 6

The Cayley table or a set o 5 elements under the operation  is given here.

 p q r s t
p s r t p q
q t s p q r
r q t s r p
s p q r s t
t r p q t s
a State with reason why the Cayley table is a Latin Square.
b Determine whether or not each o the group properties hold.
c Solve the equation ( p  x)  x = x  p.

a The Cayley table is a Latin Square because each Use the defnition o Latin Square.
element appears only once in each row and column.
b Closure is evident rom the table. The right and let Go through all the group properties
identity or each element is s. Each element is a sel- to determine i they hold.
inverse. The property o associativity does not hold,
since ( p  q)  t = r  t = p and p  (q  t) = p  r = t and t  p.
c Solutions are: q, r, s and t. In this example it is best to
substitute the elements or x as the
operation is not associative.

The example above shows a Latin Square that is not a group table.
Symmetry groups
a
We will now consider groups o plane fgures under the
composition o certain plane transormations that preserve
c
symmetrical properties.
0
Consider the symmetry in the Isle o Man moti here.

The coat of arms of the Isle of Man is the three-legged motif. Although
b
Alexander III introduced it to Scotland in the mid 13th century after he
gained control of the Manx territory, its origins go back to ancient times.
The motif carries the latin words Quocunque Jeceris Stabit , meaning it
will stand which ever way you throw it. This is thought to be a reference to
the independent and resilient spirit of the Manx people.

This fgure has three rotation symmetries about the center O o 1 20,
240 and 360 (or 0 ). We can label these transormations as ollows:
r is a clockwise rotation o 1 20 about O
s is a clockwise rotation o 240 about O
t is a clockwise rotation o 360 about O

100 The Universal Theory of Everything in Mathematics


Since symmetries are transormations, they can be combined, i.e. given two
transormations, we can perorm one ollowed by the other. For example,
on the original gure we can perorm the transormation r (rotation o
1 20 degrees through O clockwise) and ollow this with the transormation
s (rotate through 240 through O clockwise). The result is illustrated as ollows:

c a r b c s c a

o o o

b a b

Looking at the combined transormations, the result is to leave the original


gure unchanged. This same result can be obtained by rotating the original
diagram 360, or transormation t. Hence, transormation r ollowed by
transormation s is the same as transormation t, and is written in
symbols as s  r = t . This is read as transormation s ollowing
transormation r. In other words, similar to unction composition,
s  r = t is called the composition o r with s, i.e. we frst apply r, and
then apply s. Likewise, as with unction compositions, s  r = t is
perormed rom right to left.
 t r s
We will now determine i our set o rotations orms a group under
t t r s
composition o symmetry transormations by creating its Cayley table.
r r s t
We will place t rst, since it is the identity transormation. It is easy
or you to conrm the results on the right. s s t r

Notice that this table is a Latin Square, hence closure holds.


The identity is t, which is its own inverse, and r and s are mutual
inverses. Just as with unction composition, symmetry
transormation is associative.
Let us now consider the symmetries o the equilateral triangle XYZ.
There are three refective symmetries about the medians o Z
the triangle. (A median connects a vertex o a triangle to
the midpoint o the side opposite the vertex.) We can label
c
the transormations as ollows:

A: refection in median a
a O b
B: refection in median b
C: refection in median c X Y

Chapter 3 1 01
There are three rotation symmetries about O. We can label these:
I: rotation about the center 360 (or 0 ) clockwise (or anti-clockwise).
R1: rotation about the center 1 20 anti-clockwise (which is the same as
rotation about the center 240 clockwise).
R2: rotation about the center 1 20 clockwise (which is the same as
rotation about the center 240 anti-clockwise).
It might be helpul to make a cardboard copy o the triangle in order to see the
results o the various transormations.
In the three diagrams below, we see transormation B ollowed by
transormation A, and this is the same as the single transormation R1 .
Note that the median lines are fxed in space and do not rotate with the triangles.
Z X Y

c c c

B A

a O b a O b a O b

X Y Z Y Z

R1

Hence, AB = R1 , or B ollowed by A results in R1 .


We will now construct the Cayley table or the symmetries o the equilateral
triangle, and it is let to you to confrm the results in the table.

 I R1 R2 A B C
I I R1 R2 A B C
R1 R1 R2 I C A B
R2 R2 I R1 B C A
A A B C I R1 R2
B B C A R2 I R1
C C A B R1 R2 I

The Cayley table confrms that the set {I, R1 , R2, A, B, C} orms a group
under composition o transormations. The property o closure is
evident. I is the identity. I, A, B and C are all sel-inverses and R1
and R2 are mutual inverses. Composition o transormations is associative.
Since the table is not symmetrical about the main diagonal,
this group is not Abelian.

102 The Universal Theory of Everything in Mathematics


The set o six symmetries o the equilateral triangle with the binary operation
o composition o transormations is called the symmetry group o equilateral
triangles. All the symmetries o geometrical gures are elements o a larger
set o transormations called isometries, i.e. a transormation o the points
in 2D or 3D space such that distances between points remain unchanged.
Hence, under an isometry, a geometrical gure retains its shape and size,
but changes its position in space. There are our types o plane isometry:
rotation, refection, translation, and glide-refection, i.e. refection together
with a translation in the direction o the line o refection. It can be
shown that the set o all plane isometries orms a group under the
dierent transormations.

Example 7
a Construct a Cayley table or the group o symmetries y
o a square{S, } using the ollowing notation:
I: identity (rotation o 360 in either direction about the center) D C
R1 : rotation o 90 anti-clockwise about the center
R2: rotation o 1 80 anti-clockwise about the center
R3 : rotation o 270 anti-clockwise about the center
X: refection in the x-axis x

Y: refection in the y-axis


A: refection in the diagonal AC (refection in the line y = x)
A B
B: refection in the diagonal BD (refection in the line y = x)
b State whether or not the group is Abelian.

a
 I R1 R2 R3 X Y A B Enter the results o the binary
operation under the diferent
I I R1 R2 R3 X Y A B transormations into the table.
R1 R1 R2 R3 I A B Y X
R2 R2 R3 I R1 Y X B A
R3 R3 I R1 R2 B A X Y
X X B Y A I R2 R3 R1
Y Y A X B R2 I R1 R3
A A X B Y R1 R3 I R2
B B Y A X R3 R1 R2 I

b Since the table is not symmetric about the main Since we are told that this is a
diagonal, the group is not Abelian. group, we need only consider the
commutative property to determine
i it is Abelian.

Chapter 3 1 03
Symmetry groups are used throughout the study of
chemistry. The symmetry of a molecule provides
information on the energy levels of its orbital and
transitions that can occur between energy levels.
These can all be found without rigorous calculations,
which makes group theory so very powerful in the
study of the physical aspects of molecules.

Exercise 3B
1 a Copy and complete the given table so that the set {e, x, y, z} orms
a group under .
 e x y z
e e x y z
x x e
y y e
z z e

b Use the table to simpliy the ollowing:


i y  (z  x) ii (x  y)  ( y  z)

2 S = {a, b, c, d, e} under the binary operation  is defned in the table below.


 a b c d e
a a b c d e
b b c e a d
c c a d e b
d d e a b c
e e d b c a

a Simpliy:
i a  (b  c) ii (a  b)  c iii b  (d  c) iv (b  d )  c.
b Determine i {S, } has an identity element, and name it i it does.
c Determine whether each element has an inverse, and name its inverse.
d Give two reasons why {S, } does not orm a group.

1 1
3 Show that the set S ={ f, g, h} such that f (x) = x, g ( x ) = 1  , and h ( x ) =
x 1x
orms a group under unction composition. Determine i the group is Abelian.

4 Construct Cayley tables or {5 , + 5 } and {5 \{0} ,  5 } , and confrm
that both orm a group. Use the tables to solve the ollowing equations in 5 :
a x+ 4= 3 b 2x = 3 c 4x + 1 = 3
d 3(x + 1) = 1 e 4x + 1 = 2x

104 The Universal Theory of Everything in Mathematics


5 Construct an operation table or S = {2, 4, 6, 8} under  10 and
determine i it orms an Abelian group.

6 A set o six complex numbers orms a group under multiplication.


1
I one o the complex numbers is (1 + i 3 ) , fnd the other
2
fve numbers.

7 Construct Cayley tables or the symmetries o the ollowing


fgures, and show that each one orms a group. (You must
frst decide on all the symmetries that the fgure contains.)
a An isosceles triangle
b A rectangle
c A cuboid

8 Express the cube roots o unity in the orm a + bi, and show that
they orm a group under multiplication.

9 Write out the operation table or Z2  Z 2 and determine i it


orms a group under + 2.
(Z2 = {0, 1} , hence Z 2  Z 2 = {(0, 0), (0, 1), (1, 0), (1, 1)} ).

10 Show that i n = pq, where p, q are both integers greater than 1,


then (Zn ,  n ) does not orm a group.

3.2 Properties and theorems of groups and subgroups


In the frst part o this chapter, you have been working with both fnite
and infnite groups, and using some group properties intuitively.
We will now write these group properties, or theorems, and their proos.
First we will prove the right and let cancellation laws or groups.
Right and left cancellation laws for groups
Given a group {G, } and a, b, c  G
i the right cancellation law holds, i.e. a  c = b  c  a = b and
ii the let cancellation law holds, i.e. c  a = c  b  a = b.
Proof:
i a  c = b  c  (a  c)  c1 = (b  c)  c1 since c1  G.
 a  (c  c1) = b  (c  c1) by the associative property.
 a  e = b  e by the property o inverse, and
a  e = b  e  a = b by the identity property. Q.E.D.
The proo o ii is let or you to complete.

Chapter 3 1 05
Theorem 1
A group {G, } has the ollowing properties.
a The identity element or a group is unique.
b For all a  G, the inverse o a, a1 , is unique.
c For any a, b  G, the equations a  x = b and y  a = b, x, y  G, have
unique solutions in G. (For fnite groups, this means that each
element would appear only once in every row and column o its
operation table.)

Proofs:
A common strategy to prove uniqueness is to assume that uniqueness
does not hold, i.e. there are two distinct elements, and then show that
these two elements are indeed equal.
a Suppose there are two identity elements e1 and e2. Then, or any a  G ,
a  e1 = a = e1  a and a  e2 = a = e2  a. Thereore, taking each corresponding
part o both expressions separately, a  e1 = a  e2  e1 = e2 by the let
cancellation law, and, e1  a = e2  a  e1 = e2 by the right cancellation law.
Hence, uniqueness o the identity holds.

b Suppose that a  G has two inverses, a11 and a21 . It ollows then that
a  a11 = e = a11  a and a  a21 = e = a21  a . Taking each corresponding
part o both expressions separately, a  a11 = a  a21  a11 = a21 by the
let cancellation law, and a11  a = a21  a  a11 = a21 by the right
cancellation law. Hence, uniqueness o the inverse o an element holds.
c We frst need to show the existence o at least one solution or the
equations a  x = b and y  a = b .
Solving or x: a  x = b  a 1  ( a  x ) = a 1  b , since a 1  G.
 ( a  1  a )  x = a  1  b by the associative property.
 e  x = a  1  b by the property o inverse, and
 x = a  1  b by the property o identity. Hence, we obtain x = a 1  b .
Finding the solution or y is let or you to do.
Now, substituting the solution we ound or x,
a  ( a 1  b ) = ( a  a  1 )  b by the associative property,
= eb by the inverse property,
=b by the identity property.
Checking the solution o the 2nd equation is let or you to do.
To show uniqueness o these solutions, we again assume that there exist
two solutions, i.e. a  x1 = b and a  x 2 = b . Thereore a  x1 = a  x2 by
substitution, and by the let cancellation law, we conclude that x1 = x2.
Similarly, the uniqueness o y is similarly proved.
106 The Universal Theory of Everything in Mathematics
Some properties o groups
a For any a, b  G,
1
i ab = e  a = b

ii a  b = e  b = a 1
iii ab = e  ba = e
b For any a, b  G, ( a  b ) 1 = b 1  a 1
c For any a  G, ( a 1 ) 1 = a

a i a  b = e  ( a  b )  b 1 = e  b 1 since b 1  G
 a  (b  b 1 ) = b 1 by the associative and identity properties
 a  e = b 1 by the inverse property
1
 a =b by the identity property
and iii are let as exercises or you to complete.
ii
1 1 1
b The inverse o a  b is ( a  b ) . I b  a is the inverse o a  b,
then it ollows that ( a  b )  (b 1  a 1 ) must equal the identity e,
and this is what we need to confrm. (b1  a 1 )  (a  b) must
also equal e. (This latter part is let or you to do.).
( a  b )  ( b  1  a  1 ) = a  (b  b  1 )  a  1 by the associative property
= a  e  a 1 by the inverse property
1
= aa by the identity property
=e by the inverse property
The latter part is let or you do to.
Hence, b 1  a 1 is the inverse o a  b by Theorem 1 which we have
proven above, i.e. the uniqueness o the inverse.
c The inverse o a1 is (a1 ) 1 . I a is the inverse o a1 then it ollows
that a 1  a = e = a  a1 which is true by the inverse property o group G.
Hence by the uniqueness o inverse property, the result ollows.
Alternatively, since a 1  a = e, using property a with a 1 = b , ( a 1 ) 1 = b 1 = a.
Example 8

Given the group { G ,  } prove that i a  a = e or all a  G , then {G, } is Abelian.
To show that {G, } is Abelian, we need to show
that a  b = b  a or all a, b  G.
For all a, b  G, (a  b)  (a  b) = e, by the given. Use group properties and axioms to arrive
a  (b  a)  b = e by the associative property. at your results.
a  a  (b  a)  b = a  e
e  (b  a)  b = a by the given and identity property.
(b  a)  b  b = a  b by the identity property.
(b  a)  e = a  b by the given.
b  a = a  b by the identity property.

Chapter 3 1 07
Exercise 3C
1 {G, } contains exactly our elements: e, a, b, and c. State with
reasons why a  b cannot equal e, a or b, and hence must equal c.

2 Prove that i {G, } is a group and a  G, then ( a 2 ) 1 = ( a 1 ) 2 .

3 {G, #} is a group such that x # x # x # x = e , or x 4 = e , y 2 = e ,


and x # y = y # x 3 .
a Show that
i y # x = x3 # y
ii y # ( x2 # y) = x2
b Simpliy ( x # y ) # ( x 2 # y ) .

4 {G, } is an Abelian group and a n = a  a  a    a or n actors


o a, where a  G and n  Z +. Prove by mathematical induction
that ( a  b ) n = a n  b n or all a  G .

5 Show that in any group {G, } , i ( a  b ) 2 = a 2  b 2 then a  b = b  a.


6 A set S is defned as the set o all elements o a group {G, } that
commute with every element o G, i.e. a  S  a  x = x  a or
every element x  G . Prove that {S, } is also a group.

Subgroups
In question 6 o Exercise 3C you proved that a subset S o a set G
under the same binary operation as G was also a group. When a
subset o a group orms a group in its own right under the same
binary operation, then we say that the subset is a subgroup o the
given group.

Defnition
I a non-empty subset H o a set G is also a group
under , then {H, } is a subgroup o {G, } .

An example o an infnite subgroup o { R, +} is { Q , + } . However,


although +  , { + , +} is not a subgroup o { R, +} . For fnite sets,
consider Example 7, the table or the symmetries o a square, S. I we
consider a subset T o the table with the elements T = {I, R1 , R2, R3 }
we can determine i this subset T o S under the defned transormations
is a subgroup o {S, } .

108 The Universal Theory of Everything in Mathematics


 I R1 R2 R3
I I R1 R2 R3
R1 R1 R2 R3 I
R2 R2 R3 I R1
R3 R3 I R1 R2

We can see rom the table that {T, } is closed. In addition, T contains I,
the identity element o S. Each element in T has an inverse in T.
We know that {S, } is associative, hence the operation will be
associative with the elements o the subset T. So we can conclude
that {T, } orms a subgroup o {S, } under .
Notice also that the order o the subgroup is a actor o the order o the group.
When looking or possible subgroups o a given group, this act can minimize
the amount o work in our search. In the next chapter we will prove this
amous and important result, i.e. the order o a subgroup divides the order
o the group.
Thereore, in order or a set to orm a subgroup o a given group under a
given binary operation, it must also ulfll the group axioms. Any subset
o the group under the given binary operation is associative, so this property
does not need to be shown.

Theorem 2: Subgroup Theorem


A subset H o a group {G, } is a subgroup {H, } i and only i:
1 H is closed under the binary operation , i.e.
a, b  H  a  b  H
2 The identity e o {G, } is in H.
3 For all a  H , a 1  H .

Proof:
: Since {H, } is a subgroup o {G, } , then all the group properties
must hold.
: I H  G such that 1 , 2 and 3 hold, then we need only show the
property o associativity. Since all elements in H are also in G,
and or all elements in G,  is associative then {H, } is also
associative.
A corollary o the above theorem is that every group {G, } has at
least two subgroups: the group itsel and the group consisting only
o the identity.

Chapter 3 1 09
Defnitions
In many textbooks, the subgroup {{e} ,} is a proper subgroup. For
the IB syllabus however, we consider it an improper subgroup.
I {G, } is a group, then the subgroup consisting o G itsel and
the subgroup consisting o only the identity are the improper
subgroups o G. All other subgroups are proper subgroups. The
subgroup {{e} , } is also reerred to as the trivial subgroup o G.

Example 9
Show that the set S = {1 , 5, 7, 1 1 } orms an Abelian group under  1 2, and list all o its
non-trivial subgroups.

 12 1 5 7 11 Construct a Cayley table.


1 1 5 7 11
5 5 1 11 7
7 7 11 1 5
11 11 7 5 1
To show that {S, 1 2 } orms a group, the Confrm all the group properties.
ollowing properties must hold:
Closure: It is evident rom the table that
or all a , b  S , a  1 2 b  S .
Identity: It is evident rom the table that 1 is the
identity, since or all a  S , a  1 2 1 = a = 1  1 2 a .
Inverse: For all a  S there exists a 1  S such that
a  1 2 a 1 = 1 = a 1  1 2 a .

a 1 5 7 11
a1 1 5 7 11

Each element is its own inverse.


Associativity: Multiplication mod(n) is
associative.
The above confrms that {S, 1 2 } is a group.
Determine i there is symmetry about the
It is also an Abelian group, since or all
a , b  S , a  1 2 b = b  1 2 a . This is true since the
main diagonal o the Cayley table.
Cayley table is symmetric about its main
diagonal.
In addition to the set itsel, the sets o the Since the order o a subgroup must divide
non-trivial subgroups under the given operation the order o a group, we are looking only
are: {1 , 5} , {1 , 7} , {1 , 1 1 } . or subgroups o order 1, 2 and 4.

11 0 The Universal Theory of Everything in Mathematics


Example 10
Let {H, } and {K, } be subgroups o {G, } .
Prove that { H  K ,  } is a subgroup o {G, } .
H  K is i non-empty and ii a subset o G. First show that the conditions o the
defnition o subgroup are satisfed, that
i Since H and K are subgroups, then eG  H and
H  K is a non-empty subset o G.
eG  K, hence eG  H  K. H  K is non-empty.
ii Let x  H  K . Then x  H and x  K.
Since both H and K are subsets o G, x  G ,
thus H  K  G .
For { H  K ,  } to be a subgroup o {G, } it must Show that {H  K,  } satisfes the
satisy the group properties: properties o the Subgroup Theorem.
Closure, i.e. or all a, b  H  K, a  b  H  K.
Let a, b  H  K. Then a , b  H and a , b  K .
Since both {H, } and {K, } are groups, a  b  H
and a  b  K , thus a  b  H  K .
Identity: We have already proved in i that since H
and K are subgroups, eG  H and eG  K, hence
eG  H  K .
Inverse: For a  H  K, a  H and a  K .
Hence a 1  H and a 1  K , since both {H, } and
{K, } are subgroups. Hence, a 1  H  K. By the
Subgroup Theorem, thereore, { H  K, } is a
subgroup o {G, } .

We will now prove a theorem with subgroups that might be helpul


in showing that fnite or infnite subsets o a group orm a subgroup
under the given binary operation.

Theorem 3
Let{G, } be a fnite or infnite group and H a non-empty subset
o G. Then H is a subgroup o G i a  b 1  H or a, b  H.

Proo:
We are given that a, b  H  a  b 1  H .
Identity: Letting b = a  a  a 1  H, hence e  H .
Inverse: Letting a = e and b = a, then e, a  H  e  a 1  H rom the given. Since
e  a 1 = a 1 , a 1  H . In the same way, b 1  H. Hence, or a, b  H, a 1 , b 1  H .
Closure: From the above, we know that i a and b are in H then a and b1 are in
H too. Using the given, thereore, a  (b 1 ) 1  H, hence a  b  H.

Chapter 3 1 11
The ollowing is an example where you might use this theorem
instead o the Subgroup Theorem.

Example 11

Prove: Given {H, +} where H = { 4 x + 7 y x , y  Z } , {H, +} is a subgroup o { Z , + } .

x, y  Z  4 x + 7 y  Z , hence H is a non-empty First show that the condition o the


subset o Z. defnition o subgroup is satisfed, i. e. H is
a non-empty subset o Z .

Let a, b  H, a = 4 x1 + 7 y1 , b = 4 x2 + 7 y2 .
Since e = 0 x + 0 y = 0, or x  Z we have x 1 = x. To defne the inverse o an element in a
group, you must frst fnd the identity.
Hence a + b 1 = (4x1 + 7y1 )  (4x2 + 7y2) Use Theorem 3 and show that a + b 1   .
= 4(x1  x2) + 7( y1  y2)  H.

Hence H is a subgroup o {Z, +} . Write your conclusion.

Beore we examine sufcient conditions or proving that a set H is a


subgroup o G under the same binary operation, we need to dene
what is meant by the order o an element o a group.
I a  G under the binary operation , then we can use the binary
operation on a itsel, i.e. a  a. We can do this as many times as we need,
n times

e.g. a  a   a = a n . We can now ormulate the ollowing denition


and theorem.

Defnition
Let a  G where {G, } is a group. Then a is said to have fnite
order i a n = e or some n  Z + . The order o a is the least such n.
I no such n exists, the element a has infnite order.

Theorem 4
Let a be an element o a nite group {G, } . Then there exists a
smallest positive integer n such that a n = e , and n is the order o a.

11 2 The Universal Theory of Everything in Mathematics


Proof:
The set o all possible powers o a is an innite set. Since G is nite,
however, the set o possible powers o a cannot all be diferent.
Hence, i r and s are two positive integers with r < s such that a r = a s = e,
then, (or convenience sake we will omit )
a r = a s  a s a  r = a r a  r = e . Hence, there is at least one n = s  r, such that a n = e .
We have proven the existence o n, and you may want to prove its
uniqueness as an exercise.

Theorem 5
I H is a non-empty subset o a fnite group {G, } then {H, } is
a subgroup i and only i, or all a, b  H, a  b  H. In other words,
H need only be closed.

Proof:
Identity: a  b  H  a 2  H or b = a. Now, b = a 2  a 3  H.
Continuing in this way, let the order o a be n, hence a n = e , and e  H .
Inverse: Consider a n 1 . Since a n 1 a = a n = e = aa n 1, then an is the inverse o a,
and a n 1  H, a 1  H .
The condition o closure is given in the theorem, hence { H, } is a subgroup
by the Subgroup Theorem.
This is a very useul theorem, since when you are asked to show that a subset
o a fnite group is a subgroup, you need only show the property o closure!
This is particularly useul when you have a Cayley table to work rom.

Exercise 3D
1 List the proper subgroups o the given groups.
a The set containing the sets , A = {a} , B = {b} , and C = {a, b}
under the operation symmetric diference, .
b The set o unctions under unction composition, where
1 1  x 1
p (x) = x, q ( x ) = 1  , r( x ) = , and s ( x ) = .
2x 1  2x 2  2x
c The symmetry group o the rectangle.
d The set {2, 4, 8,  0,  4,  6} under   8
e {Z 6 , + 6 }

2 The set S = {1, 2, 4, 7, 8, 11, 13, 14} orms a group under the operation  15 .
a Write down the inverses and orders o each element.
b Given that the set { , 2, a, b} is a subgroup o S, nd a and b.
c Find one o the subgroups o S that also has our elements
and includes 4 but not 2.

Chapter 3 1 13
3 A group G under the binary operation  has distinct elements
{e, a, b, c,  } , where e is the identity element.
a I a  b = e and b  b = a, prove the set {e, a, b} orms a subgroup o G under .
b I a  a = b, b  b = c, c  c = a, then prove the set {e, a, b, c} does not orm a
subgroup o G under .

4 Let F be the group o all real-valued unctions with domain  under


addition o unctions. Prove that the subset o F consisting o those
unctions in F that are diferentiable orms a subgroup.

5 A group G that contains more than ten elements contains an element


q o order 10. Prove that {q, q2, q3 ,  , q10} is a subgroup o G.

6 Let {G, } be an Abelian group. Prove that i H is the set o all


elements x in G satisying the equation x2 = e, then H is a subgroup.

7 Let {G, } be a group, and a is a xed element in G. Prove that


i H is the subset o G whose elements commute with a, i.e.
H = { x  G xa = ax } , then H is a subgroup o {G, } .

3.3 Cyclic groups


In Theorem 4 we saw that i a is an element o a nite group G then the
powers o a cannot all be diferent. Consider the subgroup o the symmetries
o a square group consisting o the rotations symmetries only, and its
Cayley table.
I: identity (rotation o 0 or 360 in either direction about the center).
R : rotation through 90 anti-clockwise
R2: rotation through  80 anti-clockwise
R3 : rotation through 270 anti-clockwise

 I R1 R2 R3
I I R R2 R3
R1 R R2 R3 I
R2 R2 R3 I R
R3 R3 I R R2

We see that R1  R1 = R12 = R2 ; R1  R1  R1 = R13 = R3 ; R1  R1  R1  R1 = R14 = I .


As we proceed with higher powers, we obtain repetitions o the elements,
e.g. R1  R1  R1  R1  R1 = R15 = R14  R1 = I  R1 = R1 .

11 4 The Universal Theory of Everything in Mathematics


We can thereore rewrite the table using powers o R1 :

 I R1 R12 R13
I I R1 R12 R13

R1 R1 R12 R13 I

R12 R12 R13 I R1

R13 R13 I R1 R12

This subgroup is an example o a fnite cyclic group, because all o the


elements o the group can be written as a power o a single element.
We say that the subgroup is generated by the element R1 .

Defnitions
A group whose elements can be expressed in the orm
{e, a, a2, a3 , . . . . , an1 } is called a cyclic group o order n and is
denoted by Cn. The element a is said to generate the group and is
described as the generator o the group. It ollows thereore that a
group o order n is cyclic i an only i it contains an element o
order n.

A cyclic group can be a fnite group, as seen above, or an infnite group.

Recently it was discovered that the sequence o pitches which orms a


musical melody can be transposed (translation) or inverted (refection) and C C#/
B D
can be modeled using a cyclic group o order 12. This allows or the creation
b
b
A A#/ B

o dierent melodies by assigning unctions to the transpositions and D D Eb E


#/
inversions.
Ab
#/

F
F#/ G b G G
C C# D Eb E F F# G Ab A Bb B
0 1 2 3 4 5 6 7 8 9 10 11

Chapter 3 1 15
Example 12
a Show that the group {  5 \{ 0} ,  5 } orms a cyclic group, and fnd its generator(s).
b Find all the possible subgroups.

a
5 1 2 3 4 Construct the Cayley Table. Since we are
told that it is a group we do not need to test
1 1 2 3 4 the group properties.
2 2 4 1 3
3 3 1 4 2
4 4 3 2 1

22 = 4, 2 3 = 3, 2 4 = 1 hence the group can be Determine if the powers of an element


expressed as {1 , 2, 2 2, 2 3 } and is thereore cyclic. generate all elements of the group.
3 2 = 4; 3 3 = 2, and the group can be expressed as Determine if any other elements also
{1 , 3, 3 2, 3 3 } . generate the elements of the group.
2 and 3 are generators.
b From the table it is evident that {1 , 4} orms a Since the order of a subgroup must divide
subgroup since it is closed, 1 is the identity, the order of a group, we look only for
4 is a sel-inverse, and associativity is subgroups of order 2.
implied. Test the properties for subgroup.
Notice in Example 12 that 3 is the inverse o 2. Since 2 was a generator,
its inverse will also be a generator. The ollowing is let as a proo or the
student, and is one o the exercises at the end o this section (See Exercise 3E question 2).

Theorem 6
In a fnite group {G, } , the order o an element a is the same as
the order o its inverse a1 . (The proo is let as an exercise or you
to complete.)

We will now establish some properties o cyclic groups.

Theorem 7
Every cyclic group is Abelian.

Proof:
Let Cn be a cyclic group and let a be a generator o Cn so that
C n = { a n n  Z }. Let x and y be two elements o Cn. Hence, there exists
integers p and q such that x = a p and y = a q. Then, xy = a paq = a p + q = a q + p = a q a p = yx.
Hence, C is Abelian.

11 6 The Universal Theory of Everything in Mathematics


We now know that if we have a cyclic group it is also Abelian, but the
converse is not necessarily true.
Before discussing subgroups of cyclic groups, it is convenient to prove
the following theorem.

Theorem 8
Let {G, } be any group and let a  G . Then {H, } where
H = { a n n  Z} is the smallest subgroup of {G, } that contains
{a} , i.e. every subgroup containing {a} contains H.

Proof:
Checking the three conditions for subgroup, since a ra s = a r + s, r , s  Z ,
H is closed. Since a 0 = e , 0  Z , e  H , and since for a r  H, a  r  H , and
a r a  r = a  r a r = e , every element in H has an inverse in H. Since any subgroup
of {G, } that contains {a} must contain H, H is therefore the smallest
subgroup of G containing {a}.

Defnition
The subgroup of {G, } , H = { a n n  Z } , is the cyclic subgroup of
{G, } generated by a.

Theorem 9
A subgroup of a cyclic group is cyclic.

Proof:
(This proof is placed here to enhance understanding; it is not
required for examination purposes.)
Let C be a cyclic group generated by a and let H be a subgroup of C.
If H = {e} , then it is cyclic. If H  {e} , then a n  H, n  Z + . Let m be
the smallest integer in Z+ such that a m  H.
For c = a m to generate H, we must show that every b  H is a power
Division Algorithm
of c. Since b  H and H  C , b = a n for some n. We can express n
Theorem: If m is a
as mq + r, for 0  r < m (Division Algorithm Theorem).
positive integer and
Then a n = a mq + r = (a m ) q a r , or a r = ( a m )  q a n . n is any integer then
Since a n  H, a m  H and H is a group, both (am )  q and an are in H. there exist unique
Hence, ( a m )  q a n  H, i.e. a r  H . Since m was the smallest positive integers q and r such
that n = m q + r and
integer such that a m  H and 0  r < m, we must have r = 0. Hence
0  r < m.
n = qm and b = a n = ( a m ) q = c q . Hence b is a power of c.

Chapter 3 1 17
Example 3

A cyclic group C consists o the ollowing elements: e, x, x 2 , x 3 , x 4 , x 5 , x 6 , x 7 .


Determine:
a the elements that are generators o C, and
b the orders o the remaining elements.

a x 8 = e, (x 3 ) 8 = e, (x 5 ) 8 = e, and (x7) 8 = e are the Since the order of C is 8, we need to look


smallest such powers to equal e. Hence these for elements such that the least power of
elements are all generators o C. such an element to equal e is 8,
i. e. (x n) 8 = e, 1  n  7. This occurs when
n and 8 are relatively prime.
b Since (x2) 4, (x4) 2, and (x6) 4 all equal e, the Since the powers of the elements 2, 4, and
orders o these elements are respectively 6 are factors of 8 or factors of its multiples,
4, 2 and 4. Hence, they cannot generate C. the elements with these powers cannot
generate C.

At the beginning o this section on cyclic groups, we saw that R ,


rotation through 90 anti-clockwise, was a generator o the subgroup
o rotation symmetries o the square group. Since the order o an
element o a fnite group is the same as the order o the cyclic
subgroup generated by the element, and the order o a subgroup
must divide the order o a group, we can state the ollowing theorem.

Theorem 10: Lagranges Corollary


The order o an element o a fnite group divides the order o the group.

Theorem  0 ollows directly rom Lagranges Theorem, which we will


study and prove in Chapter 4.

Exercise 3E
1 Show that the group {Z 10, + 10} is generated by the number 7.

2 Prove by mathematical induction:


1 1 1 1
a For all a  {G , } , ( a1  a2    an ) = an  an 1    a1 , n  2 .

b For all a  {G, } , ( a n ) 1 = ( a 1 ) n , n  2.


c Hence, or otherwise, prove that the order o an element is equal
to the order o its inverse.

11 8 The Universal Theory of Everything in Mathematics


3 Z n  Z m = { ( a, b ) a  Z n , b  Z m } orms a group under the binary
operation  defned as (a1, b1)  (a2, b2) = (a1 + n a2, b1 + m b2), where
+ n and + m denote additions o integers modulo n and m, respectively.
a State the order o ( Z 4  Z 5 ,  ) , and evaluate (3, 2)  ( , 4).
b Show that { Z 2  Z 3 ,  ) is cyclic, and list any generators.
c Determine how many elements o { Z 2  Z 4 ,  } have order 4.

4 Show that
2 n1
a the nth roots o unity can be expressed in the orm {1,  ,  ,  ,  }
where  is the complex root with the smallest positive principal
argument
b the nth roots o unity orm a cyclic group under multiplication.

5 a Prove that i a group G has order p, where p is prime, then G is cyclic.


b Prove that i a group G has order pq, p, q  {Primes , then every
proper subgroup o G is cyclic.
c Find the number o generators o the cyclic group Zpq.

Review exercise
EXAM-STYLE QUESTIONS
1 a Show that the set o real numbers, excluding a single number,
orms a group under the operation  defned as a  b = a + b  ab,
and determine the single number which must be excluded
rom R.
b Hence, solve the equation 5  x = 12.

2 H is a subgroup o G and R is a relation defned on G such that or


all a , b  G , aRb  ab 1  H . Show that R is an equivalence relation.

3 Let x, a, b and c be elements o a group with identity element e.


a Solve or x: axb = c
b Solve simultaneously or x: ax2 = b and x3 = e

4 A group G with identity element e contains elements x and y such


that yx = x 2y and y 3 = e.
Prove:
a y 2 xy 2 = x 4 b x8 = x.

Chapter 3 1 19
1
5 a Given that f1 ( x ) = x, f2 ( x ) = 1  x, f3 ( x ) = ,
x
obtain expressions or f4 ( x ), f5 ( x ), and f6(x) i:
f4 ( x ) = ( f2  f3 )( x ) ; f5 ( x ) = ( f3  f2 )( x ) ; f6 ( x ) = ( f3  f4 )( x ) .

b Given that G = { f1 , f2 , f3 , f4 , f5 , f6 } orms a group with respect


to unction composition, construct its Cayley table.
c Determine the order o each element o the group.
d Find a subgroup o G containing only three elements.

6 Let {G, } be a group and a is a fxed element o G. Defne a unction


f: G  G by f ( x ) = a  x , or every x  G . Prove that fis bijective.

7 For the group { Z 1 2 , + 1 2 } :


a Find the order o the elements 4, 5 and 9.
b Show that the group is cyclic, and fnd all possible generators.

8 Let min (x, y) be the minimum value o two numbers x and y.


The operation # is defned on the set o negative integers by
x # y = min (x, y). (min(x, x) = x)
a Show that # is commutative.

b Determine which o the group axioms are satisfed.

9 Let {G, } be a group with subgroups {H, } and {K, } . Prove that
{ H  K, } is a subgroup o {G, } i and only i either H  K or K  H.

10 Find the order o a group generated by two elements a and b


i a3 = b2 = (ab) 2 = e, and fnd all subgroups o the group.

11 Construct a Latin Square o order 6 which has an identity element


and all other elements have order 2, and prove that this Latin Square
does not represent a group.

12 Let {H, } be a subgroup o {G, } . Let a  G, a  H, and


aH = { ah h  H}.
i Show that H  a H =  .
ii Show that H  aH is a subgroup o G.
iii Show that the number o elements in H  aH is twice
the number o elements o H.

120 The Universal Theory of Everything in Mathematics


Chapter  summary
Defnitions
A group {G, } is an Abelian group i G is commutative under ,
i.e. or all a, b  G, a  b = b  a.
The order | G| o a group {G, } is the number o elements in the group.
I a group has an infnite number o elements, it is said to have infnite order,
i.e. | G| = .
A Latin Square is a square array o n2 compartments such that each element
or symbol occurs exactly once in each row and column.
Symmetry groups are groups o transormations o plane fgures that preserve
symmetrical properties.
I a non-empty subset H o a group {G, } is also a group under ,
then {H, } is a subgroup o {G, } .
I {G, } is a group, then the subgroup consisting o G itsel and the subgroup
consisting o only the identity are the improper subgroups o G. All other
subgroups are proper subgroups. The subgroup {{e} , } is also reerred to
as the trivial subgroup o G.
Let a  G where {G, } is a group. Then a is said to have fnite order
i an = e or some n  Z + . The order o a is the least such n. I no such n exists,
the element a has infnite order.

The set G with a binary operation  is called a group i the ollowing our axioms
(properties) hold:
1 Closure: For all a, b  G , a  b  G
2 Identity: For all a  G, there exists an element e  G such that
a e= a= e a
3 Inverse: For each a  G there exists a1  G such that
a  a-1 = e = a1  a
4 Associativity: For all a, b, c,  G, a  (b  c) = (a  b)  c

The group G with binary operation  is denoted by {G, }

Integers modulo n and modular arithmetic


 Z n = {0,  , 2,  , n   } , n  N, n  2
 + n denotes addition (mod n), and a + n b is the remainder when a + b is
divided by n, i.e. a + n b = a + b (mod n).
  n denotes multiplication (mod n), and a  n b is the remainder when
a  b is divided by n, i.e. a  n b = ab (mod n).

Chapter 3 1 21
Properties and theorems of groups and subgroups
 Cancellation laws: Given a group {G, } and a, b, c  G:
i the right cancellation law holds, i.e. a  c = b  c  a = b and
ii the let cancellation law holds, i.e. c  a = c  b  a = b.

2 A group {G, } has the ollowing properties:


 The identity element or a group is unique.
 For any a  G, the inverse o a, a , is unique.
 For any a, b  G, the equations a  x = b and y  a = b, x, y  G, have unique
solutions in G. (For fnite groups, this means that each element would appear
only once in every row and column o its operation table.)
 For any a, b  G:
 a  b = e  a = b
 a  b = e  b = a
 a  b = e  b  a = e
 For any a, b  G, (a  b)  = b  a .
 For any a  G, (a )  = a.
3 Subgroup Theorem: A subset H o a group {G, } is a subgroup
{H, } i and only i:
1 H is closed under the binary operation , i.e. a, b  H  a  b  H.
2 The identity element e o G is in H.
3 For all a  H, a1  H.

Let {G, } be a fnite or infnite group and H a non-empty subset o G.


Then H is a subgroup o G i and only i a  b1  H or a, b  H.
Theorem: Let a be an element o a fnite group {G, } . Then a has fnite order.
Theorem: I H is a non-empty subset o a fnite group {G, } then {H, }
is a subgroup i and only i, or all a, b  H, a  b  H.
A group whose elements can be expressed in the orm {e, a, a2, a3 ,  , an }
is called a cyclic group o order n and is denoted by Cn. The element a is a
generator o the group. A group o order n is cyclic i an only i it
contains an element o order n.
Theorem: In a group {G, } the order o an element a is the same as the
order o its inverse a .
Theorem: Every cyclic group is Abelian.

122 The Universal Theory of Everything in Mathematics


Theorem: Let {G, } be any group and let a  G. Then H = {an | n  Z }
is the smallest subgroup o {G, } that contains a, i.e., every subgroup
containing a contains H.
The subgroup o {G, } above, H = {an | n  Z } , is the cyclic subgroup
o {G, } generated by a.
Theorem: A subgroup o a cyclic group is cyclic.
Theorem: (Lagranges Corollary): The order o an element o a fnite group
divides the order o the group.

Chapter 3 1 23
The classifcation
4 o groups

CHAPTER OBJECTIVES:
8.10 Permutations under composition o permutations; cycle notation or
permutations; result that every permutation can be written as a composition
o disjoint cycles.
8.11 Defnition and examples o let and right cosets o a subgroup o a group;
Lagranges theorem; use and proo o the result that the order o a fnite group
is divisible by the order o any element (Corollary to Lagranges theorem).
8.12 Defnition o group homomorphism; defnition o a kernel o a homomorphism;
proo that the kernel and range o a homomorphism are subgroups; proo o
homomorphism properties or identities and inverses; isomorphism o groups;
the order o an element is unchanged by an isomorphism.

Before you start


You should know how to: Skills check:
1 Find the partition o a set induced by 1 Find the partition o the set induced by
an equivalence relation, e.g. i the given equivalence relations:
a For a, b  , aRb  2| (a2 + b2),
A= { 5 ,  3, 1
5
, 2 , 6, 20 } i.e. 2 divides (a2 + b2).
and the equivalence relation R on A b R is an equivalence relation on
a
is defned by aRb   , fnd the   , such that or all
b
(a, b), (c, d )    ,
partition o A induced by R.
(a, b)R(c, d )  a = c. Describe how the
The partition o A induced by R is

{ }
 
equivalence relation R partitions   .
 { 5, 20 } ,  3,
1
, 6 , { 2 }  .
c The equivalence relation R on
 5 
S = {1, 2, 3,  , 10} is defned as
xRy  x  y(mod 4). Find the partition
o S induced by R.

124 The classifcation o groups


2 Determine i a unction f:A  B is 2 a Let A = {x| x  , x  0} and
surjective, injective, or both, e.g. let 3x2 + 5
let f : A  A be defned as f ( x ) = .
A = \{2} , and f : A   such that 5x2 + 3
3x Determine i fis bijective.
f( x ) = .
x+2
Determine whether or not fis bijective. b Given that f : 2  2 such that
For fto be bijective, it must be (i) injective f (a, b) = (2a + b, a  2b), show that f is
and (ii) surjective. bijective, and fnd its inverse.

(i) To show that fis injective, Injective and surjective unctions can be
represented graphically, as shown below.
Method I
We must show that f (a) = f (b)  a = b. In Figure 1, since the unction is steadily
increasing over its entire domain, the
Hence,
unction is injective. Also, i you imagine
3a 3b a horizontal line drawn anywhere
=  3 a (b + 2 ) = 3 b ( a + 2 )
a+2 b+2 through the graph, the unction will
 6a = 6b intersect such a line at only one point.
a=b
or y

Method II
We must show that fis either strictly
increasing or strictly decreasing on its
domain.
 3x 
d 
x+2 = 6
For all x  2,
dy
> 0,
dx ( x + 2) 2
dx
thereore fis strictly increasing. x

Figure 1
Hence fis injective.
Imagine that the graph in Figure 2
(ii) To show that fis surjective, we must continues to infnity at both ends. Then
show that or all b   there exists an any horizontal line drawn through the
a  A such that f (a) = b. Hence, graph will intersect it in at least one
3a point. This unction is surjective.
= b  3a = b(a + 2 )
a+2 y
 3a  ba = 2b
 a (3  b) = 3b
3b
a=
3b x
When b = 3 it is not the image o any
element in A, so f is not surjective.
Thereore f is not bijective.

Figure 2

Chapter 4 1 25
Group structures
The entire theory o groups originally grew out o an understanding
o permutations. You are amiliar with permutations as
arrangements o a given nite set. The search or solutions o
polynomial equations led the French mathematician Lagrange and
others, in the late 1 8th century, to think o permutations as bijections
rom a nite set onto itsel. However it was the French
mathematician Augustin-Louis Cauchy who developed in detail the
basic theorems o permutation theory and introduced the standard
notation we still use today.
In addition to permutation groups, we will also ocus on
isomorphisms and homomorphisms, which are unctions between
groups that preserve certain group structures. The German
mathematician Emmy Noether rst treated the ideas on group
structures in a paper published in 1 927. She is considered one o the
most amous mathematicians o our modern era.

4.1 Permutation groups


When you shufe a deck o 52 playing cards you are essentially
rearranging all o the cards, or orming permutations on the set o cards.
A permutation is thereore essentially a bijection o a set onto itsel.

126 The classifcation o groups


Defnition
A permutation o a non-empty fnite set A is a bijection rom A to A.

I we consider the set o the three natural numbers A = {1 , 2, 3}


and orm all o the possible permutations rom A onto itsel, one possible
mapping is 1  1 , 2  2, 3  3. We can illustrate this
permutation in the ollowing way:
1 2 3 
p1 =  . This mapping leaves all elements unchanged.
1 2 3 
1 2 3
Another possible mapping is 1  3, 2  2, 3  1 or p 2 =  .
3 2 1
How many possible mappings are there rom A to itsel? You should
know the answer rom your work on permutations in the core book: 3!, or 6.

Let us now complete the other our permutations:

1 2 3  1 2 33  1 2 3  1 2 3 
p3 =  , p4 =   , p5 =   , p6 =  .
1 3 2  2 1 3  3 1 2  2 3 1 

We perorm operations on permutations in the same way as


unction composition or transormations on a set o isometries.
In other words, i we want the operation p3 p4, then just as in the set
o isometries, this means p4 ollowed by p3 .
p4 maps 1 to 2, and p3 maps 2 to 3, hence p3 p4 maps 1 to 3.
p4 maps 2 to 1 , and p3 maps 1 to 1 , hence p3 p4 maps 2 to 1 .
p4 maps 3 to 3, and p3 maps 3 to 2, hence p3 p4 maps 3 to 2.
1 2 3 
Thereore, p3 p4 =   , which is p5 .
3 1 2 
Now consider p4 p3 . This means p3 ollowed by p4.
p3 maps 1 to 1 , and p4 maps 1 to 2, hence p4 p3 maps 1 to 2.
p3 maps 2 to 3, and p4 maps 3 to 3, hence p4 p3 maps 2 to 3.
p3 maps 3 to 2, and p4 maps 2 to 1 , hence p4 p3 maps 3 to 1 .
1 2 3
Thereore, p3 p 4 =   , which is p6.
2 3 1

We see already in this case that composition o permutations, just as in


unction composition, is not commutative, since p3 p4  p4 p3.
We will now determine whether the set o permutations on three
elements, S3 = {p1 , p2, p3 , p4, p5 , p6} , orms a group under composition o
permutations. Composition o permutations, just like composition o
unctions, is associative.
The set would need an identity, and clearly p1 is the identity.

Chapter 4 1 27
We now consider the inverses o the elements. Since a permutation is
a bijection rom a set onto itsel, we know that all elements have inverses.
1 3
2
Consider p2 =  1
 . Since p2 maps 1 to 3, p2 would map 3 to 1 .
 3 2 1 
1 2 3 
In the same way, p2 would map 2 to 2, and 1 to 3. Hence, p 21 = 
1

3 2 1 
which means that p2 is its own inverse. Finding the rest o the inverses is let as an exercise
beore you see the answers in Example 1 .

Example 1
Show that the set S3 o all permutations o the set {1 , 2, 3} orms a group under composition
o permutations. The defnitions o p1 , p2, etc. are those used on the previous page.

p1 p2 p3 p4 p5 p6 Work out all the permutations and put the



results in a Cayley table.
p1 p1 p2 p3 p4 p5 p6
p2 p2 p1 p5 p6 p3 p4
p3 p3 p6 p1 p5 p4 p2
p4 p4 p5 p6 p1 p2 p3
p5 p5 p4 p2 p3 p6 p1
p6 p6 p3 p4 p2 p1 p5

Closure: It is evident that the set under Ascertain the group properties.
composition o permutations is closed,
i.e. or all pi , pj  S3 , pi pj  S3 .
Identity: p1 is the identity, since or all
pi  S3 , pi p1 = p1 pi = pi .
Inverse: For all pi  S3 there exists a pj  S3
such that pi pj = pj pi = p1 .
pi p1 p2 p3 p4 p5 p6
p 1 p1 p2 p3 p4 p6 p5
i

Composition o permutations is associative. Composition of functions is associative.


Hence, S3 is a group under composition o
permutations.

Is a permutation o a set A consisting o our elements also a group?


We know rom our core work with permutations that this set would
consist o 4!, or 24 elements. It would not be practical to set up a
Cayley table or S4, so instead we will prove the ollowing theorem.

128 The classifcation o groups


Theorem 1
Let A be a non-empty set o n elements, and let Sn be the
set o all permutations o A. Then Sn orms a group under
composition o permutations, i.e. {Sn, } orms a group.

Proo:
We shall examine the group properties.
Closure: Similar to unction composition, the composition o two
permutations yields a permutation, so Sn is closed.
1 2  n The identity
Identity: The identity permutation is p1 =  .
1 2  n permutation is
Inverse: Since a permutation is a bijection, or any permutation the identity function
f(x) = x.
p  Sn there is an inverse permutation p1  Sn.
Associativity: Just as unction composition, the permutation
composition is associative.
Thereore, {Sn, } orms a group.

Defnition
Let A be the fnite set {1 , 2, 3,  , n} . The group o all
permutations o A is called the symmetric group on n elements
and is denoted by Sn.

Example 2

1 2 3 4 5 6 
Given x  S6, x =  , fnd
3 1 2 4 6 5 
a the inverse
b the order o x.

Since x maps 1  3, 2  1,
1 1 2 3 4 5 6  3  2, 4  4, 5  6, 6  5,
a x = 
2 3 1 4 6 5  x1 maps 3  1, 1  2, 2  3,
4  4, 6  5, 5  6
1 2 3 4 5 6 
x2 =  ;
2 3 1 4 5 6 
1 2 3 4 5 6  4  1 2 3 4 5 6 
x3 =  ; x =  ; Find the frst power o x that
1 2 3 4 6 5  3 1 2 4 5 6  equals e, i. e. , the identity
 1 2 3 4 5 6  6 1 2 3 4 5 6  permutation p1.
x5 =  ; x =  .
2 3 1 4 6 5  1 2 3 4 5 6 
Hence the order o x is 6, since x6 = p1 and
x, x2, x3 , x4, x5  p1 .

Chapter 4 1 29
Exercise 4A
1 Show that the subset of permutations of S4, {e, x, y, z} forms a
group, where z = xy and
1 2 3 4 1 2 3 4 1 2 3 4
e= , x =  , y =  .
1 2 3 4 3 4 2 1 2 1 4 3

2 Determine the order of the smallest subgroup of S5 containing


the element 1 2 3 4 5  .
 
1 4 3 5 2 
3 Find a cyclic subgroup of {S3 , } of order 3, and state a generator
of this subgroup.

4 The following are permutations on the set S5 .


1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
= ,  =  ,  =  
5 3 4 1 2 2 3 4 5 1 5 3 2 4 1
a Find the permutations:
i  ii  iii  2  iv  1 v ( ) 1 vi  1
b Solve for x in the following equations:
i x =  ii x = .

Permutations and cycle form


Another way of writing a permutation is in cycle form. Using the
1 2 3 
elements of S3 , since p2 =  , p can be expressed as a cycle
3 2 1  2
using the notation ( 3). This means  is mapped onto 3 and since 3 is
mapped onto  , the cycle ends. Since 2 is mapped onto 2, we can write
this as (2). We can then write the permutation as a product of cycles.
In other words, p2 = ( 3)(2) or (2)( 3). The single element that is in
brackets is mapped onto itself, i.e. the element that is invariant under the
mapping is put in its own brackets. The cycle notation for the identity
element of S3 , p , is ( )(2)(3); in other words, each element is mapped
onto itself. For simplicity of notation, the invariant element(s) will
be omitted. The identity, therefore, would be represented simply as ( ),
and for p2 its cycle form is simply ( 3).
Lets now write the other permutations of S3 using cycle notation.
1 2 3 
p3 =   , or p3 = (23).
1 3 2 
Using cycle notation, (23) means that  maps onto itself.

130 The classifcation o groups


1 2 3 
p4 =  , or p4 = (1 2).
2 1 3 
Again, this means that 3 is invariant and maps onto itsel.
1 2 3  1 2 3 
p5 =   , or p5 = (1 32); p6 =   , or p6 = (1 23).
3 1 2  2 3 1 
We will now write the element x rom Example 2 using cycle notation.
1 2 3 4 5 6
x=  = (1 3 2 )(4 )(5 6 ) , or (1 32)(56). Since the
3 1 2 4 6 5
cycles are disjoint, we can also write this as x = (56)(1 32),
i.e. whenever the cycles are disjoint, the cycle orm is commutative.
We can also write the inverse o x in cycle notation.
1 2 3 4 5 6
x 1 =   = (1 23 )(5 6 )
2 3 1 4 6 5
Again, since the cycles are disjoint, we can also write x 1 = (56)(1 23).
Notice that to fnd the inverse o an element in cycle notation,
we simply reverse the integers in the cycle. For example, inverting
the integers in (1 32) gives us (1 23), since it is understood that the
number at the end o the cycle is the same number as at the beginning
o the cycle. Thus (1 32) is 1  3  2  1 and its reverse is 1  2  3  1 .
The cycle (56), i.e. 5  6  5, is the same as (65), i.e. 6  5  6.
Let us now consider a permutation and write the permutation and its
inverse in cycle orm.
1 2 3 4 5 6 7 8 9 1 0 
Let  =  
5 1 6 8 4 1 0 7 2 9 3 
We see that 1  5  4  8  2  1 , which gives us the cycle (1 5482).
Then starting with the smallest number that we have yet to use, 3, we
have 3  6  1 0  3, giving us the cycle (3 6 1 0). (Notice that we leave
spaces in this cycle between the numbers to avoid conusion since
we have a two-digit number in our cycle.) The only remaining numbers are
7 and 9, which are invariant.
Hence,  = (1 5482)(3 6 1 0), or  = (3 6 1 0)(1 5482). Using cycle notation,
lets now fnd the inverse o . Reversing the numbers in the cycle
ater the frst number we obtain (1 2845)(3 1 0 6). In other words,
1  2  8  4  5  1 and 3  1 0  6  3. Since 7 and 9 are not
present in our cycles, they are invariant.
1 2 3 4 5 6 7 8 9 1 0 
Hence,  1 =  
2 8 1 0 5 1 3 7 4 9 6 

Chapter 4 1 31
How can we also use cycles to fnd the order o a permutation? We can
defne the length of a cycle as the number o moves required to come
ull cycle, i.e. the cycle (1 23) requires 3 moves to go rom 1 back to
1 again. Lets look again at Example 2,
1 2 3 4 5 6 
where x =  .
3 1 2 4 6 5 
We know that x can be written in cycle orm as (1 32)(56). The length
o the cycles are 3 and 2. The diagram below illustrates what this means.

1 2

5
3

We want to determine the smallest power n such that xn = p1 . Lets label


the frst cycle o length 3 as a, and label the second cycle o length 2 as b.
Every application or permutation o x moves the numbers around in a
cycle so that x would require 3 moves in cycle a to go rom 1 back to 1 .
In cycle b, x would require 2 moves to go rom 5 to 6 and back again.
This means that both 3 and 2 would need to divide n, the total number
o applications o x. Since we want both 3 and 2 to divide n, and n must
be the lowest such number, we want the lowest common multiple o 3 and 2,
which is 6. We have already seen in Example 2 that the order o x is 6.
From all the previous examples, we can summarize our fndings into
cycle properties.

Properties of cycle form


 Every permutation can be written as a product o disjoint cycles.
 Disjoint cycles are commutative.
 The order o a permutation written as a product o disjoint cycles
is the least common multiple o the lengths o the cycles.
The proofs of the above properties are not required for examination purposes,
and are not included in this course companion but you may decide to prove
them informally.

132 The classifcation o groups


Example 3

1 2 3 4 5 6 7 8 9 1 0 
Find the order o  =  .
5 1 6 8 4 1 0 7 2 9 3 
 = (1 5482)(3 6 1 0) Write the permutation in cycle form.
The length o the cycles are 5 and 3. Find the lowest common multiple of the
Since lcm (3, 5) = 1 5, the order o  is 1 5. lengths of the cycles.

Lets now consider permutation composition using cycles. Consider the


cycles a = (1 24) and b = (1 256) in S6. We can write these as permutations.
1 2 3 4 5 6 
In cycle a, since 3, 5 and 6 are invariant, a = (1 24 ) =  .
2 4 3 1 5 6 
1 2 3 4 5 6 
In cycle b, 3 and 4 are invariant, hence b = (1 256 ) =  .
2 5 3 4 6 1 
We already know how to fnd the composition o For convenience, at times
these two cycles using permutation composition. we refer to composition as
Lets concentrate now in fnding the product a product, particularly when
through the cycles, i.e. we want a  b, or (1 24)(1 256) writing it in cycle form.
in cycle orm.
As you already know, or permutation composition we move rom
right to let. The right cycle maps 1 to 2, and then the let cycle maps
2 to 4, so the composition maps 1 to 4. The right cycle then maps
2 to 5 and the let one maps 5 to 5, so 2 is mapped onto 5. The right
cycle maps 5 to 6, and the let one maps 6 to 6, so 5 is mapped onto
6. The right cycle maps 6 to 1 , and the let cycle maps 1 to 2, so 6 is
mapped onto 2.
1 2 3 4 5 6 
We can write the permutation a  b =   or (1 4)(256).
4 5 3 1 6 2 
In cycle notation, a  b = (1 24)(1 256) = (1 4)(256).
Can we arrive at the result in cycle orm without writing out
the permutation?
We see that in composing (1 24)(1 256), starting on the right cycle,
1 maps onto 2, and 2 maps onto 4 on the let cycle, so 1 maps onto
4, and well write this as an unclosed cycle, i.e. with no closing bracket:
(1 4. Then in the right cycle, 4 is mapped onto 4, and on the
let cycle 4 is mapped onto 1 , so now we can close this cycle (1 4).

Chapter 4 1 33
Then, the cycle on the right maps 2 to 5, and 5 is invariant in the
frst cycle, so 2 maps onto 5. We begin this cycle as (25. Then in the
right cycle 5 maps onto 6, and 6 is invariant in the let cycle, so 5 maps
onto 6, or (256. Now 6 maps onto  , and  maps onto 2 in the let
cycle, hence 6 maps onto 2, and we now have a ull cycle (256).
Thereore, ( 24)( 256) = ( 4)(256).
We will now compute b  a or ( 256)( 24).
Using arrows to indicate a mapping, starting on the right,   2,
then on the let 2  5, hence   5, or ( 5. Since 5 is invariant in the
right cycle, 5  5, and on the let 5  6, so 5  6, or ( 56. In the right
cycle, 6  6, and on the let 6   , hence 6   , so we have a complete
cycle ( 56). In the right cycle, 4   , and on the let   2, hence 4  2,
or (42. On the right, 4   and on the right   2, hence 4  2, so we
have a complete cycle (42). This means the 3 is invariant, and
b  a = ( 56)(42), or ( 256)( 24) = ( 56)(42).
We see again that permutation composition is not commutative,
since a  b  b  a.

Exercise 4B
1 a Write each o the ollowing permutations as a product
o disjoint cycles.
1 2 3 4 5 6 1 2 3 4 5 6 7 8
x = , y =  ,
6 3 2 5 1 4 3 4 1 2 6 7 8 5

1 2 3 4 5 6 7
z = 
1 3 2 5 4 7 6

b Write the inverses o x, y and z in cycle orm.


c Find the orders o x, y and z.

2 Write the ollowing products o cycles in permutation orm.


a on S 6: (123)(46)
b on S 7: (12)(345)(67)
c on S 8 : (245)(378)
d on S 9: (3457)(689)

3 Given that  = (136)(24) and  = (1452), both on S6,


fnd the ollowing in cycle orm:
a  1 b  c (  ) 1 d  1  1

4 Prove that the order o a cycle is equal to its length.

134 The classifcation o groups


4.2 Cosets and Lagrange's theorem
We will start this section with an important defnition needed to prove
the theorem you are already amiliar with: Lagranges theorem.

Defnition
Let {H, } be a subgroup o {G, } and let x  G. Then the set o
elements xH = {xh | h  H} is called a let coset o {H, } in G.
The set o elements Hx = {hx | h  H} is called a right coset o
{H, } in G.

We will show how this defnition works by fnding the let and right
cosets o the subgroup {3, +} o {, +} .
The let coset o 3 containing x is x + 3.
I x = 0, then 0 + 3 = { , 6, 3, 0, 3, 6,  } .
To fnd another let coset, lets take an element that is not in 3,
or example 1 . Then, 1 + 3 = { , 5, 2, 1 , 4, 7,  } .
Another element not in 3 is 2. Then, 2 + 3 = { ,4, 1 , 2, 5, 8,  } .
Now, consider the coset o k, i.e. k + 3. I k  0(mod 3), then k + 3 = 3.
I k  1 (mod 3) then k + 3 = 1 + 3. I k  2(mod 3), then k + 3 = 2 + 3.
It should be clear that there are only these three unique cosets.
Furthermore, these three let cosets partition  into let cosets o 3.
Finding the right cosets in the same manner will yield the exact
same results. However, since  is Abelian, the let coset k + 3 and
the right coset 3 + k are the same, hence the partition o  into right
cosets is the same as its partition in to let cosets.
Observe that in general, the equivalence relation R or the subgroup
{n, +} o {, +} is the same as the relation o congruence modulo n.
This means that the partition o  into cosets o n is the partition
o  into residue classes modulo n. (We do not have to distinguish
let and right cosets since addition is commutative.)

Example 4

The group {6, +} is Abelian. Find the partition o  6 into cosets o


the subgroup H = {0, 3} under addition.

One coset is {0, 3} itsel. Find the cosets containing 0, 1, 2, 


1 + {0, 3} = {1 , 4}
2 + {0, 3} = {2, 5}
The cosets are {0, 3} ,{1 , 4} ,{2, 5} . Since these three sets exhaust all of 6 , they are
the only cosets.

Chapter 4 1 35
You will have noticed that for a subgroup {H, } of an Abelian group {G, } ,
the partition of G into left cosets of H and the partition of G into right cosets
of H are the same.

Example 5

Let G = {1 , , 2, 3 , 4, 5 } under complex number multiplication be the cyclic group
i
consisting of the sixth roots of unity, where  = e 3 . Let H = {1 , 2, 4} be a subgroup
of {G, } . Find the left cosets of H.
For x  G, x = 1 , 1 H = {1 , 2, 4} Choose an x  G, e. g. x = 1 and form the left
For x  G, x = , H = { , 3 , 5 } cosets. Then choose another x  G, e. g. x = 
and form the left cosets.
The only two cosets are: These two cosets partition the group, so all
1 H = {1 , 2, 4} and other cosets would be identical to one of these
H = { ,  3 , 5 } two.

Also in Example 5, since every cyclic group is Abelian, the left and
right cosets will be the same.

We will now list some properties of cosets which you will undoubtedly
have noticed in the previous examples.

Theorem 2: Properties of cosets


For any subgroup {H, } of a group {G, } :
1 G is the union of disjoint cosets of {H, } , i.e., the group is
partitioned by the left (or right) cosets of its subgroup.
2 Every coset (left or right) of a subgroup {H, } has the same
number of elements as H.
3 Every element of G lies in one of the cosets of H in G.

The proofs of these properties are left as an exercise.


We will now consider an example where the left and right cosets are not the same.

136 The classifcation o groups


Example 6

Consider the Cayley table for S3 shown in Example 1 , and consider the subgroup {H, }
such that H = { p1 , p3 } . Find the partitions of S3 into left and right cosets of H.
Comment on your results.

1 2 3  1 2 3  1 2 3 Work out all the possible permutations.


p1 =   ; p2 =   ; p3 =  ;
1 2 3  3 2 1  1 3 2

1 2 3  1 2 3  1 2 3 
p4 =   ; p5 =   ; p6 =  
2 1 3  3 1 2  2 3 1 

p1 p2 p3 p4 p5 p6 Compose the Cayley table.



p1 p1 p2 p3 p4 p5 p6
p2 p2 p1 p5 p6 p3 p4
p3 p3 p6 p1 p5 p4 p2
p4 p4 p5 p6 p1 p2 p3
p5 p5 p4 p2 p3 p6 p1
p6 p6 p3 p4 p2 p1 p5

The left cosets of H are: Find all the left cosets of H.


p 2 H = { p 2 p 1 , p 2 p 3 } = { p 2 , p5 }
p3 H = { p3 p1 , p3 p3 } = { p3 , p1 } = H
p4H = { p4 p1 , p4 p3 } = { p4, p6}
p5 H = { p5 p1 , p5 p3 } = { p5 , p2} = p2H
p6H = { p6 p1 , p6 p3 } = { p6, p4} = p4H
The partition of S3 into left cosets of H is either
[H, p5 H, p6H], [H, p2H, p4H], [H, p2H, p6H], or
[H, p4H, p5 H].
The right cosets of H are: Find all the right cosets of H.
Hp2 = { p1 p2, p3 p2} = { p2, p6}
Hp3 = { p1 p3 , p3 p3 } = { p3 , p1 }
Hp4 = { p1 p4, p3 p4} = { p4, p5 }
Hp5 = { p1 p5 , p3 p5 } = { p5 , p4} =Hp4
Hp6 = { p1 p6, p3 p6} = { p6, p2} = Hp2
The partition of S3 into right cosets of H is either:
[H, Hp5 , Hp6], [H, p2H, p4H], [H, Hp2, Hp5 ], or
[H, Hp4, Hp6].
The partitions into left and right cosets are not the You need show only one example where
same, e.g. p2H = {p2, p5 } and Hp2 = {p2, p6}  p2H. the partitions into left and right cosets
This makes sense since {S3 , } is not Abelian. are not the same.

Chapter 4 1 37
Through the use o cosets, we are now in a position to prove Lagranges
theorem, which you have already been using in Chapter 3.

Theorem 3: Lagrange's theorem


I {H, } is a subgroup o {G, } , then the order o the subgroup
{H, } is a divisor o the order o {G, } .

Proo:
Let the order o {G, } be n and the order o {H, } be m, where m < n.
Let k be the number o cells, or sets, in the partition o {G, } into
let cosets o {H, } .
Hence, n = km, since every coset o H must also have m elements.
Thereore m is a divisor o n.
It is quite astonishing that this elegant and useul theorem
comes rom simply counting cosets, and the number o
elements in each coset!
We will now consider two amous corollaries o Lagranges theorem.

Corollary 1 The order o an element o a fnite group divides the


order o the group.

Proo:
Since the order o an element is the same as the order o the cyclic
subgroup generated by the element, the result ollows rom
Lagranges theorem.

Corollary 2 Every group o prime order is cyclic.

Proo:
Let {G, } be o prime order p. Since p > 1 there is some a  G
such that a  e. Then, the cyclic subgroup o {G, } generated
by a contains at least two elements, i.e. it has order m such that m  2.
Since by Lagranges theorem, m must divide p, then m = p.
Since {G, } is generated by a, {G, } is cyclic.
In the syllabus, the corollary to Lagranges theorem is defned as Corollary 1.

138 The classifcation o groups


Exercise 4C
1 Write out the proos or the three properties o cosets.
2 Find the let and right cosets o the ollowing subgroups:
a H = {4, +} o the group G = {, +}
b H = {4, +} o the group G = {2, +}
c H which is the set o elements generated by the element 4 in
the group { 12, + 12}
d Find in cycle orm the let and right cosets o the subgroup {H, } ,
H = {(1), (12)} , o the group G = {S3 , } , i.e. fnd gH and Hg.

3 H = { 2  {0} , +} is a subgroup o the group { 2   3 , +} .


Let (a, b) + (c, d) = (a + c (mod 2), b + d (mod 3)).
a List the sets H =  2  {0} and  2   3 .
b Find the let and right cosets o H.

4 Show that the right cosets o the subgroup { 2k, + 8} , k   8 o the


group {8 , + 8 } partitions the group.

5 Let {H, } be a subgroup o a group {G, } . Prove that i x  yH


then xH = yH.
6 Let {H, } be a subgroup o {G, } and let a  G. Prove that
aH = H i and only i a  H.

4.3 Homomorphisms
In Group Theory, we are interested in the properties induced by the
inner structure o groups. We want to make comparisons among
the groups, and decide which ones have equivalent structures,
regardless o the particular sets and binary operations that defne
particular groups.
To do this, we defne a relationship between two groups {G, } and
{H, } in terms o a mapping that relates the structures o the groups.
The groups may be fnite or infnite.

Defnition
Let {G, } and {H, } be groups. A homomorphism is
a unction f: G  H such that f (x  y) = f (x)  f (y) or all x, y  G.

Essentially this means that the operation  takes place in G while the
operation  takes place in H. These may or may not be the same
binary operations. The unction thereore defnes a relation between
these two binary operations, and hence between the two group structures.
You are already amiliar with many homomorphisms because you
have actually been using them throughout your mathematics courses,
without really reerring to them as such. Here are some examples:

Chapter 4 1 39
 The distributive property o multiplication over addition in the set
o real numbers says that or every real number c, c (x + y) = cx + cy
or all x, y  . In the language o groups we can say that c:   
where c(x) = cx is a homomorphism rom {,+} to {,+} .
 Another property o real numbers states that | xy| = | x| | y| , or x, y  .
In the language o groups, the absolute value unction :   
given by (x) = | x| is a homomorphism rom {\{0} , x} to {\{0} , x} .
 We know that or all real numbers x and y, (xy) 2 = x2y2. Again, in the
language o groups, we can say that :    such that (x) = x2 is
a homomorphism rom {\{0} , x} to {\{0} , x} .
 We know that or all real numbers x and y, 2 x + y = 2 x  2 y. Again, in
the language o groups, we can say that :    + such that (x) = 2 x
is a homomorphism between the groups {, +} and { + , } .
The unction denitions in a homomorphism need not be injective
or surjective. The third bullet point contains the squaring unction,
which is neither surjective nor injective. I we change the mapping o the
sets in the same example to :  +  , the unction is injective but
not surjective. I again we change the unction to :  +   + , this
unction is bijective.
We will now consider homomorphisms among diferent groups, both nite
and innite. In the ollowing example, we rst consider a homomorphism
between an innite group and a nite group.

Example 7

Given are the two groups {, +} and {S, } such that S = {1 , i, 1 , i} .
a Show that the unction (x) = i x denes a homomorphism between the two groups.
b Determine i the unction is injective, surjective, both or neither.
c Describe the mapping o  :   S as a partition o {, +} induced by an equivalence
relation, and dene the equivalence relation.

a Let m, n  . Then  (m + n) = i m + n = i m i n = (m) (n), Use the defnition o


hence is a homomorphism. homomorphism.
b i 4 k, k    1 , i.e. 4k = { , 8, 4, 0, 4, 8,  }  1 Use the unction to determine the
i j, j = { , 7, 3, 1 , 5, 9,  }  i elements resulting rom the
i r, r = { , 6, 2, 2, 6,  }  1 mapping.
i t, t = { , 5, 1 , 3, 7, 1 1 ,  }  i
The mapping is surjective since or all y  S there
exists an x  Z such that  (x) = y.
The mapping is not injective since many diferent
integers have the same image, or example,
 (2) = 1 =  (6).
c {, +} has been partitioned into the our cosets 4, Describe the mapping and defne
4 + 1 , 4 + 2, and 4 + 3 by the equivalence the equivalence relation.
relation xRy  (x) = ( y).

140 The classifcation o groups


For any groups {G, } and {H, } there is always at least one
homomorphism, namely the trivial homomorphism. The unction
f : G  H defned as f (x) = eH maps every element x in G onto the identity
element in H, eH. This unction is a homomorphism since
f(x  y) = eH = eH  eH = f(x)  f(y), or x, y  G.
We will now defne and prove some properties o homomorphisms,
which, loosely speaking, means that the homomorphism preserves
the identity and inverses.

Theorem 4: Properties of Homomorphisms


Let f be a homorphism rom group {G, } to group {H, } . Let a
be an element in G. Then the ollowing properties hold.
1 The homomorphism maps the identity in group G onto the
identity in group H, i.e. i eG and eH are the identity elements in
{G, } and {H, } respectively, then f(eG ) = eH.
2 The homomorphism maps the inverse o an element in group
G to the inverse o the elements image in group H, i.e. or all
a  G, f (a ) = ( f (a))  .
3 The range o the homomorphism fis a subgroup o {H, } , i.e.
f (G ) = { f ( a )  a  G } is a subgroup o {H, } .
4 The homomorphism preserves all powers, i.e. f(an) = ( f(a)) n or
all a  G.

Proofs:
1 Let f : G  H be a homomorphism rom group {G, *} to {H, } .
Then, or all a in G, f (a) = f (a  eG ) = f (a)  f (eG ) by defnition o
homomorphism and the identity element eG. By defnition o eH,
f (a)  f (eG ) = f (a)  eH. Thereore by the let cancellation law, f (eG ) = eH.
2 By defnition o inverse and property 1, f(a  a1) = f(a1  a) = f(eG ) = eH
or all a in G. By defnition o homomorphism,
f(a)  f(a1) = f(a1)  f (a) = eH or every a in G. Thereore, by the group
property o uniqueness o inverses, f(a1) = ( f(a)) 1.
3 In order or f (G) = { f (a)| a  G} ,{ f (G), } to be a subgroup o {H, } ,
the ollowing properties must hold:
Closure: This property ollows rom the defnition o homomorphism,
and rom closure o {G, } .
Identity: The range o fcontains the identity, i.e. eH = f(eG )  f(G).
Inverse: This ollows rom property 2 o homomorphisms, and rom
the inverses in {G, } .
So { f(G), } is the subgroup o {H, } .

Chapter 4 1 41
The proof of property 4 using mathematical induction is left as an exercise.

Property 3 is actually a corollary of part i of a more general theorem.


The proof of the theorem is left for you to do.
Let {G, } and {H,  } be groups with subgroups {G 0, } and {H0,  } respectively.
If f : G  H is a homomorphism, then
i f(G 0 ) = {f(x) | x  G 0}, {f(G 0),  } is a subgroup of {H,  }, and
ii f (H0 ) = {x  G | f(x)  H0}, {f1 (H0), } is a subgroup of {G, }
1

Example 8

Let f be a homomorphism rom group {G, } to {H, } . Furthermore,


let f be surjective. Prove that i {G, } is Abelian, then {H, } is Abelian.
Let c, d  H. You need to show that for
Since f is surjective, there exist elements a, b  G such that c, d  H, c  d = d  c.
f(a) = c and f(b) = d.
By defnition o homomorphism, f(a  b) = f(a)  f(b) = c  d.
Furthermore, since {G, } is Abelian,
f(a  b) = f(b  a) = f(b)  f(a) = d  c.
Hence, c  d = d  c

The kernel of a homomorphism


There is no simple way o showing that a homomorphism between
two groups is surjective. There is, however, an important theorem
that is useul in showing that it is injective.

Theorem 5
A homomorphism f:{G, }  {H, } is injective i and only i the
unique solution to f(x) = eH is x = eG.

Proof:
: f(x) = eH  f(x) = f(eG) by Theorem 4 and by the assumption
that fis injective, x = eG.
: Let x = eG be the only solution o f(x) = eH. Suppose that f (a) = f (b)
or a, b  G. Then, f(a) = f(b)  f(a)  f(b)1 = eH  f(a  b1 ) = eH.
Since a  b1 = eG, a = b and f is injective.

An injective homomorphism is called a monomorphism, and a surjective


homomorphism is called an epimorphism.

142 The classifcation o groups


In Example 7 we saw how the homomorphism f (x) = i x rom group
{, +} to group {S, } , S = {1 , i, 1 , i} , partitioned the set o
integers  according to the image o each integer in S, i.e.
{..., 8, 4, 0, 4, 8, ...}  1
{..., 7, 3, 1 , 5, 9, ...}  i
{..., 6, 2, 2, 6, 1 0, ...} 1
{..., 5, 1 , 3, 7, 1 1 , ...}  i
The set o elements rom , { , 8, 4, 0, 4, 8, } that are mapped
onto the identity in S, e = 1 , is called the kernel o the homomorphism f.
The ollowing denition thereore shows how the identity appears as
the value o a homomorphism.

Defnition
Given the group homomorphism f:{G, }  {H, } the kernel o
the homomorphism f, ker( f ), is dened as the set o all elements
o G which are mapped to eH, i.e. ker ( f) = {a  G| f (a) = eH} . ker(f) eH f(G)

You saw rom the example that the kernel o f(x) = i x ormed a
f
subgroup o {, +} . We shall now prove this observation. G H

Theorem 6
The kernel o a homomorphism f:{G, }  {H, } is a
subgroup o {G, } .

Proof:
We will show that the subgroup properties hold.
Identity: By Theorem 4, f(eG) = eH  eG  ker ( f ).
Closure: Let a, b  ker( f ) or some a, b  G. Then, by denitions
o homomorphism and kernel, f(a  b) = f(a)  f(b) = eH  eH = eH.
Hence, a  b  ker( f ).
Inverse: Let a  ker( f ) or some a  G. Then by property 2 o
homomorphism, f(a1 ) = ( f(a)) 1 = eH1 = eH. Hence, a1  ker ( f ).
Hence ker( f ) is a subgroup o {G, } , since associative property
holds or all the elements o G.
As an exercise, you may want to work out the same proo using a
diferent subgroup theorem, e.g. show that i a,b are elements o ker ( f ),
then a  b1 is an element o ker( f ).
It is interesting to note that the kernel can be useul in solving
equations. For example, consider the solutions or the equation z3 = 8i.
We can change this to an example with homomorphisms. Let us
consider f:{\{0} ,  }  {\{0} ,  } such that f (z) = z3 or z  \{0} .
We can easily show that f is a homomorphism,
since f (z1 z2) = (z1 z2) 3 = z1 3 z23 = f (z1 ) f (z2). Using De Moivres theorem,

we can nd one solution to the equation z3 = 8i, z1 = 2 cis   .
6 
Chapter 4 1 43
The elements o the kernel o the homomorphism are the solutions to the equation
 2   4 
z3 =  , since ker( f ) =  , f (z) = z3 . Thereore ker( f ) = K =  1, cis   , cis  .
  3   3 
Hence, the solutions to our original equation are elements o the coset
  5 3  
z K =  2 cis   , 2 cis   , 2 cis   .
 6   6   2 

Exercise 4D
1 a Show that f :{\{0} , }  {\{0} , } is a homomorphism,
and determine the kernel, when:
1
i f(x) = | x| ii f ( x ) =
x
b Show that f :{, +}  {, +} is not a homomorphism when:

i f(x) = the largest integer x ii f(x) = x + 1

2 Let {, +} and {C, +} be groups such that C is the set o continuous
1
unctions with domain [0, 1]. Show that f : C  , f (c) = 0 c (x)dx or
c  C, is a homomorphism.

3 Given two groups {, +} and { 2, + 2} , show that f:    2 is a


 0, x  even numbers
homomorphism i or x  , f ( x ) = 
 1, x  o dd numbers
4 Prove part 4 o Theorem 4 by mathematical induction.

5 Prove that the composition o homomorphisms is a homomorphism,


i.e. i f :G  H and g : H  K, then ( g  f ): G  K is a homomorphism.

6 Let f:{G, }  {H, } be a homomorphism. Prove the inverse


image o a subgroup o {H, } is a subgroup o {G, } .

7 Let f :G  H defne a group homomorphism. Let K = ker ( f ).


Prove f 1( f(a)) = {x  G | f (x) = f(a)} is the let coset aK o K and is
also the right coset Ka o K, i.e. the two partitions o G into let
and right cosets o K are the same. (Hint: Use the double inclusion
method or proving two sets are equal.)

4.4 Isomorphisms
In Chapter 3 we saw that there was only one way to construct
a Cayley table or a group o order 2 and a group o order 3. In other
words, interchanging a row or a column did not change any o the
results o the operation. For example, the Cayley table or a group o
order 3 is shown next, and beside it is the same table where the columns
or elements e and b have been interchanged.

144 The classifcation o groups


 e a b  b a e
e e a b e b a e
a a b e a e b a
b b e a b a e b

A quick check o the results o the operations on all elements in the


2nd table will show that the results are the same as those in the 1 st table.
Thereore we say that the two tables are structurally equivalent.
We now consider the question How many diferent groups o order 4
are there? Consider two Cayley tables that you have already seen
o two cyclic groups o order 4: S = {1 , i, 1 , i} under multiplication,
and { 4, + 4} .
 1 1 i i +4 0 1 2 3
1 1 1 i i 0 0 1 2 3
1 1 1 i i 1 1 2 3 0
i i i 1 1 2 2 3 0 1
i i i 1 1 3 3 0 1 2

Notice the ollowing:


 The orders o the groups are the same.
 The identity is in the frst row and column o each table.
 I we consider a unction  which maps the elements o the frst group
to the elements o the second group as ollows, 1  0, i  1 , 1  2,
 i  3, we obtain the 2nd table.
 The unction  maps the results o the binary operation on the
elements o the frst table onto the corresponding results o the binary
operation in the 2nd table.
To expand upon this last bullet point, let us consider  (1  i ).
We see that (1  i) = (i) = 3, i.e.  : i  3.
Furthermore, (1 ) + (i ) = 2 + 1 = 3, hence (1  i) = (1 ) + (i ).
I you test all the other pairs o elements you will see that
(a  b) = (a) +  (b), or a, b  S and  (a),  (b)  4.
Additionally we can consider the order o the elements in the tables.
These tables show the orders o the elements in both groups:
{S, }
element 1 i 1 i
order 1 4 2 4
{4, + 4}
element 0 1 2 3
order 1 4 2 4

Chapter 4 1 45
As you can see, each o these groups has an identity, one element o
order 2, and two generators, i.e. two elements o order 4.
The unction fmaps 1  0, i  1 , 1  2, i  3, i.e. fmaps elements
whose orders are equal onto each other.
We could reconstruct both tables using one o the generators,
e.g. i in the rst group and 1 in the second group.
In the rst group e = 1 = i 4, and the in second group e = 0 = 1 4,
i.e. 1 4 = 1 + 1 + 1 + 1 = 4(mod 4) = 0.
 e i i2 i3 +4 e 1 12 13
e e i i2 i3 e e 1 12 13
i i i2 i3 e 1 1 12 13 e
2 2 3
i i i e i 1 2
1 2
1 3
e 1
3 3 2
i i e i i 13 13 e 1 12

Then, both tables can essentially be expressed by the same table using
a generator a, and are structurally equivalent to this table.
 e a a2 a3
e e a a2 a3
a a a2 a3 e
a2 a2 a3 e a
a 3
a 3
e a a2
We could also have achieved the same efect by mapping the generators
diferently, e.g. f : i  1 and f : i  3. The identities are still corresponding
elements, as well as the only element o order 2 in both groups.
The mapping o these elements remains the same, i.e. f: 1  0 and f: 1  2.
We leave it to you to construct the Cayley tables using this new mapping,
i.e. the row and column o the elements i and i would need to be
interchanged. You will notice again that the new Cayley table is
structurally equivalent to the rst one we constructed. We conclude that all
cyclic groups o order 4 are structurally identical.
Is there a group o order 4 that is not cyclic, i.e. that is not structurally
equivalent to the cyclic group o order 4? Consider the Abelian group
in Example 9 rom Chapter 3:
{S,  1 2} , S = {1 , 5, 7, 1 1 } . Here is the Cayley table:

12 1 5 7 11
1 1 5 7 11
5 5 1 11 7
7 7 11 1 5
11 11 7 5 1

146 The classifcation o groups


The identity is 1 , and the order o the elements 5, 7 and 1 1 is 2.
This group is clearly not cyclic, although it is Abelian. Thereore this
group o order 4 is not structurally equivalent to our two cyclic groups
o order 4. In group theory this group is called the Klein our-group,
or KV, where the subscript V stands or the German word or our  vier.
Its defnition is KV = {a,b | a2 = b2 = (ab) 2 = e} . It is the smallest
non-cyclic group. Another example o the KV group is rom Chapter 3,
Exercise 3B, question 7(b), the symmetries o a rectangle.
This group is structurally equivalent to {S,  1 2 } . There are only
two groups o order 4, the cyclic group and the Klein our-group.
All groups o order 4 will be structurally equivalent to one o these
two groups, i.e. the cyclic group o order 4 or the KV group.

The Klein 4 - group is the subgroup V (Vierergruppe)


o the permutation group S 4. The group consists o
the ollowing 4 permutations written in cycle notation:
the identity permutation (1), (12)(34), (13)(24) and
(14)(23). The group is named ater the German
mathematician Felix Klein, who was an early pioneer
in Group Theory applied to Geometry. He also devised
the amous topological fgure, the Klein bottle, an
impossible fgure with no inside.

We will now defne what we mean by structurally equivalent.

Defnition
An isomorphism is a bijective homomorphism,
i.e. given groups {G, } and {H, } , f : G  H is an isomorphism i
and only i
i f is bijective, and
ii f is a homomorphishm, i.e. or all a, b  G, f (a  b) = f (a)  f (b).

In other words, to show that two groups are isomorphic, you must
show that the homomorphism is both injective and surjective.
The bijection guarantees that the sets have the same size, or cardinality,
and the homomorphism guarantees that the groups have the
same structure.
As you have already seen, to show that two fnite sets are isomorphic
we need only show that their Cayley tables are structurally equivalent,
i.e. their tables can be shown to be structurally the same by rearranging
or swapping columns or rows.
We will next show a worked-out example or fnite sets.

Chapter 4 1 47
Example 9
a Determine i any o the ollowing three groups are isomorphic by
constructing their Cayley tables.
 { + }
6, 6
 Symmetries o an equilateral triangle as defned in Chapter 3,
on page 1 01 , i.e. S = {I, R1 , R2, A, B, C}
 { \{0} ,  }
7 7
b Reconstruct the Cayley table(s) to show the equivalent structure o
the isomorphic groups.
a Creating the Cayley tables or each group:
+6 0 1 2 3 4 5
0 0 1 2 3 4 5
1 1 2 3 4 5 0
2 2 3 4 5 0 1
3 3 4 5 0 1 2
4 4 5 0 1 2 3
5 5 0 1 2 3 4

 I R1 R2 A B C
I I R1 R2 A B C
R1 R1 R2 I C A B
R2 R2 I R1 B C A
A A B C I R1 R2
B B C A R2 I R1
C C A B R1 R2 I

7 1 2 3 4 5 6
1 1 2 3 4 5 6
2 2 4 6 1 3 5
3 3 6 2 5 1 4
4 4 1 5 2 6 3
5 5 3 1 6 4 2
6 6 5 4 3 2 1

148 The classifcation o groups


All three groups have the same order. Since {S, } Notice the structures of the Cayley
is not Abelian, and the other two groups are, the table, e. g. Abelian.
only possibility or an isomorphism is between
{6, + 6} and { 7\{0} ,  7} .
The orders o the elements in each table are: Determine the orders of the elements
in both groups.
a  6 0 1 2 3 4 5
order 1 6 3 2 3 6
b  7 1 2 3 4 5 6
order 1 3 6 3 6 2

The groups are cyclic. The identities are Map elements of similar orders.
corresponding elements, hence 0  1 . The only
elements o order 2 are also corresponding elements,
so 3  6.
Mapping the generators 1  3 and 5  5 leaves two
possible mappings or the remaining elements: 2  2
and 4  4, or 2  4 and 4  2.
Mapping the generators 1  5 and 5  3 again
leaves the two possible mappings o 2  2 and 4  4,
or 2  4 and 4  2.
b Using the frst mapping, we can leave + 6 as is, and
reconstruct the 7 table. Rearrange one of the tables so
that the corresponding elements
7 1 3 2 6 4 5
are in the same positions.
1 1 3 2 6 4 5
3 3 2 6 4 5 1
2 2 6 4 5 1 3
6 6 4 5 1 3 2
4 4 5 1 3 2 6
5 5 1 3 2 6 4

How many groups o order 6 are there? You have worked with the
cyclic group o order 6 above. In Chapter 3 you worked with the
symmetries o the equilateral triangle, and saw that it ormed a
non-Abelian group o order 6. From a previous theorem we know
that i a group is cyclic, it must be Abelian, or i a group is not Abelian,
it cannot be cyclic. Again as with order 4, there are two distinct groups
o order 6.

Chapter 4 1 49
We can classiy fnite groups by isomorphism classes, i.e. the number
o distinct groups o a particular order.
 All groups o prime order are cyclic, hence there is only one class
o groups whose order is a given prime number.
 There are two distinct groups or classes o order 4 and order 6:
one cyclic group and one non-cyclic group.
 There are fve distinct groups o order 8, three o which are cyclic.
 There are two distinct groups o order 9, and both are cyclic.
 There are two distinct groups o order 1 0, one cyclic and one non-cyclic.
Although none o the above results are needed or examination purposes,
you might want to research this urther and investigate how many classes
o groups there are o a given order greater than 1 1 .
We shall now illustrate an example showing an isomorphism between infnite sets.

Example 10

Show that the mapping f : x  2 x rom the set o integers  to the set
S= { 1
8
,
1
4
,
1
2
}
, 1, 2, 4, 8,  is an isomorphism between the groups {, +} and {S, } .

Since f (x + y) = 2x+ y = 2 x  2 y = f (x)  f (y), Show that f is a homomorphism, i. e.


f defnes a homomorphism between determine if f maps the sum of two
the two groups. elements in  to the product of two
elements in S.
Method I
Show that f is bijective, i. e.
f is injective i f(a) = f(b)  a = b.
injective and surjective.
2a = 2 b  a = b, hence f is injective.
Method II
To show fis injective, we can make use o
Theorem 5, showing that the unique solution
o f (x) = es is e. Assume that the solution is not
unique, i.e. x  y such that f(x) = es and f(y) = es.
Hence, 2 x = 2 y  x = y, which is a contradiction.
Thereore the solution is unique, and is e = 0.
f is surjective i or every b  S there exists an a  
lo g b
such that f(a) = b. I 2 a = b  a = = log 2 b   .
lo g 2
Hence f is bijective.
Thereore f is an isomorphism.

150 The classifcation o groups


We see thereore rom Example 1 0 that to show f :{G, }  {H, }
is an isomorphism, we need to show that:
 f is a homomorphism
 ker( f ) = e , i.e. f is injective.
G
 f is surjective.

Since an isomorphism is a homomorphism, all the properties o a


homomorphism will apply. In addition to these, there is another
important property specifc to isomorphisms which you will already
have observed in the examples so ar.

Theorem 7
I f : {G, }  {H, } is an isomorphism, the order o a  G is equal to
the order o f(a)  H or every a  G.

Proof:
Let n be the order o a. Thereore by defnition, n is the smallest
positive integer such that an = e. Then:
(f (a))n = f (a)  f (a)  ...  f (a)
= f ( a
 a    a ) by defnition o isomorphism,
n times

n
= f (a )
= f (eG), since the order o a is n
= eH
We now must show that n is the least positive integer such that ( f (a)) n = eH.
Let the order o f (a) be m, m < n. Then, by defnition, ( f (a))m = eH .
Hence, eH is the image o both an and am . Since am = an = eG, and n is
the smallest such integer, then this is a contradiction with the assumption
that m < n. Hence n  m and the order o a  G is the same as the order
o f (a)  H.
This property is very useul or showing that two groups are not
isomorphic, i.e. i the orders o the elements o the groups do
not match, then the groups are not isomorphic. The ollowing example
illustrates how to use this property.
Example 11

Determine whether or not {, +} and { \ {0} , } are isomorphic.

The only element in  with fnite order is the identity, 0, Determine if all elements in both
whose order is 1 . groups have the same orders.

In  \ {0} , the identity 1 has order 1 , and the element 1


has order 2, i.e. there are at least two elements with fnite
order. Hence, the groups are not isomorphic.

Chapter 4 1 51
Is the converse o this theorem true, i.e. i the orders o the elements
o two groups are the same, are the groups isomorphic? This is
not true, and the smallest group with this property has order  6,
i.e. there exist two groups o order  6 whose elements have the same
orders but which are not isomorphic. You may want to research
this very important result urther.

Exercise 4E
1 Determine which o the ollowing groups o order 6 are isomorphic.
a The symmetry group o the equilateral triangle
b The set {1, 2, 4, 5, 7, 8} under  9
c The set {2, 4, 6, 8, 10, 12} under  14
d The permutation group with the ollowing elements:

1 2 3 4 5  1 2 3 4 5  1 2 3 4 5 
p1 =   ; p2 =  4 5 3 1 2  ; p3 =  3 2 4 1 5 
1 2 3 4 5     

1 2 3 4 5  1 2 3 4 5  1 2 3 4 5 
p4 =   ; p5 =  3 5 1 4 2 ; p6 =  
4 2 1 3 5    1 5 4 3 2 

2 Prove that the mapping f :  +  , f (x) = ln x, is an isomorphism


between the groups { + , } and {, +} .

3 Prove that the mapping f : x  x1 is an isomorphism o a group {G, }


onto itsel i and only i {G, } is Abelian.

4 Let {H, } be a subgroup o {G, } , and let M = {x1hx | h  H} be a


subset o G or some given element x  G.
a Prove that M is closed under the operation  o the group G and
that each element o M has an inverse under  in M.
Hence, deduce that M is a subgroup.
b Show that M is isomorphic to H.
a b
5 Let (ab) denote a cycle defned by the permutation  
b a
a Consider the permutations on the set {1, 2, 3, 4} . Let p1 = (1),
p2 = (12)(34), p3 = (13)(24) and p4 = (14)(23) be our o these
permutations. Write out the Cayley table or the set
P = { p1, p2, p3 , p4} under composition o permutations.
b Prove that {P, } is an Abelian group.
c Determine whether or not {P, } is isomorphic to { 4,+ 4} .

6 Given that two groups {G, } and {H, } are isomorphic, prove that
{G, } is Abelian i and only i {H, } is Abelian.

152 The classifcation o groups


Review exercise
EXAM-STYLE QUESTIONS
1. a Let G be a group o order 6 such that it contains no elements o
order 6. State Lagranges theorem and hence prove by contradiction
that at least one o the elements will have order 3.
b Let G be o order n, and g be an element o G that has order k.
Write down a cyclic subgroup o order k and use Lagranges
theorem to show that gn = e.
2 a Let { n, + n} be the cyclic group o integers under + n. Write down
the elements o this group, and identiy a generator or the group.
b Let { n, } be the group whose elements are the nth roots o
unity under multiplication. Write down the elements o this
group, and show that the group is cyclic. Write down a generator
o the group. 2 ix 

c Show that f :{ n, + n}  { n,  } , f ( x ) = e n , x  n


is an isomorphism.
3 Let G be a set o isomorphic groups, i.e. G1  G2 (G1 is isomorphic to G2)
or all G1,G2  G. I f : G1  G2, show that the relation on G dened
by  is an equivalence relation.
4 Prove that the mapping f : x  x2 is an isomorphism o a group {G, }
i and only i {G, } is Abelian.
5 Prove that i f :{G, }  {G, } is a homomorphism with kernel K then
f (x) = f ( y) i and only i y = xk or some k  K.
6 Let G be the group o permutations S3 and H is a subgroup o
G such that H = {(1), (12)} . Find the let and right cosets o H in G.
7 Let G be a group. Prove that the relation on G dened as xRy  x = y or
x = y1 is an equivalence relation, and write down the equivalence classes.
8 Prove that the groups {, +} and {+ , } are isomorphic.
9 a Show that {S, } , S = {2 a3 b | a, b = } , a, b   orms a group.
b Show that {S, } is isomorphic to the group {, +} ,
 = {a + bi | a, b  Z, i = 1 } .
10 Express p = (13256)(23)(46512) as a product o disjoint cycles.
11 Explain why f :  12   10, f (x) = 3x (mod 10) is not a homomorphism.
12 Given the permutation group S3 , nd a subgroup {H, } and an element
o g such that gH  Hg.
13 Let G ={\{0} , } .
a Show that f: x  x n is a homomorphism or all n   + .
b Determine ker( fn).
c Determine n so that the mapping is an isomorphism.
14 Let S be the set o polynomials in x with real coefcients under addition.
Dene the mapping f : p(x)  P (x) =  p ( x ) dx such that P (0) = 0.
Show that f is a homomorphism, and determine its kernel.
Chapter 4 1 53
Chapter  summary
A permutation o a non-empty fnite set A is a bijection rom A to A.
Theorem 1: Let A be a non-empty set o n elements, and let Sn be the set o all
permutations o A. Then Sn orms a group under composition o permutations.
Let A be the fnite set { , 2, 3,  , n} . The group o all permutations o A is the
symmetric group on n elements and is denoted by Sn.

Permutations and cycle form


 Every permutation can be written as a product o disjoint cycles.
 Disjoint cycles are commutative.
 The order o a permutation written as a product o disjoint cycles is
the least common multiple o the lengths o the cycles.
Let {H, } be a subgroup o {G, } and let x  G. Then the set o elements
xH = {xh h  H} is called a left coset o {H, } in G. The set o elements
Hx = {hx h  H} is called a right coset o {H, } in G.
Properties of cosets: For any subgroup {H, } o a group {G, } :
1 G is the union o disjoint cosets o {H, } .
2 Every coset (let or right) o a subgroup {H, } has the same number o elements as H.
3 The group is partitioned by the let (or right) cosets o its subgroup.
4 Every element o G lies in one o the cosets o H in G.

Lagranges theorem: I {H, } is a subgroup o {G, } , then the order


o the subgroup {H, } is a divisor o the order o {G, } .

Corollaries to Lagranges theorem:


1 The order o an element o a fnite group divides the order o the group.
2 Every group o prime order is cyclic.
Let {G, } and {H, } be groups. A homomorphism is a unction f : G  H such that
f (x  y) = f (x)  f ( y) or all x, y  G.
Properties of homomorphisms: Let f be a homomorphism rom group {G, } to group
{H, } . Let a be an element in G. Then the ollowing properties hold.
1 The homomorphism maps the identity in group G onto the identity in group H,
i.e. i eG and eH are the identity elements in {G, } and {H, } respectively,
then f (eG ) = eH .
2 The homomorphism maps the inverse o an element in group G to the inverse o the
elements image in group H, i.e. or all a  G, f (a1) = ( f (a)) 1.
3 The range o the homomorphism fis a subgroup o {H, } , i.e. or
f (G) = { f (a)  a  G} , { f (a), } is a subgroup o {H, } .
4 The homomorphism preserves all powers, i.e. f (an ) = ( f (a)) n or all a  G.

154 The classifcation o groups


Theorem: A homomorphism f :{G, }  {H, } is injective i and only i the
unique solution to f (x) = eH is x = eG.
Given the group homomorphism f :{G, }  {H, } the kernel o f, ker( f ),
is defned as the set o all elements o G which are mapped to eH ,
i.e. ker( f ) = {a  G f (a) = eH}
Theorem: The kernel o a homomorphism f :{G, }  {H, } is a subgroup o G.
Given the groups {G, } and {H, } , f : G  H is an isomorphism i and only i
i f is bijective, and
ii f is a homomorphism, i.e. or all a, b  G, f (a  b) = f (a)  f (b).

Theorem: I f :{G, }  {H, } is an isomorphism, the order o a  G


is equal to the order o f (a)  H.

Chapter 4 1 55
Exercise D
Answers 1 A  B = {(1, p), (1, q), (2, p), (2, q), (3, p), (3, q)}
Chapter 1 B  A = {( p, 1), ( p, 2), ( p, 3), (q, 1), (q, 2), (q, 3)}
Skills check The two products are not equal since the
1 a 24 Cartesian product is made up of ordered pairs
and hence (1, q)  (q, 1) etc
Exercise A 2 a Tabulate the Cartesian product A  B
1 a A \ B = {b, c, d}
AB 1 2 3 4 5 6
b B \ A = {i, o, u}
c A  B = {b, c, d, i, o, u} 1 (1, 1) (1, 2) (1, 3) (1, 4) (1, 5) (1, 6)
d (A  B) \ (A  C) = {a, e} 2 (2, 1) (2, 2) (2, 3) (2, 4) (2, 5) (2, 6)
e A  (B  C) = {a, b, c, d, e} 3 (3, 1) (3, 2) (3, 3) (3, 4) (3, 5) (3, 6)
Exercise B 4 (4, 1) (4, 2) (4, 3) (4, 4) (4, 5) (4, 6)
1 Many examples possible, such as: The ordered pairs in bold represent A  A  A  B
P = {{red cards} ,{black cards} } b i R = { (1, 1 ), (1, 2 ), (1, 4 ), (1, 6 ), (2, 1 ), (2, 3 ), (2, 5 ),
P = {{number cards} , {picture cards} } (3, 2 ), (3, 4 ), (4, 1 ), ( 4, 3 )}
2 a P is a partition.
b Q is not a partition. ii R = {(1, 1 ), ( 2, 4 )}
c B is not a partition since 2 is an element iii R = { (1, 3 ), (1, 4 ), (1, 6 ), (2, 4 ), (2, 5 ), (3, 5 ),
of both sets. (3, 6 ), ( 4, 6 )}
3 a Partition b Partition c Partition
iv R = {(1, 1 ), (1, 2 ), (1, 3 ), (1, 4 ), (1, 5 ), ( 2, 1 ), ( 2, 2 ),
4 Many examples possible such as
( 2, 3 ), ( 2, 4 ), (3, 1 ), (3, 2 ), (3, 3 ), ( 4, 1 ), ( 4, 2 )}
a {{x  , x  10} , {x  | x  {0, 1, 2, 3, ,10} } }
or {{ , e} , { x  | x   , x  e } } 3 A  B = {( a, p ), ( a, q ), (b, p ), (b, q )}
b {{ x | x   } , { x | x   , x   }} n ( A  B ) = 4  n ( P ( A  B )) = 2 4 = 1 6
or {{primes} , {x| x  R, x is not a prime 4 a A  B = {( a, 1 ), ( a, 2 ), ( a, 3 ), (b, 1 ), (b, 2 ), (b, 3 )}
number} }
A  C = {( a, 3 ), ( a, 4 ), (b, 3 ), (b, 4 )}
c {..., [  3,  2[, [  2, 1 [, [ 1 , 0[, [0, 1 [, [1 , 2[,
( A  B )  ( A  C ) = {( a, 3 ), (b, 3 )}
[2, 3[, ... }
b B  C = {3 }
 A  ( B  C ) = {( a, 3 ), (b, 3 )}
Exercise C c (A  B)  (A  C ) = A  (B  C )
6 a AB 5 A  C = {( a, c ) | a  A, c  C}
B  C = {(b, c ) | b  B, c  C}
Since A  B it follows that a  A  a  B
A B  ( a, c )  B  C for all a  A , c  C .
Therefore A  C  B  C .
6 List the Cartesian product S  S
S S 0 2 4 6 8
A '  B'
0 (0, 0) (0, 2) (0, 4) (0, 6) (0, 8)
U
2 (2, 0) (2, 2) (2, 4) (2, 6) (2, 8)
4 (4, 0) (4, 2) (4, 4) (4, 6) (4, 8)
A B 6 (6, 0) (6, 2) (6, 4) (6, 6) (6, 8)
8 (8, 0) (8, 2) (8, 4) (8, 6) (8, 8)

The elements of R are the ordered pairs in bold.

156 Answers
10 R = {(2, 1), (4, 2), (8, 3), (16, 4), (32, 5), (64, 6), 6 Equivalence classes
(128, 7), (256, 8), (512, 9), (1024, 10)} [1 ] = { x | x 2  1 = 3 k, k   + } = {1, 2, 4, 5, 7, 8, 1 0, 1 1 ...}

xR 1 y  y = lo g 2 x =
ln x [3 ] = { x | x 2  3 = 3 k, k   + } = {3, 6, 9, 1 2, 1 5 ....}
ln 2
1
n(R ) = 1 0 7 R partitions the Cartesian plane into lines parallel
to the y-axis.
Exercise E 8 [(1, 2)] = { ( x , y ) | 2 x = y } = { (1, 2), (2, 4 ), (3, 6), (4, 8). . . }
1 Since R is reexive, symmetric and transitive it b
[ ( a , b ) ] = { ( x , y ) | bx = ay } = { ( x , x )} , which represents
ollows that it is an equivalence relation. a
2 Since R is reexive, symmetric and transitive it sets o straight lines passing through the origin.
ollows that it is an equivalence relation. 1
9 [ (1, 1 ) ] = { ( x , y ) | xy = 1} = { ( x , )| x   \ {0} }
3 Since R is not transitive it ollows that it is not an x

equivalence relation. ab
[ ( a , b ) ] = { ( x , y ) | xy = ab } = { ( x , ) | x   \ {0} } , which
4 Since R is not reexive and not transitive it x
represents a set o rectangular hyperbolas with
ollows that it is not an equivalence relation.
the x and y axes as asymptotes.
5 Since R is reexive, symmetric and transitive it
ollows that it is an equivalence relation. 10 b [ 0 ] = { x | x  0  } = 
9 One example rom not reexive or not symmetric
is enough to show that R is not an equivalence
relation.
c 3
 4  { 3
= x | x  = n, n  
4 }
10 Since R is reexive, symmetric and transitive it {
= ...,
9 5 1 3 7 1 1
4
,
4
, , , , , ...
4 4 4 4
}
ollows that it is an equivalence relation
a = {x | x  }
a
d  b  = n, n  
b

= {x | x = , n  }
Exercise F nb + a
1 a The equivalence classes induced by R: b
[set] = {set, car, sea, sun} R partitions  into ractions with denominator
[bike] = {bike, wave} b and numerator an innite arithmetic
[table] = {table, chair} progression depending on a and with
[tennis] = {tennis, stairs} common diference b.
b The equivalence classes induced by R:
[set] = {set, stairs, sea, sun}
Review Exercise
[table] = {table, tennis}
[chair] = {chair, car} 1 i A\B
[bike] = {bike} U
[wave] = {wave}
2 a R partitions the set o line segments into sets
o segments o equal length. A B
b R partitions the set o all polygons into sets
o polygons with same number o sides,
i.e. {triangles} , {quadrilaterals} , etc
3 R partitions the set o parabolas into sets A  (U \B)
containing parabolas with vertex tangent to the U
line y = c.
4 The relation partitions    into concentric
circles with centre at the origin. A B
5 Equivalence classes:
[1 ] = { x | x + 2 = 3 k, k   + } = {1, 4, 7, 1 0,...}
[ 2 ] = { x | x + 4 = 3 k, k   + , k  2} = { 2, 5, 8,...}
In the lower o the two diagrams above, the area
[3 ] = { x | x + 6 = 3 k, k   + , k  3} = {3, 6, 9,...}
shaded in both directions represents A  (U \ B ).

Answers 1 57
ii Similarly, Venn diagrams demonstrating 9 b First consider the equivalence class [(a, 0)] where a
two expressions that are each the symmetric is a constant
diference o A and B. [(a, 0 ) ] = { ( x , y ) | x = a , y  0 = 5 k } = { 0,  5,  1 0,  1 5, . . . }
2 Venn Diagrams suitably drawn to show
[ ( a, 1) ] = { ( x , y ) | x = a , y  1 = 5 k } = { . . . ,  1 4,  9,  4, 1, 6, 1 1, . . . }
i A  (B  C ) = (A  B)  (A  C )
[ ( a, 2) ] = {( x , y )| x = a, y  2 = 5 k } = {...,  1 3,  8,  3, 2, 7, 1 2, ...}
ii A  (B  C ) = (A  B)  (A  C )
3 b The relation R partitions the Argand [ ( a, 3) ] = {( x , y )| x = a, y  3 = 5 k } = {. .. ,  1 2,  7,  2, 3, 8, 1 3, . .. }
diagram into lines passing through the [ ( a, 4 ) ] = {( x , y )| x = a, y  4 = 5 k } = { ... ,  1 1,  6,  1, 4, 9, 1 4, ... }
origin, since or every each particular angle,
 , all the complex numbers having  as an 10 S is not transitive.
argument lie on a straight line passing 11 b AB R CD i and only i AB and CD are
through the origin and make an angle  parallel line segments o the same length
with the positive real axis. and oriented in the same direction. Thus an
4 a A = {2, 3, 5, 7, 11, 13, 17, 19} equivalence consists o all translations o
B = {1, 2, 3} a given directed line segment.
C = {0, 1, 2, 3}
D = {2, 0, 2} Chapter 2
E = {1, 0, 1, 2, 3} Skills Check
b i True because n(A) = 8, n(D) = 3, n(E ) = 5 1 a 0 < t< 3 b t> 3
ii False because D  A = { 2, 0} (3  2 x ) ln x
2 a f 1 ( x ) = where x  1 b f 1 ( x ) =
 n ( D  A ') = 2 (x 1) ln 2

f 1 ( x ) = ln  ) 
1
iii True as evident rom list above. c
2
(x + x2 + 8
iv True D \ B = {2, 0} and neither o these
elements are in A Exercise A
v False C  E = (C  E ) \ (C  E ) = { 1} 2 a Not a unction b Not a unction
c Not a unction
5 a i R is not reexive
3 a Is a unction b Not a unction
ii R is symmetric c Not a unction
iii R is not transitive. 5 a not surjective b injective
b R is not an equivalence relation because it is 6 i injective, not surjective
not reexive and it is also not transitive. ii injective, not surjective
iii injective and surjective
6 a Since R is reexive, symmetric and transitive 8 i surjective, not injective
it is an equivalence relation. ii surjective, not injective
b i C 0 = [ 0 ] = { x | x 3  0 (mod 5 )} = { 5 , 1 0 , 1 5 , 20 . . . } iii not surjective, not injective
9 a Range 1 < f (x) < 1
ii C1 = [1 ] = { x | x 3  1 (mod 5} = { x | x 3 = 5 k + 1}
b y
= {6, 11, 16, 21, 26, 31, ...}
2
7 a ii [z2  3z + 4] consists o all polynomials o 1
the orm z2  3z + c
0 x
b ii [z2  3z + 4] consists o all polynomials o 10 8 6 4 2 2 4 6 8 10
1
the orm z2 + bz + 4
2
8 b [2] = {2, 4, 6,  }
[1] = {1, 3, 5, 7 } 10 a f is not surjective b fis injective
c 5 355
 5(mod 8) c g is surjective d g is injective

158 Answers
2 2
Exercise B 12 a e (ln(2x1)) b ln(2e x  1)
2
1 g  f ( a, b ) = b c 2 ln (2x  1) d e 4(ln(2x1))
3 a i f  g( x ) = x 2 ii g  f( x ) = 2 x e 2e (ln(2x1))2
b f  g is not injective and not surjective g   fis Exercise C
injective and surjective
2
4 a f 1(n, 1) = n + 1 1 a ( f  g)(x) = x  2, ( g  f ) ( x ) = x + , no
3
c f  g ( n , m ) = ( n + m  1, 1 ), g  f ( n ) = n b ( f  g)(x) = x , ( g  f ) ( x ) = x , yes
5 a not injective and not surjective therefore not
bijective. c ( f  g)(x) = x , ( g  f ) ( x ) = x , yes

b f  f ( x, y ) = ( xy ( x + y ), x + xy + y ) 2 a f 1(x) = e x
x 1
6 b f  g( x ) = 1 g  f( x ) =  x if x is rational
1x x b f 1 ( x ) = 
d fand g are both self inverses   x if x is irrational
1
e f g and g  fare inverses of each other 3 a i ( f  g) ( x ) = arccos(ln x )

7 b f 1 ( x , y ) = 

3
y y 
, x3 
ii (g 1
 f 1 ) ( x ) = arccos(ln x )
 x x 

8 a f (x) = 8e > 0 for all x  [1,  [


2x
Exercise D
 x +3  1 a Binary operation  not closed
b f 1 ( x ) = ln  
 4  b Binary operation  closed
c Binary operation  closed
9 a y
d Binary operation  closed
10
f(x) = ln x 2 S is not closed under addition or multiplication
5
or division
0 x 3 i A is closed under addition and multiplication
20 15 10 5 5 10 15 20
x 5 ii B is not closed under addition but closed
f(x) =
e 10
under multiplication
b The function is steadily increasing over the 4  0 1 2 3  0 1 2 3
whole range and so it is a bijection. 0 0 1 2 3 0 0 0 0 0
ex x 1 1 1 2 3 0 1 0 1 2 3
c f 1 ( x ) = 
e
x
x >1 2 2 3 0 1 2 0 2 0 2

3 3 0 1 2 3 0 3 2 1
10 a i  ii {0}
b 1 Closed Closed
c k = 1  solutions (1, 0) and (1, 0) 5 X is closed under addition
k = 2  no solutions X is closed under subtraction
11 a 16  n(mod 8) = n(mod 8) X is closed under composition

b
n
 8 if n is even 6  1 1 i i
2
1 1  1 i i
| n  7| if n is odd
n 1 1 1 i i
c 8 if n is even
2 i i i 1 1
17  n if n is odd i i i 1 1
n
d  (m o d 8 ) if n is even
2 7  + is closed under 
1  n(mod 8) if n is odd
e | n(mod 8)  8| 8  1 2 3  1 2 3
n 1 1 1 1 1 1 1 2
f if n is even
2 2 3 4 5 2 1 1 2
| 9  n| if n is odd
3 5 7 9 3 2 2 4
Not closed Not closed

Answers 1 59
9 Not closed under addition Identity e = 0
Closed under multiplication Inverse of 0 is 0
10 a S  S = {(1, 1), (1, 2), (2, 1), (2, 2)} Inverse of 1 is 3
b  1 2 Inverse of 2 is 2
Inverse of 3 is 1
1 3 6 Not closed
4 a 2 4 6 8
2 6 12
2 4 8 2 6
c  (1, 1 ) (1, 2 ) ( 2, 1 ) ( 2, 2 ) 4 8 6 4 2
(1, 1 ) (3, 3 ) (3, 6 ) ( 6, 3 ) ( 6, 6 ) 6 2 4 6 8
(1, 2 ) (3, 6 ) (3,, 1 2 ) ( 6, 6 ) ( 6, 1 2 ) 8 6 2 8 4
( 2, 1 ) ( 6, 3 ) ( 6, 6 ) (1 2, 3 ) (1 2, 6 ) Identity e = 6
( 2, 2 ) ( 6, 6 ) ( 6, 1 2)) (1 2, 6 ) (1 2, 1 2 ) Inverse of 6 is 6
Inverse of 2 is 8
Exercise E
Inverse of 4 is 4
1 a not commutative not associative
Inverse of 8 is 2
b commutative not associative
b Identity e = 2
c not commutative not associative
Inverse a 1 = 4
3  f1 f2 f3 f4 a
f1 f1 f2 f3 f4 c There is no identity.
f2 f2 f1 f4 f3
6 b a
f3 f3 f4 f1 f2
f4 f4 f3 f2 f1 Review Exercise
From Cayley table we can see that S is closed 3 Bijection
 
because every element in Cayley table is in S.
f 1 ( x , y ) = 
x
, arcsin y 
Composition of functions is commutative 
 (1  y ) 2 

because the main diagonal of the Cayley table is
a line of symmetry. 4 Identity e = (1, 0)
4 a not commutative b not associative. 6 a Not injective and not surjective.
b The function f becomes invertible when
Exercise F
the domain is restricted to [0, ) and the
1 commutative associative e= 0 co-domain to [ 1 , 19
].
2 commutative associative e = (1, 1) 2 6
3 commutative associative e = (0, 0)  ln( x  1 ) 
 6 
4 not commutative associative e = (1, 0) f 1 ( x ) = arccos  
 ln 3 
5 commutative associative no identity  

Exercise G 7 a i commutative
1 b Identity = 1 ii associative
c Inverse a1 = a  2 b Identity e = 0
2 c Identity = 1 8 a Range  1 + 12 , 1 + e 2 
a  bi  e 
d Inverse ( a + bi ) 1
= 2 2
a +b b i Not injective since f (x) = f (x + 2n), n  .
3 * 0 1 2 3 ii Not surjective since the range of f (x)  
0 0 1 2 3 e.g. there is no x   such that f (x) = 10.
1 1 2 3 0
k = , A =  1 + 2 , 1 + e 2 
1
c i
2 2 3 0 1  e 
3 3 0 1 2 ii g 1 ( x ) = arccos ln x  1 ( )
The main diagonal is a line of symmetry, so the
iii x   1 + 2 , 1 + e 2 
1
operation is commutative.  e 

160 Answers
Chapter 3 4 {Z 5, + 5} :

Skills check +5 0  2 3 4
2 1 x
1 a f ( g ( x )) = ln( x + 1 ) b f ( g ( x )) = e 1 0 0  2 3 4

c g ( f ( x )) = [ln( x + 1 )]
2
d 1 1
f ( g ( x )) = e
x
1   2 3 4 0
2 2 3 4 0 
2 a The binary operation on the given set is
closed. No other properties hold. 3 3 4 0  2

b The binary operation on the given set is 4 4 0  2 3


closed and commutative. {Z 5 \ {0} ,  5} :
c The binary operation on the given set is
closed and commutative. 5  2 3 4
  2 3 4
Exercise A
2 2 4  3
3 a not a group
b not a group 3 3  4 2
c is a group 4 4 3 2 
d is a group
a x= 4 b x= 4 c x= 3
5
4 c d x= 1 e x= 2
3

5 b i (1, 2) ii (0.75, 2.8) 5


 10 2 4 6 8
c not Abelian
2 4 8 2 6
Exercise B 4 8 6 4 2
1 a 6 2 4 6 8
 e x y z
e e x y z 8 6 2 8 4

x x y z e It does orm an Abelian group.


1 1
y y z e x 6 1;  (1 + i 3 );  (1  i 3 )
2 2
z z e x y 7 a R = refection in the median
I = R2
b i y ii e

2 a i e ii e  I R
I I R
iii b iv c
R R I
b The identity element is a.
c right inverses b Symmetries o the Rectangle
I: Identity Transormation
x a b c d e X: Refection in the x-axis
x 
a d b c e Y: Refection in the y-axis
H: Rotation o 180 degrees about its center.
let inverses
x a b c d e  I X Y H
x  a c d b e I I X Y H

d Let and right inverses are not equal; X X I H Y


 is not associative. Y Y H I X
H H Y X I

Answers 1 61
c Symmetries o a cuboid 2 a
a  2 4 7 8  3 4
I: Identity transormation
X: Refection in the y-z plane a  8 4 3 2  7 4
Y: refection in the x-z plane order  4 2 4 4 2 4 2
Z: refection in the x-y plane
H1: rotation o 180 degrees about x-axis b a = 4; b = 8
H2: rotation o 180 degrees about y-axis c {1, 4, 7, 13} or {1, 4, 11, 14}
H3 : rotation o 180 degrees about z-axis Exercise 3E
C: central inversion
4 a 20; (0, 1) b (1, 1); (1, 2)
 I X Y Z H H2 H3 C c 4 elements have order 4: (0, 1); (0, 3);
I I X Y Z H H2 H3 C (1, 1); (1, 3)

X X I H3 H2 C Z Y H Review exercise
7
1 a 1 b x= 
Y Y H3 I H Z C X H2 4

Z Z H2 H I Y X C H3 3 a x = a1cb1 b x = b1a
1
5 a f4 ( x ) = 1  ; f5 ( x ) = 1 ; f6 ( x ) = x
H H C Z Y I H3 H2 X x x 1 x 1
H2 H2 Z C X H3 I H Y b Let f1 = 1; f2 = 2; etc.

H3 H3 Y X C H2 H I Z   2 3 4 5 6
C C H H2 H3 X Y Z I   2 3 4 5 6
2 2  4 3 6 5
9
+2 (0, 0) (0,  ) ( , 0) ( ,  ) 3 3 5  6 2 4
(0, 0) (0, 0) (0,  ) ( , 0) ( ,  ) 4 4 6 2 5  3
(0,  ) (0,  ) (0, 0) ( ,  ) ( , 0) 5 5 3 6  4 2
( , 0) ( , 0) ( ,  ) (0, 0) (0,  ) 6 6 4 5 2 3 
( ,  ) ( ,  ) ( , 0) (0,  ) (0, 0)
c
f  2 3 4 5 6
{Z 2  Z 2, + 2} does orm a group. order  2 2 3 3 2

Exercise C d { f1, f4, f5}


3 b x 3
7 a order = 3; order = 12; order = 4
Exercise D b 1, 5, 7 and 11
8 b The operation # is closed, associative
1 a {, A} ; {, B} ; {, C}
and has an identity e = 1. Not all elements
b {p, r}
have inverses.
c Symmetries o the Rectangle
10 Order o the group is 6.
I: identity Transormation
Subgroups: {e} , {e, a, a2} ; {e, b} ; {e, ab} ;
X: Refection in the x-axis
{e, a2b} ; {e, a, a2, b, ab, a2b}
Y: Refection in the y-axis
H: Rotation o 180 degrees about its center. 11 Many answers possible, such as:
 e a b c d f
 I X Y Z
e e a b c d f
I I X Y Z
a a e c d f b
X X I H Y
b b c e f a d
Y Y H I X
c c d f e b a
Z H Y X I
d d f a b e c
Subgroups: {I, X} ; {I, Y } ; {I, H}
d {8, 10} ; {4, 10, 16} f f b d a c e
e {0, 3} ; {0, 2, 4} The Latin Square is not associative.

162 Answers
Chapter 4 b
1 2 3 4 5 6 7 
 
Skills check 2 1 4 5 3 7 6 
1 a
R partitions Z into two sets: even integers and 1 2 3 4 5 6 7 8 
odd integers. c  
1 4 7 5 2 6 8 3 
b Each ordered pair (a, b)  Z  Z belongs to
an equivalence class consisting of all ordered 1 2 3 4 5 6 7 8 9 
d  
pairs with integer coordinates lying on a 1 2 4 5 7 8 3 9 6 
vertical line passing through (a, b).
c The partition of S induced by R is {{1, 5, 9} , 3 a (163)(24)
{2, 6, 10} ,{3, 7} ,{4, 8} } b (1236)(45)
3 a not bijective c (1632)(45)
d (1632)(45)
1  2 a + b a  2b 
b f (a, b ) =  ,  Exercise C
 5 5 
2 a Left cosets: 4Z; 1 + 4Z; 2 + 4Z; 3 + 4Z
Exercise A
Right cosets are the same as the left cosets.
2 3
b Left cosets: 4Z; 2 + 4Z
  1 2 3  1 2 3  1 2 3  
3  ,  ,   . Right cosets are the same as the left cosets.
  1 2 3  2 3 1  3 1 2  
c H = {0, 4, 8}
1 2 3 1 2 3 Left cosets: H; 1 + H = {1, 5, 9} ;
Both   and   are generators.
2 3 1 3 1 2 2 + H = {2, 6, 10} ; 3 + H = {3, 7, 11}
Right cosets are the same as left cosets.
1 2 3 4 5
4 a i  =  3  d
 4 1 2 5 
xG Left coset xH Right coset Hx
1 2 3 4 5
ii  =   ( ) {( ), ( 2)} {( ), ( 2)}
1 4 5 2 3 
( 3) {( 3), ( 23)} {( 3), ( 32)}
1 2 3 4 5
iii   2
=   (23) {(23), ( 32)} {(23), ( 23)}
4 1 5 3 2 
3 a H = {(0, 0), (1, 0)} ; Z 2 x Z 3 = {(0, 0), (1, 0),
1 2 3 4 5
iv  1 =  2  (0, 1), (1, 1), (0, 2), (1, 2)}
 4 3 1 5  b Left cosets: H = {(0, 0), (1, 0)} ;
1 2 3 4 5 (0, 1) + H = {(0, 1), (1, 1)} ;
v ( )  1 =   (0, 2) + H = {(0,2), (1,2)} .
 4 1 3 2 5 

Right cosets are the same as the left cosets.
1 2 3 4 5 Left and right cosets are equal: Z 2  {0} ;
vi   1 =  5 
 4 2 1 3  Z 2  {1} ; Z 2  {2}

1 2 3 4 5 Exercise D
b i x =  
5 2 3 1 4   a i { , } ii { }

1 2 3 4 5 Exercise E
ii x =  
4 1 2 5 3  1 a and d; b and c
Exercise B 5 a Let p1 = 1; p2 = 2; p3 = 3; p4 = 4
1 a x = (1645)(23); y = (13)(24)(5678);
 1 2 3 4
z = (23)(45)(67)
1 1 2 3 4
b x 1 = (1546)(23); y 1 = (13)(24)(5876);
z 1 = (23)(45)(67) 2 2 1 4 3
c order = 4; order = 4; order = 2 3 3 4 1 2
1 2 3 4 5 6  4 4 3 2 1
2 a  
2 3 1 6 5 4  c The groups are not isomorphic.

Answers 1 63
Review exercise 10 (124)(35)
2 a {0, 1, 2,  , n  1} ; 1 is a generator. 12 H = {(1), (12)} and g (13), for example.
b {1, ,  2,  ,  n1} ;  is a generator.  {1 } if n is o dd
13 b ker( fn ) = 
6 right cosets: {e, (12)} ; {(13), (132)} ; {(23), (123)}  {1,  1 } if n is even
left cosets: {e, (12)} ; {(13), (123)} ; {(23), (132)} c fn is an isomorphism when n is odd.
7 [x] = {x, x1} 14 ker( f ) only contains the zero polynomial.

164 Answers
Index

Page numbers in italics reer to review exercises.

A cyclic groups 11419, 122 unctions as relations 4850


Abel, Niels Henrik 88, 89 cyclic subgroups 117, 123 identity element e 789, 85
Abelian groups 89, 121 defnition 115, 117 identity unctions 56, 702, 85
associative law 17, 45 Theorem 10 118 injective unctions 50, 84
associativity 76, 85, 143 Theorem 6 116 inverse unctions 615
Theorem 7 11617 inverse o an element 7981,
B Theorem 8 117 85
bijection 52, 127, 154 Theorem 9 117 onto unctions 51
binary operations 725, 85 properties o binary
D operations 767
defnition 73
properties o binary De Morgans Laws 16, 20 properties o unctions 669
operations 767 associative law 20 surjective unctions 51, 84
Bourbaki, Nicolas 47 distributive law 20
G
Dedekind, Richard 47
C Descartes, Ren 21 Galois, Evariste 88, 89
cancellation laws 81, 1058, 122 disjoint sets 7, 15, 44 generators 115
Theorem 10 812 distributivity 76 Global Positioning Systems
Cantor, Georg 3, 4, 47 domains 48, 84 (GPS) 22
cardinality 4, 44 Gdel, Kurt 3
Cartesian plane 21, 22 E group structures 126
Cartesian product o two sets elements 48 Group Theory 8889
2122, 45 inverse o an element 7981, 85 groups 8990, 121
defnition 22 empty set 4, 44 Abelian groups 89, 121
Cauchy, Augustin-Louis 47, 126 epimorphism 142 Cayley tables 97
Cayley tables 73, 97, 1447 equal sets 512 cyclic groups 11419, 122
Cayley, Arthur 73, 97 axiom 6 fnite groups 947, 122
chemistry 104 defnitions 6 generators 115
classifcation o groups 1245, 153 equality o unctions 508 groups o integers modulo n
cosets and Lagranges defnition 50, 51, 52 98100, 121
theorem 1359, 154 Theorem 1 557 infnite groups 904
group structures 126 equivalence classes 3242, 45 investigation 92
homomorphisms 13944, 1545 defnition 33 Latin Square 96, 97, 121
isomorphisms 14452, 155 Theorem 6 35 order 94, 112, 121, 122
permutation groups 12630 equivalence relations 2527, 45 right and let cancellation
permutations and cycle defnition 25 laws 81, 85, 1058, 122
orm 1302, 154 Euler diagrams 14 subgroups 10813
properties o cycle orm 1324 Euler, Leonhard 14, 47 symmetry groups 1005, 121
co-domains 48, 84 Theorem 1 1067
Cohen, Paul 3 F Theorem 4 11213
commutativity 76, 80, 85 fnite groups 947, 122 Theorem 5 113
complement sets 7, 15, 44 defnition 94
composition o unctions 5961, 85 fnite order 112, 122 H
Theorem 2 59 fnite sets 9 Hilbert, David 3
Continuum Hypothesis 3 unction 46, 834 Hilbert Hotel 4
contrapositive statements 50 bijective unctions 52 homomorphisms 13942, 1545
cosets 1359, 154 binary operations 725, 85 defnitions 139, 143
defnition 135 cancellation laws 813 injective homomorphisms
Theorem 2: properties o composition o unctions 5961, 142
cosets 136, 154 85 kernel o a homomorphism
cycle orm 1302 equality o unctions 508 1424, 155
length o a cycle 132 evolution o the unction surjective homomorphisms
properties o cycle orm 1324 concept 47 142

Index 1 65
homomorphisms (continued) N set theory 2, 424, 47
Theorem 4: properties o Noether, Emily 89, 126 Cartesian product o two
homomorphisms 141, 154 sets 2122
Theorem 5 142, 155 O equivalence classes and
Theorem 6 143, 155 partitions 3242, 45
onto unctions 51
trivial homomorphisms 141 language o sets 34
order 94, 112, 121, 122
partitions 1214
I
P power set P(S) 9
identity element e 789, 85 relations 2332
denition 78 partitions 1214, 3242, 45
Russells Paradox 11
Theorem 8 78 denition 12
set denitions and operations
identity unctions 56, 702, 85 permutation groups 12630
45
denition 70 denitions 127, 129
set properties 1621
Theorem 6 70 identity permutation 129
Venn diagrams 1416
Theorem 7 70 permutations and cycle
well-dened sets, equal sets and
improper subgroups 110, 121 orm 1304, 154
set diference 512
innite groups 904, 122 symmetric groups 129
subgroups 10813, 121
innite order 112 Theorem 1 129, 154
cyclic subgroups 117, 123
injective unctions 50, 84 power set P(S) 9
denition 108, 110
injective homomorphisms 142 proper subgroups 110, 121
improper subgroups 110, 121
integers modulo n 98100, 121 properties o binary
proper subgroups 110, 121
intersection o two sets 7, 44 operations 767
Theorem 2 109, 122
inverse unctions 615 denitions 76
Theorem 3 111
denition 61 properties o unctions 669
trivial subgroups 110, 121
Theorem 3 612 investigation 67
surjective unctions 51, 84
inverse o an element 7981, 85 properties o composite
surjective homomorphisms
denition 79 unctions 68
142
Theorem 9 80 Theorem 4 66
symmetric diference 7, 15, 44
inverse relations 23 Theorem 5 67
symmetric groups 129, 154
Islamic art 87 R symmetric relations 25
isometries 103 symmetry groups 1005, 121
range 48, 84
isomorphisms 14452, 155
reexive relations 25 T
denition 147
relations 2324
Theorem 7 151, 155 transitive relations 25
denition 23
trivial homomorphisms 141
K equivalence relations 2527,
trivial subgroups 110, 121
Klein bottle 147 45
Klein-4 group 147 unctions as relations 4850 U
Klein, Felix 147 inverse relations 23 union o two sets 7, 44
modular congruence 2731, 45 universal set 7
L reexive relations 25 universal theory o everything
Lagrange, Louis 126 symmetric relations 25 8687, 11920
Lagranges Corollary 118 transitive relations 25 cyclic groups 11419
Lagranges theorem 135, 138, 154 right cancellation law 81, 85, Group Theory 8889
theorem 3 138 1058, 122 groups 89105
Latin Square 96, 97, 121 right cosets 135, 154 properties and theorems o
let cancellation law 81, 85, 1058, Russell, Bertrand 11 groups and subgroups
122 Russells Paradox 11 105114
let cosets 135, 154
S V
Leibniz, Gottried Wilhelm 47
set diference 512, 15, 44 Venn diagrams 1416
M denitions 7 Venn, John 14
modular arithmetic 98, 121 set properties 1621
modular congruence 2731, 45 Theorem 1 17 W
denition 29 Theorem 2 17 well-dened sets 512
Theorem 5 2930 Theorem 3 1718 denitions 5
monomorphisms 142 Theorem 4 1819 Weyl, Hermann 88

166 Index
MATHEMATICS HIGHER LEVEL :
SETS, RELATIONS AND GROUPS
The most comprehensive and accurate coverage of the Sets, Relations and Authors
Groups Option for HL, with unrivalled support straight from the IB. Offering a Josip Harcet
rigorous approach and supported by a full set of worked solutions online, Lorraine Heinrichs
this book will fully challenge learners to drive top achievement.
Palmira M ariz Seiler
O xford course books are the only DP resources developed with the IB.
This means that they are: M arlene Torres-Skoumal
 The most accurate match to IB specifications
 Written by expert and experienced IB examiners and teachers
 Packed with accurate assessment support, directly from the IB
 Truly aligned with the IB philosophy
Free support material online at
www.oxfordsecondary.co.uk/ibmathhl

Extensive challenge material thoroughly


stretches learners, supporting the highest levels
of comprehension

Examples and investigations help to put complex


theory into practice

Also available
978 0 19 839012 1 978 0 19 839011 4

How to get in contact:

1 web www.oxfordsecondary.co.uk/ib
email schools.enquiries.uk@oup.com
tel +44 (0)1 536 452620
fax +44 (0)1 865 31 3472