Set Theory

Alexandre Borovik
0N1 (MATH19861)
Mathematics for Foundation Year
http://www.maths.manchester.ac.uk/
~
avb/
math19861.html
Lecture Notes
0N1 Mathematics Arrangements for the Course 2
Arrangements for the Course
Aims of 0N1
A basic course in pure mathematical topics for members
of the foundation year.
Key ingredient: language of Mathematics, including spe-
cic use of English in Mathematics..
Brief description
13 lectures: Sets. Denition, subsets, simple examples, union,
intersection and complement. De Morgans Laws. Ele-
mentary Logic; universal and existential qualiers. Proof
by contradiction.
5 lectures: Polynomials. Equations, relations between roots
and coecients. Remainder Theorem. Repeated Fac-
tors.
2 lectures: Proof by induction.
2 review lectures at the end of the course, Week 12.
Textbooks:
AC Croft and R Davison, Foundation Maths, Prentice
Hall
Bostock, Chandler and Rourke, Further Pure Mathe-
matics, Stanley Thornes
S Lipschitz, Set Theory and Related Topics, McGraw-
Hill
Richard Hammack, Book of Proof, http://www.people.
vcu.edu/
~
rhammack/BookOfProof/index.html
Detailed lecture notes will be provided as course progresses.
Course Webpage:
http://http://www.maths.manchester.ac.uk/
~
avb/math19861.
html
Arrangements
* As of 201314 academic year

Two lectures in weeks 15 and 712:
Monday 16:00, Renold/C016
Friday 13:00, Renold/C016
Learn to take lecture notes!
One tutorial (in small groups) in weeks 2-5 and 7-12.
Friday 11:00
Exercise sheets are distributed a week before tuto-
rial.
Work on your own, in the tutorial discuss your
solutions with an Instructor.
Solutions to exercises are distributed after the tu-
torial.
Hours of private study : 68.
5-minute multiple choice tests: 10 minutes at the end
of each of 10 tutorials are reserved for the test (but experi-
ence shows that clearing the desks from books, etc, and then
collection of scripts takes some time, so the actual test time
is about 5 minutes).
Two questions, each costing 2%, so that all together
they make
10 2 2% = 40% of the total mark for course.
One problem is on material from the previous week,
another from all all previous weeks at random.
Please notice that one of the tests will be on the last
day of classes, Friday 13 December.
* The date is for 201314 academic year

You have been warned: if you decide to go home
early, you loose, with every missed test, 4% of
your mark for the course.
Test Resits: No Resits or Reworks
All medical notes, honourable excuses, etc.: submit to
Foundation Year Oce. Their decision is nal.
The Lecturer will not look into any detail.
If Foundation Studies Oce decides that you have a
valid reason to miss a test, the total for tests will be
adjusted in proportion to your marks for those tests
that you sat.
Rules for Tutorial Tests
1. Students will be admitted to the test only after showing
an ocial University ID card. No ID No Test.
2. During the test, the ID card has to be positioned at the
corner of the students desk and ready for inspection.
3. After the test has started, the examiner checks the IDs.
4. Students are not allowed to leave the room until the end
of the test. Examiners will not collect their scripts until
the test is over.
5. No books or any papers other than test paper are al-
lowed to be kept on the table.
6. Examiners should remove any remaining formulae from
the blackboard/whiteboard.
7. If a student breaches any of this rules, or behaves noisy,
etc., Examiners are instructed:
7.1 conscate the oenders test script;
7.2 write across the script: report to the Lecturer;
7.3 make note of the oenders name and ask him/her
to leave the room, quietly;
7.4 after the test, immediately report the incident to
the Lecturer.
The Lecturer will take care of further necessary actions.
2 hour examination (January):
Weighting within course 60%.
Ten problems, you choose and solve six of them. Each
problem costs 10%; notice 6 10% = 60%.
Unlike tests, examination problems are not multiple
choice.
You will have to give not only an answer, but justify it
by a detailed calculation and/or a complete proof.
Past exam papers are available at
http://www.maths.manchester.ac.uk/
~
avb/math19861.
html
E-mail policy:
Questions are welcome, e-mail them to
[email protected]
When relevant, the Lecturer will send a response (with-
out naming the author of the original question) to all
the class.
Ensure that the subject line of your message is meaning-
ful. Always include the name of the course e.g. 0N1
Tutorial. Otherwise your message will be deleted as
spam.
Use your university e-mail account. The lecturer will
delete, without reading, e-mails from outside of the Uni-
versity.
Questions from students
These lecture notes include some of the questions
my students asked by students in the course. These
parts of notes contains no compulsory material
but still may be useful.
Notation and Terminology:
Some textbooks use notation and terminology
slightly dierent from that used in the lectures.
These notes make use of marginal
comments
* marginal = written on the margin,

the empty space at the side of a
pagelike this one.
margin = edge, border
* comment = remark, commentary,
note
like this one to give other words which are fre-
quently used in English mathematical literature
in the same sense as the marked word. Outside
of mathematics, the usage of such words could be
dierent. Also, the choice of words very much de-
pends on the sentence in which they are used.
0N1 Mathematics Lecture 1 7
Lecture Notes
Lecture 1
Sets
A set is any collection of objects, for example, set of numbers.
The objects of a set are called the elements of the set.
A set may be specied by listing its elements. For exam-
ple, {1, 3, 6} denotes the set with elements 1, 3 and 6. This
is called the list form for the set. Note the curly brackets.
* Typographical terms:
{ opening curly bracket
} closing curly bracket
We usually use capital letters A, B, C, etc., to denote sets.
The notation x A means x is an element of A.
But
* Alternatively we may say x belongs
to A or A contains x.
x A means x is not an element of A.
Example. 1 {1, 3, 6}, 3 {1, 3, 6}, 6 {1, 3, 6} but 2
{1, 3, 6}.
A set can also be specied in predicate form
, that is * or descriptive form

by giving a distinguished property of the elements of the set
(or an explicit
description of the elements in the set). For * explicit = specic, denite

example, we can dene set B by
B = {x : x is a possitive integer less than 5}.
The way to read this notation is B is the set of all x such
that x is a positive integer less than 5. The curly brackets
indicate a set and the colon
: is used to denote such * Typographical terms:

: colon
that, and, not surprisingly, is read such that.
Two sets are equal
if they have exactly the same ele- * We also say: two sets coincide.
ments. Thus
{1, 2, 3, 4} = {x : x is a possitive integer less than 5}.
In list form the same set is denoted whatever order the el-
ements are listed and however many times each element is
listed. Thus
{2, 3, 5} = {5, 2, 3} = {5, 2, 3, 2, 2, 3}.
Note that {5, 2, 3, 2, 2, 3} is a set with only 3 elements: 2, 3
and 5.
Example.
{x : x is a letter in the word GOOD } = {D, G, O}.
The set {2} is regarded as being dierent from the number
2. A set of numbers is not a number. {2} is a set with only one
element which happens to be the number 2. But a set is not
the same as the object it contains: {2} = 2. The statement
2 {2} is correct. The statement {2} {2} is wrong.
The set
{x : x is an integer such that x
2
= 1}
has no elements. This is called an empty set
. It was said * Some books call it null set.

earlier that two sets are equal if they have the same elements.
Thus if A and B are empty sets we have A = B. (Mathe-
maticians have found by experience that this is the correct
viewpoint.) Thus there is only one empty set, THE empty
set.
It is usually denoted by . Thus * Notice the use of denite article THE.

{x : x is an integer such that x
2
= 1} = .
Consider the sets A and B where A = {2, 4} and B =
{1, 2, 3, 4, 5}. Every element of the set A is an element of the
set B. We say that A is a subset of B and write A B, or
B A. We can also say that B contains A.
* Also: A is contained in B,
A is included in B.
* This section contains no compulsory

material but still may be useful.
1. My question is: Are all empty sets equal? No matter the condi-
tions. For example is
{x : x is positive integer less than zero}
equal to
{x : x is an integer between 9 and 10}
Answer. Yes, all empty sets are equal. To see that in your
example, let us denote
A = {x : x is positive integer less than zero}
and
B = {x : x is an integer between 9 and 10}
So, I claim that A = B. If you do not agree with me, you have to
show that A is dierent from B. To do so, you have to show me
an element in one set that does not belong to another set. Can
you do that? Can you point to an oending element if both sets
have no elements whatsoever?
Indeed, can you point to a positive integer less than zero which
is not an integer between 9 and 10? Of course, you cannot,
because there are no positive integers less than zero.
Can you point to an integer between 9 and 10 which is not a
positive integer less than zero? Of course, you cannot, because
there are no integers between 9 and 10.
Hence you cannot prove that A is not equal to B. Therefore you
have to agree with me that A = B.
Lecture 2
Subsets
_
'
A B
Figure 1: Diagram of A B or B A
This is a simple example of Venn diagram for showing
relationships between sets.
Some basic facts:
A A for every set A. Every set is a subset of itself.
The empty set is a subset of every set: A for any
set A.
If A B and B C then A C.
* We say that is a transitive relation

between sets. Notice that the relation
being an element of is not transitive.
relation = connection, bond
If A B and B A then A = B.
Example. The subsets of {1, 2, 3} are
, {1}, {2}, {3}, {1, 2}, {1, 3}, {2, 3}, {1, 2, 3}.
Note: dont forget the empty set and the whole set {1, 2, 3}.
Thus {1, 2, 3} has 8 subsets.
Theorem
If A is a set with n elements then A has 2

n
sub- * The word theorem means a statement
that has been proved and therefore be-
came part of mathematics. We shall also
use words proposition, lemma, corollary:
they are like theorem, but a proposition
is usually a theorem of less importance,
while lemma has no value on its known
and is used as a step in a proof of a theo-
rem. Corollary is something that easily
follows from a theorem.
sets. Here,
2
n
= 2 2 2
with n factors.
Proof.
Let A = {a
1
, a
2
, . . . , a
n
}. How many are there ways
* The word proof indicates that an ar-
gument establishing a theorem or other
statement will follow.
to choose a subset in A? When choosing a subset, we have
to decide, for each element, whether we include this elements
into our subset or not. We have two choices for the rst
element: include and do not include, two choices for the
second element, etc., and nally two choices for the n
th
ele-
ment:
2 2 2
choices overall.
If A B and A = B we call A a proper subset of B and
write A B to denote this.
* If A B, we also write B A.
Example. Let A = {1, 3}, B = {3, 1}, C = {1, 3, 4}. Then
A = B true
A B false
C A false
A B true
A C true
C C false
B A true
A C true
(Compare with inequalities for numbers: 2 2 true, 1 2
true, 2 < 2 false, 1 < 2 true.)
A set with n elements contains 2
n
1 proper subsets.
Finite and Innite Sets
A nite set is a set containing only nite number of elements.
For example, {1, 2, 3} is nite. If A is a nite set, we denote by
|A| the number of elements in A. For example, |{1, 2, 3}| = 3
and || = 0.
A set with innitely many elements is called an innite set.
The set of all positive integers (also called natural numbers)
N = {1, 2, 3, . . . , }
is innite; the dots indicate that the sequence 1, 2, 3 is to be
continued indenitely.
* indenitely = for ever, without end

The set of all non-negative integers
is also innite: * There is no universal agreement about

whether to include zero in the set of
natural numbers: some dene the nat-
ural numbers to be the positive inte-
gers {1, 2, 3, . . .}, while for others the
term designates the non-negative in-
tegers {0, 1, 2, 3, . . .}. In this lecture
course, we shall stick to the rst one
(and more traditional) convention: 0 is
not a natural number.
N
0
= {0, 1, 2, 3, . . . , }.
More examples of innite sets:
Z = { . . . , 2, 1, 0, 1, 2, . . . } (the set of integers)
{ . . . , 4, 2, 0, 2, 4, . . . } (the set of all even integers)
{ . . . , 3, 1, 1, 3, . . . } (the set of all odd integers)
Q denotes the set of all rational numbers (that is, the num-
bers of the form n/m where n and m are integers and
m = 0),
R the set of all real numbers (in particular,
2 R and
R),
C the set of all complex numbers (that is, numbers of the
form x + yi, where x and y are real and i is a square
root of 1, i
2
= 1).
* The letters
ABCDEFGHIJKLMOPRSTUVWXYZ
are called blackboard bold and where
invented by mathematicians for writing
on a blackboard instead of bold letters
ABC. . . which are dicult to write
with chalk.
They are all innite sets. We have the following inclusions:
N N
0
Z Q R C.

1. > (c) Let U = {u, v,w, x, y, z}.
> (i) Find the number of subsets of U.
> (ii) Find the number of proper non-empty subsets of U.
>
> i think the answer of question (ii) should be 63,
> not 62 which is given by
> exam sample solution. how do u think about it
Answer. The answer is 62: there are 2
6
= 64 subsets in U
altogether. We exclude two: U itself (because is is not proper)
and the empty set (because it is not non-empty.
2. > My question
> relates to one of the mock exam questions,
> worded slightly differently.
>
> Question: List the 8 subsets of {a,b,c,d} containing {d}?
Answer. A very good questionhow to list in a systematic way
all subsets of a given set? I emphasise the word systematic, this
means that if you do the same problem a week later, you get
exactly the same order of subsets in the list.
There are several possible approaches, one of them is to use the
principle of ordering words in a dictionary; I will illustrate it on
the problem
list all subsets in the set {a, b, c}.
In my answer to that problem, you will perhaps immediately
recognise the alphabetic order:
{}
; {a}, {a, b}, {a, c}, {a, b, c}; {b}, {b, c}; {c}. * {} is the empty set
Returning to the original question,
List the 8 subsets of {a, b, c, d} containing {d},
we have to add the element d to each of the sets:
{d}; {a, d}, {a, b, d}, {a, c, d}, {a, b, c, d}; {b, d}, {b, c, d};
{c, d}.
Lecture 3
Operations on Sets
B
A
A B A B
Figure 2: Sets A and B and their intersection A B and union
A B.
Suppose A and B are sets. Then AB denotes the set of
all elements which belong to both A and B:
A B = { x : x A and x B}.
A B is called the intersection of A and B.
* The typographic symbol is some-

times called cap. Notice that the
name of a typographical symbol for an
operation is not necessary the same as
the name of operation. For example,
symbol plus is used to denote addition
of numbers, like 2 + 3.
Example. Let A = {1, 3, 5, 6, 7} and B = {3, 4, 5, 8}, then
A B = {3, 5}.
A B denotes the set of all elements which belong to A
or to B:
A B = { x : x A or x B}.
A B is called the union of A and B.
* The typographic symbol is some-

times called cup.
Notice that, in mathematics, or is usually understood in
the inclusive sense: elements from A B belong to A or to
B or to both A and B; or, in brief, to A and/or B. In some
human languages, the connective
or is understood in the * Connective is a word like or, and,

but, if, . . .
exclusive sense: to A or to B, but not both A and B. We
will always understand or as inclusive and/or. In
particular, this means that
A B A B.
Example. Let A = {1, 3, 5, 6, 7}, B = {3, 4, 5, 8}, then
A B = {1, 3, 4, 5, 6, 7, 8}.
If A and B are sets such that AB = , that is, A and B
have no elements in common, we say that A is disjoint from
B, or that A and B are disjoint
(from each other). * Or that A and B do not intersect.

Example. A = {1, 3, 5}, B = {2, 4, 6}. Here A and B are
disjoint.
In any application of set theory all the sets under consider-
ation will be subsets of a background set, called the universal
set. For example, when working with real numbers the uni-
versal set is the set R of real numbers. We usually denote the
universal set by U.
U is conveniently shown as a frame when drawing a
Venn diagram.
_
'

A
_
B
U
Figure 3: The universal set U as a background set for sets A
and B.
All the sets under consideration are subsets of U and so
can be drawn inside the frame.
Let A be a set and U be the universal set. Then A
(called
the complement
of A and pronounced A prime) denotes * Notice that the complement A
is
sometimes denoted A and pronounced
not A, or A (pronounced A bar), or
A
c
(A compliment)
the set of all elements in U which do not belong to A:
A
= { x : x U and x A}.
U
A
A
Figure 4: The shaded area is the complement A
of the set A.
Example. Let U = {a, b, c, d, e, f}, A = {a, c}, B = {b, c, f},
C = {b, d, e, f}. Then
B C = {b, c, d, e, f},
A (B C) = {c},
A
= {b, d, e, f}
= C,
A
(B C) = C (B C)
= {b, d, e, f}
= C.
It will be convenient for us to modify predicate notation:
instead of writing
{ x : x U and x satises . . . }
we shall write
{ x U : x satises . . . }
Example.
{ x Z : x
2
= 4 } = { 2, 2 }.
Boolean Algebra
When dealing with sets, we have operations , and

. The
manipulation of expressions involving these symbols is called
Boolean algebra (after George Boole, 18151864). The iden-
tities of Boolean algebra
are as follows. (A, B and C denote * Or laws of Boolean algebra.

arbitrary sets all of which are subsets of U.)
A B = B A
A B = B A
_
commutative laws (1)
A A = A
A A = A
_
idempotent laws (2)
A (B C) = (A B) C
A (B C) = (A B) C
_
associative laws (3)
A (B C) = (A B) (A C)
A (B C) = (A B) (A C)
_
distributive laws
(4)
A (A B) = A
A (A B) = A
_
absorbtion laws (5)
identity laws:
A U = A A U = U
A = A A =
(6)
complement laws:
(A
= A A A
= U
=
A A
= U
= U
(7)
(A B)
= A
(A B)
= A
_
De Morgans laws (8)
We shall prove these laws in the next lecture. Meanwhile,
notice similarities and dierences with laws of usual arith-
metic. For example, multiplication is distributive with respect
to addition:
a (b + c) = (a b) + (a c),
but addition is not distributive with respect to multiplication:
it is not true that
a + (b c) = (a + b) (a + c).
Notice also that the idempotent laws are not so alien to
arithmetic as one may think: they hold for zero,
0 + 0 = 0, 0 0 = 0.
Sample Test Questions
* Marking scheme: 2 marks for a cor-

rect answer, 0 for an incorrect answer or
no answer.
1. Let X = {x R : x
4
1 = 0}. Which of the following sets is equal
to X?
(A) {1} {1} (B) {1} (C) {1} {1}
Answer: (C), because the set X equals {1, 1}.
2. Let X = {x R : x
3
= x
2
}. Which of the following sets is equal
to X?
(A) (B) {0, 1} (C) {0, 1, 1}
3. How many subsets of {a, b, c, d} contain {d}?
(A) 6 (B) 8 (C) 15
Answer: (B), because each of the subsets of {a, b, c, d} that con-
tain {d} can be obtained from a (unique!) subset of {a, b, c} by adding
element d. But the set of three elements {a, b, c} contains 2
3
= 8 subsets.
4. How many subsets of {a, b, c, d, e} contain {b, e}?
(A) 8 (B) 15 (C) 6
Questions from Students

1. This question appear to refer to the following problem from a
test:
Which of the following sets is nite?
(A) {1, 2} R (B) {x R : x
2
< 9} (C) [0, 1] [
1
2
,
3
2
]
A student wrote:
> what is the definition of finite and infinite sets?
> because the question that
> you gave us today confused me:
> I think all answers could be correct,
> for example, answer b is -3<x<3 and
> I think it is correct.
Answer. A nite set is a set containing only nite number of
elements. For example, 1,2,3 is nite. A set with innitely many
elements is called an innite set.
The set that you mentioned,
{x R : 3 < x < 3}
is innite: there are innitely many real numbers between 3 and
3.
For example, take a real number which has decimal expansion
1.2345 . . .
No matter how you continue write more digits after the decimal
point (and this can be done in innitely many ways), you will have
a number which is bigger than 3 and smaller that 3. Therefore
the set
{x R : 3 < x < 3}
is not nite.
However, the set
{x Z : 3 < x < 3}
is nite, it equals {2, 1, 0, 1, 2} and therefore has 5 elements.
2. > sorry to disturb you I have got one more question
> Given that A and B are intersecting sets,
> show following on venn
> diagram:A, AUB, AUB, and AnB
>can you please do these in the lecture
Answer is the following sequence of Venn diagrams:
Sets A, B, U.
Sets A
, A B
, A
.
Set A
.
3. > Dear Sir,
> Can you please help me with the following question?
[6 marks] Let
A = {x R : x
4
+ x > 2 }
B = {x R : x
3
< 1}
and
C = {x R : x
8
> 1}.
(i) Prove that A B C.
> Can you say that A and B are disjoint as they do not meet?
> And therefore the Empty Set is a subset of C
Answer: It would be a valid argument if A and B were indeed
disjoint. But they are not; one can easily see that 2 belongs to
both A and B.
A correct solution: Assume x A B. Then x A and x B.
Since x A, it satises
x
4
+ x > 2.
Since x B, it satises
x
3
< 1
which implies x < 1 which is the same as 1 > x. Adding the last
inequality to the inequality x
4
+ x > 2, one gets
x
4
+ x + 1 > 2 + x
which simplies as
x
4
> 1.
Both parts of this inequality are positive, therefore we can square
it and get
x
8
> 1.
But this means that x C. Hence A B C.
4.
> Say for eg you have a situation whereby you have
>
> A U AU B
>
> Does this simply to A U U (which is U) or A U B? Because i no A U A is
> Union but i get confused when simplifying these when you have AU B. is
> it Union or is it B?
Answer: You are mixing the union symbol and letter U used
to denote the universal set. The correct calculation is
AA
B = (AA
)B = U B = U,
I set it in very large type to emphasise the dierence between
symbol and letter U. The answer is U, the universal set.
5.
> Was just wandering about a note I took in your lecture that doesnt
> seem right. I might have copied it down wrong but I wrote:
>
> A = Any integer B = Any Real Number
>
> A union B = any integer
>
> Was just wandering wether that should be,
> A union B = any real number
Answer: Of course, you are right: if A = Z and B = R then
A B = B and A B = A.
I believe I gave in my lecture both equalities and also a general
statement:
If A B, then A B = B and A B = A.
0N1 Mathematics Lecture 4 Set Theory 23
Lecture 4
Set theory
The identities in (1)-(7) of the previous lecture are called the
laws of Boolean algebra. Several of them are obvious
because * obvious = evident, self-evident

of the denitions of , and

. The others may be veried
* to verify
= to check, to conrm, to validate
by drawing Venn diagrams. For example, to verify that
A (B C) = (A B) (A C),
we draw the following diagrams.
A
B
C
(a) A, B, C
A
B
C
(b) B C
A
B
C
(c) A (B C)
A
B
C
(d) A B
A
B
C
(e) A C
A
B
C
(f) (A B) (A C)
Equality holds
because diagrams (c) and (f) are the same. * holds = is true
Because of the associative laws in (1) of the previous lec-
ture, we can write ABC and ABC with unambiguous
* unambiguous = unmistakable, de-

nite, clear
ambiguous = vague, unclear, uncertain
meanings. But we must not write A B C or A B C
without brackets. This is because, in general
* A good example when the use of a

word in mathematics is dierent from its
use in ordinary speech. In the usual lan-
guage in general means as a rule,
in most cases. In mathematics in
general means sometimes. For exam-
ple, in mathematics the phrases Some
people are more than 100 years old and
In general, people are more than 100
years old are the same.
A (B C) = (A B) C,
A (B C) = (A B) C.
(Give your examples!)
To prove the property A B for particular
sets A
* particular = individual, specic
and B we have to prove that every element of A is an element
of B (see denition of ). Sometimes this is clear.
But if
* clear = obvious, self-evident
not proceed as in the next examples.
Example. Let
A = {x R : x
2
3x + 2 = 0}.
Prove that A Z.
Solution. Let x A. Then
x
2
3x + 2 = 0,
(x 1)(x 2) = 0,
x =
_
1 or
2
x Z.
To prove A = B for particular sets A and B we have to
prove A B and then B A.
Recall that a segment [a, b] of the real line R is dened as
the set
* Typographical symbols:
[ opening square bracket
] closing square bracket
[a, b] = { x R : a x b }.
Example. Let A = [1, 2] and
B = [0, 2] [1, 3].
Prove that
A = B. * prove that . . .
= show that . . . , demonstrate that . . .
Solution. We rst prove that
[1, 2] [0, 2] [1, 3].
Let x [1, 2]. Then 1 x 2. Hence 0 x 2 and
1 x 3. Hence x [0, 2] and x [1, 3]. Hence
x [0, 2] [1, 3],
and, since x is an arbitrary
element of [1, 2], this means that * arbitrary = taken at random

[1, 2] [0, 2] [1, 3].
Now we prove that
[0, 2] [1, 3] [1, 2].
Let x [0, 2] [1, 3]. Then x [0, 2] and x [1, 3]. Hence
* hence
= therefore, for this reason, thus, conse-
quently, so
0 x 2 and 1 x 3. Therefore x 1 and x 2. For
this reason 1 x 2. Consequently, x [1, 2].
Comment: In the lecture, an alternative
method was * alternative = other, another, dierent

used for solving a similar problem. It is based on a graphic
representation of segments [a, b] on the real line R.
- , , , ,
0 2
1 3
R
One can immediately see
from this picture that * see = observe, notice

[0, 2] [1, 3] = [1, 2].
Similarly, an interval (a, b) of the real line R is dened as
the set
(a, b) = { x R : a < x < b }.
Example. Notice that
[0, 1] [1, 2] = {1}
while
(0, 1) (1, 2) = .
Do not mix notation {a, b}, [a, b], (a, b)!
Some problems solved with the help of Venn
diagrams
Venn diagrams can be used to solve problems of the following
type.
Example. 100 people are asked about three brands of soft
drinks called A, B and C.
(i) 18 like A only (not B and not C).
(ii) 23 like A but not B (and like C or dont like C).
(iii) 26 like A (and like or dont like other drinks).
(iv) 8 like B and C (and like A or dont like A).
(v) 48 like C (and like or dont like other drinks).
(vi) 8 like A and C (and like or dont like B).
(vii) 54 like one and only one of the drinks.
Find how many people like B and nd how many people dont
like any of the drinks.
For solution, we draw a Venn diagram. Let
a be number of people liking A only
b be number of people liking B only
c be number of people liking C only
d be number of people liking A and B but not C
e be the number of people who like A and C, but not B.
f be the number of people who like B and C, but not A.
g be the number of people who like all tree products A,
B, and C.
h be number of people liking none of the drinks,
as shown on the Venn diagram below.
From (i)(vii) we get
(i) a = 18
(ii) a + e = 23
(iii) a + d + e + g = 26
(iv) f + g = 8
(v) c + e + f + g = 48
(vi) e + g = 8
(vii) a + b + c = 54
We also have
(viii) a + b + c + d + e + f + g + h = 100
Now (i) gives
a = 18, (ii) gives e = 5, (vi) gives g = 3, * gives = yields

(iii) gives d = 0, (iv) gives f = 5, (v) gives c = 35, (vii) gives
b = 1, (viii) gives h = 33.
Therefore the number of people who like B is
b + d + f + g = 9,
and the number of people who like none is h = 33.
Sample test question
1. X and Y are sets with the following three properties.
(i) X
has 12 elements.
(ii) Y
has 7 elements.
a
b
c
e
d
f g
h
A
B
C
(iii) X Y
has 4 elements.
How many elements in X
Y ?
(A) 6 (B) 8 (C) 9
Answer. (C).
Brief solution. Denote x = |X Y
|, y = |X
Y | (this is what
we have to nd), z = |X Y |, t = |(X Y )
| (make a Venn diagram!),

then
|X
| = y + t = 12
|Y
| = x + t = 7
|X Y
| = x = 4
Excluding unknowns, we nd t = 3 and y = 9.
Detailed solution. Recall that we use notation |A| for the num-
ber of elements in a nite set A.
Denote x = |X Y
|, y = |X
Y | (this is what we have to nd),

z = |X Y |, t = |(X Y )
|, see a Venn diagram below.

_
'

x
X
z
_
y
t
Y
U
Then
|X
| = y + t = 12
|Y
| = x + t = 7
|X Y
| = x = 4
So we have a system of three equations:
y + t = 12
x + t = 7
x = 4
Excluding unknowns, we nd t = 3 and y = 9.
This last step can be written in more detail. Substituting the value
x = 3 from the third equation into the second equations, we get
4 + t = 7,
which solves as t = 3. Now we substitute this value of t in the rst
equation and get
y + 3 = 12;
solving it, we have y = 9.
2. Which of the following sets is innite?
(A) {0, 1} R (B) {x R : x
2
< 4} (C) [0, 1]
_
4
3
,
3
2
Answer. (B). Indeed, this set is {2 < x < 2} is innite. The set
A is nite because it is a subset of a nite set {0, 1}. The set C is empty
and therefore nite.
3. Which of the following sets is nite?
(A) {0, 1}R (B) [0, 1][
1
2
,
3
2
] (C) {x R : x
2
< 9}
4. Let X, Y and Z be sets such that Y X. Which of the following
must be true?
(A) X Z Y Z
(B) Y
(C) X (Y Z) = Y (X Z)

1. > [7 marks] A survey was made of 25 people to ask about
> their use of products A and B. The following infor-
> mation was recorded: 14 people used only one of the
> products; 9 people did not use B ; 11 people did not
> use A.
> (i) How many people used A?
> (ii) How many people used both products?
Answer. A solution is straightforward: denote
a number of people using A but not B
b number of people using B but not A
c number of people using both A and B
d number of people not using any product
(it is useful to draw a Venn diagram and see that a, b, c, d corre-
spond to its 4 regions).
Then
14 people used only one of the products means a+b = 14
9 people did not use B means a+d = 9
11 people did not use A means b + d = 11
Finally, a + b + c + d = 25.
Thus you have a system of 4 linear equations with 4 variables:
a + b = 14
a + d = 9
b + d = 11
a + b + c + d = 25
and it is easy to solve; I leave it you to work out details. Answer:
a = 6, b = 8, c = 8, d = 3.
0N1 Mathematics Lecture 5 Propositional Logic 31
Lecture 5
Propositional Logic
A statement (or proposition) is a sentence which states or
asserts
something. It is either true or false. If true, we say * assert = state, claim

that the statement has truth value T. If false, it has truth
value F .
Example.
London is the capital of England has truth value T.
2 2 = 5 has truth value F .
Are you asleep? is not a statement.
Mathematically we do not distinguish between statements
which make the same assertion, expressed dierently. For
example, The capital of England is London is regarded as
equal to London is the capital of England.
We use p, q, r, . . . to denote statements.
Conjunction. If p and q are statements then p and q is a
new statement called the conjunction of p and q and written
pq. According to mathematical convention,
pq has truth * convention = custom, agreement

value Twhen both p and q have truth value T, but p q has
truth value F in all other cases.
Here is the truth table: * The typographical symbol is called

wedge. It is used not only in logic, but in
some other areas of mathematics as well,
with a completely dierent meaning.
p q p q
T T T
T F F
F T F
F F F
Examples.
Suppose p is 2 is even and q is 5 is odd. Then p q
is 2 is even and 5 is odd. Since p has truth value
Tand q has truth value T, p q has truth value T(1st
row of the table).
3 is odd and 2 is odd has truth value F (see 2nd row
of the truth table).
If we know q is true but pq is false we can deduce that
p is false (the only possibility in the truth table).
p q is sometimes expressed without using and. For
example, Harry is handsome, but George is rich is the same,
mathematically, as Harry is handsome and George is rich.
Disjunction. p or q is called the disjunction of p and q
and written p q. Truth table:
p q p q
T T T
T F T
F T T
F F F
In eect, this table tells us how or is used in mathemat-
ics: it has the meaning of and/or (inclusive meaning of
or).
* The typographical symbol is called

vee
Note that p q is true if at least one of p and q is true. It
is only false when both p and q are false.
* In Computer Science, the exclusive

version of or is also used, it is usually
called XOR (for eXclusive OR) and is
denoted p q. Its truth table is
p q p q
T T F
T F T
F T T
F F F
We shall not use XOR in our course.
Examples.
Suppose p is 4 is odd and q is 5 is odd. Then p q
is 4 is odd or 5 is odd. Since p has truth value F and
q has truth value T, p q has truth value T(3rd row of
truth table).
3 > 4 or 5 > 6 has truth value F (see 4th row of truth
table).
Negation. The statement obtained from p by use of the
word not is called the negation
of p and is written p. * Symbols sometimes used to denote

negation: p, p.
p is sometimes called the opposite of
p
For example, if p is I like coee then p is I dont like
coee. The truth value of p is the opposite of the truth
value of p.
p p
T F
F T
Example. 2 is odd is false, but 2 is not odd is true.
Conditional. Suppose p and q are statements. The state-
ment If p then q, denoted p q, is called a conditional
statement.
The truth values to be given to p q are open * There is a huge number of ways to
express if p then q, for example
p implies q
p leads to q
p yields q
q follows from p
q is a consequence of p
q is a necessary condition for p
p is a sucient condition for q
q is true provided p is true
p entails q
to some debate but the mathematical convention is as follows.
p q p q
T T T
T F F
F T T
F F T
You must work according to this table whether
you like it or not!
The convention is that p q is true when p is false,
regardless of the truth value of q. Rough explanation: when
p is false there is nothing wrong with p q because it means
if p then q and so makes an assertion only when p is true.
Another explanation: some conditional statements can be
thought of as statements of promise. For example:
if I have no cold, Ill come to class.
Here p is I have no cold and q is Ill come to class. If p is
false, that is, if I have cold, you would agree that I have kept
my promise even if I have not come to class (in which case q
is false).
* This example is expanded at the end

of this lecture.
Perhaps the most surprising is the third row of the table.
You may think of it as the principle of the absolute priority of
Truth: Truth is Truth regardless of how we came to it or from
whom we heard it. This is because our statements are about
the world around us and are true if they describe the world
correctly.
For a statement to be true, it is not necessary to * This is why in the literature, our rule
for implication is sometimes called ma-
terial implication: it is about material
world.
receive it from a source of authority or trust.
Statements of promise also give a good explanation. Re-
turning to the phrase if I have no cold, Ill come to class, you
would agree that if I have cold (p is F ) but nevertheless came
to class (q is T), I have kept my promise and told the truth;
hence F T is T.
Examples.
Suppose p is 4 > 1 and q is 3 = 5. Then p q is
If 4 > 1 then 3 = 5. This is false because p is true
and q is false (see 2nd row of truth table).
If 3 = 5 then 2 = 0 is true (see 4th row of truth
table).
If 3 = 5 then 2 = 2 is true (see 3rd row of truth
table).
If p q has truth value F we can deduce that p is true
and q is false (only the second row of the truth table
gives p q false).
Statements of the form p q usually arise only when
there is a variable or unknown involved.
Example. If x > 2 then x
2
> 4 is a true statement,
whatever the value of x. For example, when x = 3, x
2
= 9 > 4
and when x = 4, x
2
= 16 > 4. The statement is regarded as
true, by convention, for values of x which do not satisfy x > 2.
For numbers like x = 1 we do not care whether x
2
> 4 is
true.
The following example illustrates dierent expression of
p q in English. Let p be x > 2 and q be x
2
> 4. Then
all of the following expresses p q.
If x > 2 then x
2
> 4.
x
2
> 4 if x > 2.
x > 2 implies x
2
> 4.
x > 2 only if x
2
> 4.
x > 2 is sucient condition for x
2
> 4.
x
2
> 4 is necessary condition for x > 2.

1. > Im having a bit of trouble with the propositional
> logic conditional statement.
> Surely if p implies q and p is false but q is true,
> the statement that p implies q is false?
> I know you said we would have trouble with this but
> ive found it difficult to trust my own logical
> reasoning when working out subsequently more
> complex compound statements. Could you suggest a
> more logical way of approaching this concept?
Answer. I expand my example with interpretation of implication
as promise.I am using here a large fragment from Peter Subers
paper Paradoxes of Material Implication,
http://www.earlham.edu/peters/courses/log/mat-imp.htm.
It is important to note that material implication does conform to
some of our ordinary intuitions about implication. For example,
take the conditional statement,
If I am healthy, I will come to class.
We can symbolize it, H C. The question is: when is this
statement false? When will I have broken my promise?
There are only four possibilities:
H C H C
T T T
T F F
F T T
F F T
In case #1, I am healthy and I come to class. I have clearly kept
my promise; the conditional is true.
In case #2, I am healthy, but I have decided to stay home and
read magazines. I have broken my promise; the conditional is
false.
In case #3, I am not healthy, but I have come to class anyway.
I am sneezing all over you, and youre not happy about it, but I
did not violate my promise; the conditional is true.
In case #4, I am not healthy, and I did not come to class. I did
not violate my promise; the conditional is true.
But this is exactly the outcome required by the material implica-
tion. The compound is only false when the antecedent
is true * In the conditional statement H C ,

the rst term H is called antecedent, the
second C consequent.
and the consequence is false (case #2); it is true every other time.
Many people complain about case #4, when a false antecedent
and a false consequent make a true compound. Why should this
be the case?
If the promise to come to class didnt persuade you, heres an
example from mathematics.
If n is a perfect square, then n is not prime.
I hope youll agree that this is a true statement for any n. Now
substitute 3 for n:
If 3 is a perfect square, then 3 is not prime.
As a compound, it is still true; yet its antecedent and consequent
are both false.
Even more fun is to substitute 6 for n:
If 6 is a perfect square, then 6 is not prime.
it is a true conditional, but its antecedent is false and consequent
is true.
> Unfortunately, case #4 seemed perfectly logical to me. It was case #3
> which I found illogical. If I told you that I would come to class IF I
> was not sick, and yet I came to class despite being sick, surely my
> promise was not honoured? If I had said I MAY not come to class if I am
> sick then I would always be honouring my promise so long as I came to
> class when I was well... Is this a more appropriate way to think about
> it? Would I have problems using the may component?
Answer. An excellent question. I wish to emphasise:
Propositional Logic is designed for communication with machines,
it gives only very crude description of the way how natural hu-
man language. Such constructions as I MAY are too subtle for
Propositional Logic to capture their meaning.
Therefore we have to live with rules of material implica-
tion as they are: they present a best possible compromise
between language for people and language for machines.
Logical constructions of the kind I MAY are studied in a more
sophisticated branch of logic, Modal Logic. I simply copy the
following description of Modal Logic from Wikipedia:
A modal logic is any system of formal logic that at-
tempts to deal with modalities. Traditionally, there
are three modes or moods or modalities of the
copula to be, namely, possibility, probability, and ne-
cessity. Logics for dealing with a number of related
terms, such as eventually, formerly, can, could, might,
may, must, are by extension also called modal logics,
since it turns out that these can be treated in similar
ways.
But we are not studying Modal Logics in our course. However,
they are taught in Year 4 of School of Mathematics: Course
MATH43032 Non-Standard Logicshttp://www.maths.manchester.ac.uk/undergraduate/ugstudies/units/2008-
09/level4/MATH43032/ by Professor J. Paris.
Lecture 6
Propositional Logic, Continued
Notice that p q and q p are dierent; q p is called
the converse of p q.
Example. Let p = x > 2 and q = x
2
> 4. Then p q
is If x > 2 then x
2
> 4 TRUE. But q p is If x
2
> 4
then x > 2. This is FALSE (for x = 3, for example).
Biconditional p if and only if q is denoted by p q. The
truth table is as follows.
* Notice that, in mathematical litera-

ture and blackboard writing, the expres-
sion if and only if is sometimes abbre-
viated i. p q p q
T T T
T F F
F T F
F F T
So, if p and q are both true or both false then p q is
true: otherwise it is false.
The biconditional p q can be expressed as
p if and only if q
or
p is a necessary and sucient condition for q.
For example,
x > 2 if and only if x + 1 > 3
is the same as
For x > 2 it is necessary and sucient that x + 1 > 3.
p q may be thought of as a combination of p q and
q p.
The symbols , , , and are called connectives.
Compound
statement may be built up from statements * compound = complex, composite

p, q, r, . . .
by means of connectives. We use brackets for punctuation as
in
(p q) ( r q).
We take the convention that applies only to the part of
the expression which comes immediately after it. Thus rq
means (r) q, which is not the same as (r q).
The truth value of a compound statement involving p, q, r, . . .
can be calculated from the truth values of p, q, r, . . . as follows.
Example. Find the truth table of
* Some typographic terminology: in the

expression
(p (q r))
the rst opening bracket and the last
closing bracket (they are underlined)
match each other. This is another pair
of matching brackets:
(p (q r)).
(p (q r)).
(We take 8 rows because there are 3 variables p, q, r, . . . each
with two possible truth values.)
p q r q r p (q r) (p (q r))
T T T T T F
T T F T T F
T F T T T F
T F F F F T
F T T T T F
F T F T T F
F F T T T F
F F F F T F
Please always write the rows in this order, it will help you
to easier check your work for errors.
(We get each of the last 3 columns by use of the truth
tables for , and .)
This can also be set out as follows.
(p (q r))
F T T T T T
F T T T T F
F T T F T T
T T F F F F
F F T T T T
F F T T T F
F F T F T T
F F T F F F
(The truth values for p, q, r (8 possibilities) are entered
rst:
(p (q r))
T T T
T T F
T F T
T F F
F T T
F T F
F F T
F F F
Then the other columns are completed in order 5, 3, 1.)
Example. Find the truth table of p (q p).
p q q q p p (q p)
T T F T T
T F T T T
F T F T F
F F T F F
or
p ( q p)
T T F T T T
T T T F T T
F F F T T F
F F T F F F
Tautologies. The statements
p p and (p (p q)) q
have the following truth tables.
p p p p
T F T
F T T
p q p q p (p q) (p (p q)) q
T T T T T
T F F F T
F T T F T
F F T F T
Only Toccurs in the last column. In other words, the
truth value of the statement is always T, regardless of the
truth values of its components p, q, r, . . .. A statement with
this property is called a tautology.
Examples.
(i) Let p = It is raining. Then p p is Either it is
raining or it is not raining. This is true regardless of
whether it is raining or not.
(ii) Let p = x > 2 and q = y > 2. Then
(p (p q)) q
is If x > 2, and x > 2 implies y > 2, then y > 2. This
is true because
(p (p q)) q
is a tautology: the meanings of p and q are not impor-
tant.
We can think of tautologies as statements which are true
for entirely logical reasons.
Contradictions. A statement which is always F regardless
of the truth values of its components p, q, r, . . . is called a
contradiction. (Only F occurs in the last column of the truth
table).
Example. p p. It is raining and it is not raining.
Logically equivalent statements. Let X and Y be two
statements built up from the same components p, q, r, . . .. If
the truth value of X is the same as the truth value of Y for
every combination of truth values of p, q, r, . . . then X and
Y are said to be logically equivalent. In other words X and
Y are logically equivalent if the nal columns of their truth
tables are the same.
Sample test questions
1. Given that pq is T and q r is F, which of the following statements
is T?
(A) (p q) r (B) (p q) r
(C) ( p q) r
2. Which of the following statements is a tautology?
(A) (p q) ( p q)
(B) (p q) ( p q)
(C) (q p) ( p q)

1. When drawing truthtables, i found that there are
2 types of u can do, is the correct method putting
in T or F values underneath each of the symbols
or I have seen in our notes that the answer can
still be found without finding out each symbol
and by breaking up the particular question.
for example the question ~q --> ( p ---> q)
can a truth table be written in the exam as this:
p q ~q (p-->q) ~q-->(p-->q)
T T F F T
T F T T T
etc. Will you be given full marks for this method
or must u include values for each symbol?
My answer: either way of composing truth tables is valid, can be
used in the exam and be given full marks.
But please, try to write in a neat and comprehensible way, so
that table looks like a table and is not stretched diagonally all
over page.
2. > can I have a simple English sentence illustrates
> this statement p -> (q -> p)?
My answer: quite a number of English sentences built around an
expression even without or even if belong to this type. For
example, the turkey is good, even without all the trimmings:
p is the turkey is good
q is without all the trimmings.
The statement p (q p) becomes
the turkey is good, and for that reason, even without all the
trimmings, the turkey is still good.
Lecture 7
Logically equivalent statements
Let X and Y be two statements built up from the same com-
ponents p, q, r, . . .. If the truth value of X is the same as
the truth value of Y for every combination of truth values of
p, q, r, . . . then X and Y are said to be logically equivalent.
In other words X and Y are logically equivalent if the nal
columns of their truth tables are the same.
Example.
p q p q (p q) p q p q
T T T F F F F
T F F T F T T
F T F T T F T
F F F T T T T

Columns and are the same, i.e.
for every choice of * i.e. = that is,

truth values for p and q, (p q) and p q have the
same truth values. Thus (p q) and p q are logically
equivalent.
Notation. If X and Y are logically equivalent statements
we write X Y .
Example. (p q) p q.
A particular case of this is shown by taking
p = You are French and
q = You are a woman.
Then
(p q) = You are not a French woman and
p q = Either you are not French or you are not a
woman.
The logical equivalence
(p q) p q
is analogous to the set theory identity
(A B)
= A
.
In fact it is remarkable that if we replace by , by ,

by , U by T(to denote a tautology) and by F (to denote a
contradiction) then all the rules of Boolean algebra turn into
logical equivalences.
p q q p
p q q p
_
p p p
p p p
_
idempotent laws (2)
p (q r) (p q) r
p (q r) (p q) r
_
p (q r) (p q) (p r)
p (q r) (p q) (p r)
_
distributive laws
(4)
A (A B) = A
A (A B) = A
_
absorbtion laws (5)
p T p p T T
p F p p F F
(6)
(p) p p p F T F
p p T F T
(7)
(p q) p q
(p q) p q
_
De Morgans laws (8)
They may all be proved by means of truth tables as we
did for
(p q) p q.
Similarly:
p q p q (9)
(p q) (p q) (q p) (10)
We call (1)(9) the fundamental logical equivalences.
Rules 8 and 9 enable us to rewrite and entirely in terms
of , and . Expressions involving , and can be
manipulated by means of rules (1)(7).
Example. Simplify p (p q).
p (p q)
by (4)
(p p) (p q)
by (1)
(p p) (p q)
by (7)
T (p q)
by (1)
(p q) T
by (6)
p q.
To determine whether or not statements X and Y are logically
equivalent we use truth tables. If the nal columns are the
same then X Y , otherwise X Y .
If we are trying to prove X Y we can either use truth
tables or we can try to obtain Y from X by means of funda-
mental logical equivalences (1)(10).
Example. Prove that q p p q.
We could use truth tables or proceed as follows
q p
by (9)
q p
by (7)
q p
by (1)
p q
by (9)
p q.
1. Which of the following statements is logically equivalent to
p (p q)?
(A) q p (B) p q (C) F

1. > In the exam are we going to receive a formula
> sheet with rules of boolean algebra?
Answer. Yes, you are. AB
0N1 Mathematics Lecture 8 Predicate Logic 48
Lecture 8
Predicate Logic
Many mathematical sentences involve unknowns or vari-
ables.
Examples.
(i) x > 2 (where x stands for an unknown real number).
(ii) A B (where A and B stand for unknown sets).
Such sentences are called predicates. They are not state-
ments because they do not have a denite truth value: the
truth value depends on the unknowns.
Examples.
(i) x > 2 is Tfor x = 3, 3
1
2
, etc., F for x = 2, 1, etc.
(ii) A B is Tfor A = {1, 2}, B = R.
A B is F for A = {1, 2}, B = {2, 3, 4}.
We can write p(x), q(x), . . . for predicates involving an
unknown x, p(x, y), q(x, y), . . . when there are unknowns x
and y, p(A, B), q(A, B), . . . when there are unknowns A and
B, etc.
Examples.
(i) Let p(x) denote the PREDICATE x > 2. Then p(1)
denotes the STATEMENT 1 > 2 (truth value F ) while
p(3) denotes the STATEMENT 3 > 2 (truth value T).
(ii) Let p(x, y) denote x
2
+ y
2
= 1. Then p(0, 1) denotes
0
2
+ 1
2
= 1 (true) while p(1, 1) denotes 1
2
+ 1
2
= 1
(false).
The logical connectives , , , , can be used to
combine predicates to form compound predicates.
Examples.
(i) Let p(x) denote x
2
> 5 and let q(x) denote x is posi-
tive. Then p(x) q(x) denotes the predicate x
2
> 5
and x is positive.
(ii) Let p(x, y) denote x = y
2
. Then p(x, y) denotes x =
y
2
.
(iii) Let p(A, B) denote A B and let q(A) denote A
{1, 2} = . Then q(A) p(A, B) denotes the predicate
If A {1, 2} = then A B.
We can calculate truth values as follows.
Example. Let p(x, y) denote x > y and let q(x) denote
x < 2. Find the truth value of the predicate (p(x, y) q(x))
when x = 3 and y = 1.
Solution. We need to nd the truth value of the of the
statement
(p(3, 1) q(3)).
Now p(3, 1) is Tand q(3) is F . Therefore p(3, 1) q(3) is F .
Therefore
(p(3, 1) q(3))
is T.
let p(x) denote the predicate x > 1 and let q(x) denote the
predicate x {0, 1, 2}. Which of the following statements is
true?
(A) p(1) q(1)
(B) p(1) p(1)
(C) (p(2) q(2))
Solution: Notice that
p(1) is T, q(1) is F , p(1) is F , p(2) is T, q(2) is T
Therefore statements (A), (B), (C) become
(A) T F , (B) T F , (C) (T T),
of which (B) is T.
0N1 Mathematics Lecture 9 Quantifiers 51
Lecture 9
Quantiers
Many statements in mathematics involve the phrase for all
or for every or for each: these all have the same meaning.
Examples.
(i) For every x, x
2
0.
(ii) For all A and B, A B = B A.
If p(x) is a predicate we write (x)p(x) to denote the state-
ment For all x, p(x). Similarly, (x)(y)p(x, y) denotes
For all x and all y, p(x, y).
Examples.
2
0. Then (x)p(x) denotes For
every x, x
2
0 or x, For each x, x
2
0.
(ii) Let p(A, B) denote A B = B A. Then
(A)(B)p(A, B)
denotes For all A and B, A B = B A.
When we write (x)p(x) we have in mind that x belongs
to some universal set U. The truth of the statement (x)p(x)
may depend on U.
Example. Let p(x) denote x
2
0. Then (x)p(x) is true
provided that the universal set is the set of all real numbers,
but (x)p(x) is false if U = C because i
2
= 1.
Usually the universal set is understood from the context.
But if necessary we may specify it:
For every real number x, x
2
0
may be denoted by (x R)p(x) instead of (x)p(x).
If p(x) is a PREDICATE then
(x)p(x) is a STATEMENT.
(x)p(x) is true if p(x) is true for every x U, whereas
* whereas = while
(x)p(x) is false if p(x) is false for at least one x U.
Similar remarks apply to (x)(y)p(x, y), etc.
Examples.
2
0 where U = R. Then (x)p(x) is
true.
(ii) The statement For every integer x, x
2
5 is false.
Here U = Z but there is at least one x Z for which
x
2
5 is false, e.g. x = 1.
(iii) Let p(x, y) denote
If x y then x
2
y
2
,
where U = R. Then (x)(y)p(x, y) is false. Take, for
example, x = 1 and y = 2. Then p(x, y) becomes
If 1 > 2 then 1 > 4.
Here 1 > 2 is Tbut 1 > 4 is F . From the truth table
for we see that If 1 > 2 then 1 > 4 is F . Hence
(x)(y)p(x, y) is F .
(iv) For all x and all y, if x y then 2x 2y is T.
The symbol is called the universal quantier: it has the
meaning for all, for every or for each.
We now also study , the existential quantier: it has
the meaning there is (at least one), there exists or for
some.
Examples.
2
5, where U = R. Then (x)p(x)
denotes
There exists a real number x such that x
2
5.
This can also be expressed as
x
2
5 for some real number x.
(ii) The statement
There exist sets A and B for which (AB)
= A
may be denoted by
(A)(B)p(A, B)
where p(A, B) denotes the predicate (AB)
= A
,
or
(A)(B)((A B)
= A
).
If p(x) is a PREDICATE then (x)p(x) is a STATE-
MENT.
(x)p(x) is true if p(x) is true for at least one x U,
whereas
(x)p(x) is false if p(x) is false for all x U.
Examples.
(i) Let U = R. The statement (x)x
2
5 is Tbecause
x
2
5 is Tfor at least one value of x, e.g. x = 3.
(ii) Let p(x) denote x
2
< 0, where U = R. Then (x)p(x)
is F because p(x) is F for all x U.
(iii) (x)(y)(x + y)
2
= x
2
+ y
2
(where U = R) is T: take
x = 0, y = 0 for example.
Statements may involve both and .
Example. Consider the following statements.
(i) Everyone likes all of Beethovens symphonies.
(ii) Everyone likes at least one of Beethovens symphonies.
(iii) There is one Beethovens symphony which everyone
likes.
(iv) There is someone who likes all of Beethovens sym-
phonies.
(v) Every Beethovens symphony is liked by someone.
(vi) There is someone who likes at least one of Beethovens
symphonies.
If we let p(x, y) denote the predicate x likes y where x
belongs to the universal set of all University of Manchester
students and y belongs to the universal set of all Beethovens
symphonies then the statements become:
(i) (x)(y)p(x, y)
(ii) (x)(y)p(x, y)
(iii) (y)(x)p(x, y)
(iv) (x)(y)p(x, y)
(v) (y)(x)p(x, y)
(vi) (x)(y)p(x, y)
All have dierent meanings: in particular, (x)(y) is not
the same as (y)(x).
Example. Consider the statements
(i) (x)(y)x < y and
(ii) (y)(x)x < y
where U = R.
Statement (i) is true but statement (ii) is false. Note that
(i) states that whatever number x we choose we can nd a
number y which is greater than x (e.g. y = x + 1). But (ii)
states that there is a number y which is simultaneously greater
than every number x: this is impossible because, with x = y,
x < y does not hold.
1. For real numbers x and y let p(x, y) denote the predicate x = y.
Which of the following statements is false?
(A) (x)(y)p(x, y)
(B) (x)(y)p(x, y)
(C) (x)(y)p(x, y)
Solution: (C) is false, because every number is equal to itself
and therefore the formula (x)(y)p(x, y) which means
there us a number x such that every real number y
is not equal to x
cannot be true.
Another solution: (C) is F because its negation (x)(y)p(x, y)
is T. This can be seen because
(x)(y)p(x, y) (x)(y) p(x, y),
which means
for every x there is y such that x = y
which is obviously T.
Why are (A) and (B) true?
(A) is (x)(y)x = y is true because you can take x = 1 and
y = 2.
(B) is (x)(y)x = y, or
for every x there exists y such that x = y
this is true, because you may take for such y the value y = x +1.

1. > I can not differentiate the true from the false
> when it comes to different arrangements of
> quantifiers or variables after the quantifier.
>
> For example:
> Let the Universal set be Z.
>
> (i) For all x there exists an integer y such that y^2=x.
>
> (ii) For all y there exists an integer x such that y^2=x.
>
> Which one of those statements is true?
> which one is false?
> are they both false or true?
Answer: (ii) is true, (i) is false.
Why (i) is false? If it is true, then, since it is true for all x, it has
to be true for x = 2. So let us plug x = 2 into the statement:
For x = 2 there exists an integer y such that y
2
= x.
but this is the same as to say
there exists an integer y such that y
2
= 2.
But this obviously false there is no such integer y.
Why is (ii) true? Because, for every y, we can set x = y
2
.
For example,
for y = 1 there exists an integer x such that 1
2
= x (indeed,
take x = 1);
for y = 2 there exists an integer x such that y
2
= x (indeed,
take x = 4);
2
= x (indeed,
take x = 9);
2
= x (indeed,
take x = 16);
2
= x (indeed,
take x = 25).
0N1 Mathematics Lecture 10 Predicate Logic Logical equivalences 57
Lecture 10
Logical equivalences
Statements can be formed from predicates by means of a mix-
ture of connectives and quantiers.
Examples.
(i) Let p(x, y) denote x < y and let q(y) denote y = 2.
Then
(x)(y)(p(x, y) q(y))
denotes
For all x there exists y such that x < y and
y = 2.
(This is T).
(ii) Let p(x) denote x > 2 and let q(x) denote x
2
> 4. Then
(x)(p(x) q(x))
denotes
For all x, if x > 2 then x
2
> 4.
(True).
(iii) Let p(x) denote x > 2 and let q(x) denote x < 2. Then
we may form
((x)p(x) (x)q(x)) (x)(p(x) q(x)).
This is F because
(x)p(x) (x)q(x)
is Tbut
(x)(p(x) q(x))
is F . T F gives F .
As in propositional logic, we say that two statements X
and Y are logically equivalent, and write X Y , if X and
Y have the same truth value for purely logical reasons.
Example. (x)p(x) (x)p(x). We dont need to
know the meaning of p(x).
Fundamental logical equivalence (6) of propositional logic
is
p p.
This can be applied to predicate logic to show that
(x)p(x) (x)p(x),
(x)(y)p(x, y) (x)(y)p(x, y),
etc. We can use all of the fundamental logical equivalences
(1)(10) in this way, plus two additional equivalences:
(11) (x)p(x) (x) p(x).
(12) (x)p(x) (x) p(x).
Example of (11). Let U be the set of all University of
Manchester students. Let p(x) denote x is British. Then
(x)p(x) denotes
It is not true that every University of Manchester
student is British
and (x) p(x) denotes
There is a University of Manchester student who
is not British.
These are logically equivalent.
Example of (12). Let U = Z. Let p(x) denote x
2
= 2.
Then (x)p(x) denotes
It is false that there exists x Z such that x
2
=
2
and (x) p(x) denotes
For all x Z, x
2
= 2.
These are logically equivalent.
Example. Prove that
(x)(y)(p(x, y) q(x, y)) (x)(y)(p(x, y) q(x, y)).
Solution.
(x)(y)(p(x, y) q(x, y))
by (11)
(x) (y)(p(x, y) q(x, y))
by (11)
(x)(y) (p(x, y) q(x, y))
by (9)
(x)(y) (p(x, y) q(x, y))
by (8)
(x)(y)(p(x, y) q(x, y))
by (7)
(x)(y)(p(x, y) q(x, y))
Perhaps, the very rst line in this solution needs a com-

ment: we apply rule
(11) (x)p(x) (x) p(x)
with the formula p(x) = (y)(p(x, y) q(x, y)) highlighted
by use of a boldface font.
1. Two of the following statements are contradictions (that is, they
are false no matter how we interpret the predicate p(x, y)). Mark
the statement which is not a contradiction.
(A) (x)(y)(p(x, y) p(x, y))
(B) (x)(y)p(x, y) (y)(x)p(x, y)
(C) (x)(y)p(x, y) (y)(x)p(x, y)
Solution. One of the students wrote:
> I did not understand question 1 of the mock test for this week.
> Please could you translate the sentences and explain the
> answer, as when i tried to do so each seemed non-sensical.
Of course, two of the statements are non-sensical because they are
contradictions: they are never true. For example, (A) is always
false because
p p
is always false, it says p is true if and only if p is true, or p is
true if and only if not p is true; the phrase can be even re-told
as p is true and false simultaneously. Of course, the sentence
p is true and false simultaneously is false no matter what is the
statement p.
Meanwhile, the statement (B),
(x)(y)p(x, y) (y)(x)p(x, y)
is true if we take the set of real numbers R for the universal do-
main and and interpret the predicate p(x, y) as x < y.
2. Two of the following statements are tautologies (that is, they are
true no matter how we interpret the predicate p(x, y)). Mark the
statement which is not a tautology.
(A) (x)(y)p(x, y) (y)(x)p(x, y)
(B) (x)(y)p(x, y) (y)(x)p(x, y)
(C) (x)(y)p(x, y) (x)(y)p(x, y)
Solution. Statements (A) and (C) are always true.
Indeed, to see that (A)
is true observe that both universal state- * I added a detailed explanation on a

request from a student; of course, you do
not have to write it down in a multiple
choice test.
ments
(x)(y)p(x, y) and (y)(x)p(x, y)
can be expressed by a phrase which does not mention names x
and y for variables:
all elements in U are in relation p.
For example, take for U the set of all people and for p(x, y) the
relation x and y are friends; then both
(x)(y)p(x, y) and (y)(x)p(x, y)
become
all people are friends [to each other and themselves]
and
(x)(y)p(x, y) (y)(x)p(x, y)
becomes
if all people are friends then all people are friends,
which is obviously true. And notice that the actual meaning of
the predicate p(x, y) does not matter here.
* If you think that all people are

friends is too optimistic an assertion,
repeat the same argument with the fa-
mous Latin proverb homo homini lupus
est, a man is a wolf to [his fellow] man
[and himself]. I repeat: the actual
meaning of the predicate p(x, y) does not
mater.
Similarly, (C) reads in our interpretation as
if all people are friends then there is someone who
befriends someone,
and again the actual meaning of predicate p(x, y) does not matter.
Statement (B) is false when we take R for the universal domain
and interpret the predicate p(x, y) as x < y.

1. > Since I can not use symbols,
> A=for all and E=there exists.
>
> If there was a statement like this:
>
> ~((Ax)(Ey)(p(x,y)^(Ey)~q(y)))
>
> If I want to simplify this,
> I multiply the negation inside the brackets,
> but I am not sure of what would happen,
> will the negation be multiplied by both (Ax)(Ex) ?
> as it will be
> ~(Ax)~(Ey) ~((p(x,y)^(Ey)~q(y))??
Answer. I am afraid it works dierently. Here is a sequence of
transformations:
((x)(y)(p(x, y) (y) q(y))) (x) (y)(p(x, y) (y) q(y))
(x)(y) (p(x, y) (y) q(y))
(x)(y)(p(x, y) (y) q(y))
(x)(y)(p(x, y) (y) q(y))
(x)(y)(p(x, y) (y)q(y)).
0N1 Mathematics Lecture 11 Methods of Proof 62
Lecture 11
Methods of Proof
* Recommended reading: Book of Proof

by Richard Hammack, Chapter 4.
I. Statements of the form (x)p(x)
To prove (x)p(x) is Twe must prove that p(x) is Tfor all
x U. (The method will vary.)
To prove (x)p(x) is F

we must show that there exists * We also say: disprove (x)p(x);
refute (x)p(x).
at least one x U such that p(x) is F for this x. Such a value
of x is called a counterexample to the statement (x)p(x).
Example. Prove that
For all real numbers x, x
2
3x + 2 0
is false.
Note that
x
2
3x + 2 = (x 1)(x 2).
If 1 < x < 2 then x 1 is positive: x 1 > 0, and x 2 is
negative: x2 < 0, so their product (x1)(x2) is negative:
(x 1)(x 2) < 0.
Thus any number x with 1 < x < 2 is a counterexample: the
statement is false. For a concrete
value of x, we can take * concrete = specic, existing in real-

ity or in real experience; perceptible by
the senses.
x = 1
1
2
. One counterexample is enough: we do not have to
show that
x
2
3x + 2 0
is false for all x.
Example. Prove that the statement
For all sets A, B and C,
A (B C) = (A B) C

is false.
We try to nd a counterexample by experiment. Try A =
, B = , C = {1}. Then
A (B C) =
but
(A B) C = {1}.
Thus A = , B = , C = {1} gives a counterexample: the
statement is false.
NOTE. One counterexample is enough to prove that a
statement is false.
Lecture 12
Methods of proof, continued
II Statements of the form (x)(p(x) q(x))
An example is
For all x, if x > 2 then x
2
> 4.
In practice such a sentence is often expressed as
If x > 2 then x
2
> 4
where the phrase For all x is taken as obvious. However, in
symbols, we should write
(x)(p(x) q(x)).
Example. If A B then A B = B is shorthand
for * shorthand = abbreviation

For all A and all B, if A B then A B = B,
written as
(A)(B)(p(A, B) q(A, B))
where p(A, B) denotes A B and q(A, B) denotes AB = B.
To prove that (x)(p(x) q(x)) is Twe need to prove
that p(x) q(x) is Tfor each element x of U. The truth
table for shows that p(x) q(x) is automatically Twhen
p(x) is F . Therefore we only need to prove that p(x) q(x)
is Tfor elements x of U such that p(x) is T. We take an
arbitrary value of x for which p(x) is Tand try to deduce
that q(x) is T. (The method will vary.) It then follows that
(x)(p(x) q(x)) is T.
Example. Prove the statement
If x is an even integer then x
2
is even.
We rst note the mathematical denition of an even in-
teger: an integer n is even if there exists an integer m such
that n = 2m. Let x be an even integer. Then there exists an
integer y such that x = 2y. Hence
x
2
= 4y
2
= 2(2y
2
).
However, 2y
2
is an integer. Therefore, with m = 2y
2
we have
x
2
= 2m which proves that x
2
is even.
To prove that (x)(p(x) q(x)) is F we have to show
that there exists x U such that p(x) q(x) is F . The
truth table of shows that p(x) q(x) can only be F when
p(x) is Tand q(x) is F . Thus we have to show that there
exists x U such that p(x) is Tand q(x) is F . This will be a
counterexample to (x)(p(x) q(x)).
Example. Prove that the statement
If x is a real number such that x
2
> 4 then x > 2
is false. Let x = 3. Then x
2
> 4 is Tbut x > 2 is F . Thus
x = 3 is a counterexample: the statement is false.
Contrapositive and Converse
* Recommended reading: Book of Proof

by Richard Hammack, Chapter 5.
By the method of truth tables we can prove
* Do that as an exercise!
p q q p.
q p is called the contrapositive of p q. It
follows that
(x)(p(x) q(x)) (x)(q(x) p(x)).
(x)( q(x) p(x)) is called the contrapositive of
(x)(p(x) q(x)).
To prove a statement p q or (x)(p(x) q(x)) it is
enough to prove the contrapositive. Sometimes this is easier.
Example. Prove the statement
If x is an integer such that x
2
is odd then x is
odd.
The contrapositive is
If x is an integer such that x is not odd then x
2
is not odd.
However not odd is the same as even. So the contrapos-
itive is
If x is an even integer then x
2
is even.
We have already proved this.
q p is called the converse of p q.
Similarly,
(x)(q(x) p(x))
is called the converse of
(x)(p(x) q(x)).
The converse is NOT equivalent to the original statement.
Example. Let p be You got full marks and let q be You
passed the exam.
p q is If you got full marks you passed the exam.
The contrapositive q p is
If you did not pass the exam you did not get full
marks.
The converse q p is
If you passed the exam you got full marks.
q p is equivalent to p q, but q p is not.
Example. The statement
If x > 2 then x
2
> 4
is true, but the converse
If x
2
> 4 then x > 2
is false.
* Indeed, give a counterexample!

III. Statements of the form (x)(p(x) q(x))
We make use of the logical equivalence
p q (p q) (q p).
Thus to prove that p q is Tit is sucient to prove two
things
(i) p q is T
(ii) q p is T.
To prove that p q is F it is sucient to prove that either
p q is F or q p is F .
Similarly to prove that (x)(p(x) q(x)) is Twe usually
proceed in TWO STEPS.
(i) We prove (x)(p(x) q(x)).
(ii) We prove (the converse) (x)(q(x) p(x)).
In order to prove (i) we follow the method described in II
above: we take an arbitrary x such that p(x) is Tand try to
deduce that q(x) is T. Than to prove (ii) we take an arbitrary
x such that q(x) is Tand try to deduce that p(x) is T.
To prove that (x)(p(x) q(x)) is F we prove that
(x)(p(x) q(x)) is F
or
(x)(q(x) p(x)) is F .
1. Of the following three statements, two are contrapositive to each
other; mark the statement which is not contrapositive to the other
two.
(A) If an integer is odd then its square is odd.
(B) If an integer is even then its square is even.
(C) If the square of an integer is odd then the integer is odd.
Solution. Let p be the statement the integer is odd and q be
the statement its square is odd. The the statements become
(A) p q; (B) p q; and (C) q p.
Hence (B) and (C) are contrapositive to each other, while (A) is
not contrapositive to (B) or (C). You have to click (A).
0N1 Mathematics Lecture 13 Numbers 69
Lecture 13
Numbers
Some facts about numbers
Suppose m and n are integers. We say that n is divisible by
m if there is an integers k such that n = km.
* Alternative language:
n is divisible by m
m divides n
m is a factor of n
m is a divisor of n
m | n
n is a multiple of m
are all the same.
Example. 24 is divisible by 6. 24 is not divisible by 7. This
is because 24 =
24
7
7 but
24
7
is not an integer.
15 is divisible by 3 because 15 = (5)(3).
A prime number is a positive integer p where p > 1 and p
has no positive factors except 1 and p.
Prime numbers: 2, 3, 5, 7, 11, 13, 17, 19, . . .
Two basic properties of prime numbers.
1. Let p be a prime number and m and n be integers. If p
divides mn then p | m or p | n.
Example. If mn is even (i.e. 2 | mn) then m is even or n
is even.
Example. If 7 | mn then 7 | m or 7 | n.
This property does not hold for non-prime numbers,
e.g. * Non-prime integers n > 1 are called

composite numbers.
6 | mn 6 | m 6 | n
is false. Counterexample: m = 2, n = 3.
2. Every positive integer greater than 1 can be written as
a product
of prime numbers in a unique way except * can be written as a product of prime

numbers
= can be decomposed into a product of
prime numbers
= can be factorised into a product of
prime number
for the order of factors.
This statement will be partially proven in the last lecture
of this course, Page 101 (using mathematical induction): we
shall prove existence, but not uniqueness of factorisation.
Example. 126 = 2 3 7 = 7 2 3.
Let m and n be positive integers. The greatest common
divisor (gcd) of m and n is the largest positive integer which
divides both m and n.
Example. The gcd of 252 and 120 is 12.
Compare the prime factors: 252 = 2 2 3 3 7,
120 = 2 2 2 3 5. The gcd is 2 2 3 = 12.
If the gcd of m and n is 1 we say that m and n are coprime
or relatively prime, or that m and n have no common factors.
Example. 175 and 44 have no common factors. (175 =
5 5 7, 44 = 2 2 11.)
A rational number is a number which can be written as
m
n
where m and n are integers and n = 0.
Example.
22
7
and
3
10
are rational. and
2 are not.
Let a be a rational number. Assume a > 0. We know we
can write
a =
m
n
If m and n have common factor it can be cancelled out so
that we can write
a =
m
where m
and n
have no common factor.

Example.
30
42
=
15
21
=
5
7
.
5 and 7 have no common factor.
Hence every positive rational number can be written in
the form
m
n
where m and n have no common factor.
1. Let p be a prime number and let a, b and c be positive integers.
Suppose that ab and ac are divisible by p but b+c is not divisible
by p. Which of the following statements can be proven from these
assumptions?
(A) a is divisible by p;
(B) b is divisible by p;
(C) abc is divisible by p
2
.

1. > I was wondering if there was any method of finding
> the gcd without just simply guessing random primes.
Next week, I will show you a method (Euclids algorithm) which
works both for numbers and polynomials. It does not requires
factorisation.
> I noticed that the ON1 exam is non-calculator
> so its not a simple as typing values into
> the calculator and seeing if it divides.
Exam involves only problems which can be solved by mental arith-
metic. Be assured, no calculator is needed.
For example, how could the gcd for 270 and 1575 be found with
ease? Notice that 270 = 1027 = 23
3
5 and 1575 = 5315 =
5 5 63 = 5 5 3 3 7 = 3
3
5
2
7.
Hence gcd(270, 1575) = 45. As simple as that.
0N1 Mathematics Lecture 14 By contradiction 72
Lecture 14
Proof by contradiction
* Recommended reading: Richard

Hammack, Book of Proof, Chapter 6.
Suppose we want to prove some statement q. Assume
that q is false, i.e. assume q is true. Try to deduce from
q a statement which we know is denitely false. But a true
statement cannot imply a false one. Hence q must be false,
i.e. q must be true.
* The same can be formulated dier-

ently: notice that
( q F ) q
is a tautology (I leave to you its proof as
an exercise). Therefore if we prove
q F ,
q will follow.
Example. Let p be a prime number and m a positive in-
teger. Suppose that m is divisible by p. Then m + 1 is not
divisible by p.
Proof. Suppose, in order to get a contradiction, that m+1
is divisible by p. We can write m + 1 = ap where a is an
integer. We are also given that m is divisible by p. Hence we
can write m = bp where b is an integer. Then
(m + 1) m = ap bp = (a b)p
i.e. (a b)p = 1. Therefore 1 is divisible by p. This is false.
Hence our assumption that m+1 is divisible by p is false. So
m + 1 is not divisible by p.
Example. There are innitely many prime numbers.
Proof. Assume, in order to get a contradiction, that the
statement is not true. Then there are only nitely many prime
numbers. Let these be p
1
, p
2
, . . . , p
n
.
Consider the number m = p
1
p
2
p
n
(product). Clearly,
m is divisible by p
i
for all i. Hence, by the previous example,
for all i, m+1 is not divisible by p
i
. But m+1 is divisible
by some prime number. If p
1
, . . . , p
n
are only prime numbers
then m + 1 is divisible by p
i
for some i. Hence we came
to the obviously false statement
for all i, m+1 is not divisible by p
i
and m+1
is divisible by p
i
for some i.
Hence the statement that there are only nitely many
prime numbers is false. So there are innitely many prime
numbers.
Example. Let p be a prime number. Then

p is irrational.
Proof. Suppose, to get a contradiction, that

p is rational.
Then we can write
p =
m
n
where m and n are integers with no common factors. Hence
n
p = m
and, after squaring both sides of the equation, we get
n
2
p = m
2
But n
2
is an integer, therefore m
2
is divisible by p, i.e.
p | m
2
.
But (basic properties of prime numbers)
p | m
2
p | m p | m,
that is, p | m. we can write m = kp where k is an integer:
n
2
p = k
2
p
2
n
2
= k
2
p
But k
2
is an integer, therefore n
2
is divisible by p, i.e. p | n
2
.
Basic properties of prime numbers give p | n. So we now have
m and n have no common factors but p | m
and p | n.
This is a false statement. Hence

p is rational must be
false, i.e.

p is irrational.
Tick the correct box:
1. (A simplied version of the next question.)
1. Let p be a prime number and let x and y be positive integers
such that x/y is an integer. What can NOT be proven from the
fact that both x and y are divisible by p ?
(A) xy is divisible by p;
(B) x/y is divisible by p;
(C) x + y is divisible by p.
Answer: (B).
First solution: Let us argue: since p | x then x = pk for some
integer k and xy = pky is divisible by p. Hence (A) is proven.
Also, since p | y then y = pl for some integer l and
x + y = pk + pl = p(k + l)
is divisible by p. Hence (C) is also proven. Therefore only (B)
remains under suspicion and you can tick box (B).
You can also see that (B) is answer from:
Second solution:
Take x = y = p. But x/y = p/p = 1 is * As you can see, the second solutions
of easier problem 1 and harder problem
2 are equally easy.
not divisible by p. Hence (B) is false in that particular case and
therefore cannot be proven from the assumption that both x and
y are divisible by p .
2. (A harder version of the previous question.) Let p be a
prime number and let x and y be positive integers such that x/y
is an integer. What can NOT be proven from the fact that both
xy and x y are divisible by p ?
(A) both x and y are divisible by p;
(B) x/y is divisible by p;
(C) x + y is divisible by p.
Answer: (B).
First solution: Let us argue: since p | xy and p a prime, p | x
or p | y. If p | x then x = pk for some integer k. Then, since
p | x y,
x y = pl
for some integer l. Substituting x = pk, we have
pk y = pl andy = pk pl = p(k l),
hence p | y and (A) is true. But then p | x + y and (C) is true.
Hence, in that case, only (B) can be false.
The case when p | y is is similar: it leads to the conclusion that
(A) and (C) are true.
Second solution:
Take x = y = p. The p | xy and p | x y, * As you can see, the second solutions

of easier problem 1 and harder problem
2 are equally easy.
But x/y = p/p = 1 is not divisible by p. Hence (B) is false
in that particular case and therefore cannot be proven from the
assumption that both xy and x y are divisible by p .
3. Of the following three statements, two are converse to each other;
mark the statement which is not converse to the other two.
(A) If a cat is black then its kittens are black.
(B) If the kittens of a cat are not black then the cat is not
black.
(C) If a cat is not black then its kittens are not black.
Answer: (A).
Solution: Let p means cat is black, q means kittens are
black. Then the statements can be written as
(A) p q;
(B) q p;
(C) p q.
By denition, (B) and (C) are converse to each other.
0N1 Mathematics Lecture 15 Polynomials 76
Lecture 15
Polynomials
Some useful formulae
(a b)(a + b) = a
2
+ ab ba b
2
= a
2
b
2
(a b)(a
2
+ ab + b
2
) = a
3
+ a
2
b + ab
2
a
2
b ab
2
b
3
= a
3
b
3
In general,
a
n
b
n
= (a b)(a
n1
+ a
n2
b + + ab
n2
+ b
n1
)
Special case with b = 1:
a
n
1 = (a 1)(a
n1
+ a
n2
+ + a + 1)
e.g.
a
3
1 = (a 1)(a
2
+ a + 1)
a
2
1 = (a 1)(a + 1)
Polynomials
Let x be a variable standing for any number (real or complex).
The expression
a
n
x
n
+ a
n1
x
n1
+ + a
1
x + a
0
where a
n
, . . . , a
0
are numbers and a
n
= 0 is called a polyno-
mial of degree n.
For example, 2x
3
3x +
1
2
is a polynomial of degree 3,
while
(1 + i)x
2
2ix + 1
is a polynomial of degree 2.
A polynomial of degree 1 is called a linear polynomial :
ax + b (a = 0).
A polynomial of degree 2 is called a quadratic polynomial :
ax
2
+ bx + c (a = 0).
If f(x) is a polynomial and t is a number then f(t) denotes
the value obtained by substitution t for x in f(x).
For example, if f(x) = x
3
2x
2
+ 1 then
f(1) = 1 2 + 1 = 0
f(2) = 8 8 + 1 = 1
f(1) = 1 2 + 1 = 2
f
_
1
2
_
=
1
8

1
2
+ 1 =
1 4 + 8
8
=
5
8
f(i) = i + 2 + 1 = 3 i
Number t is a root (or zero) of f(x) if f(t) = 0.
If f(x) and g(x) are polynomials we say that f(x) is di-
visible by g(x) if there is a polynomial h(x) such that f(x) =
g(x)h(x).
Alternative terminology: f(x) is divisible by g(x), g(x) is
a factor (or divisor) of f(x), g(x) divides f(x), g(x) | f(x).
For example,
(2x 3)(x
2
+ 3x 1) = 2x
3
+ 6x
2
2x 3x
2
9x + 3
= 2x
3
+ 3x
2
11x + 3
So 2x
3
+3x
2
11x +3 is divisible by 2x 3 and x
2
+3x 1.
Another example:
(x
n
1) = (x 1)(x
n1
+ x
n2
+ + x + 1),
therefore x
n
1 is divisible by x 1.
More generally, if t is any number then
(x
n
t
n
) = (x t)(x
n1
+ tx
n2
+ + t
n2
x + t
n1
),
so x
n
t
n
is divisible by x t. For example,
x
3
8 = (x 2)(x
2
+ 2x + 4).
Lecture 16
Polynomials, continued
Division with remainder
If f(x) is a polynomial and ax + b is a linear polynomial we
can investigate whether ax + b is a factor of f(x) by long
division.
For example, if ax + b = 2x + 1 and f(x) = 2x
3
3x + 2
then
x
2
1
2
x
5
4
2x +1 | 2x
3
+0x
2
3x +2
2x
3
+x
2
x
2
3x
x
2
1
2
x
5
2
x +2
5
2
x
5
4
13
4
which means that
2x
3
3x + 2 = (2x + 1)
_
x
2
1
2
x
5
4
_
+
13
4
.
The remainder is
13
4
. The polynomial 2x
3
3x + 2 is not
divisible by 2x + 1.
In general we can write
f(x) = (ax + b)g(x) + R
where R is the remainder upon division of f(x) by ax + b.
Remainder Theorem. When f(x) is divided by ax+b the
remainder is f(
b
a
).
Proof. Write
f(x) = (ax + b)g(x) + R
where R is a constant. Substitute x =
b
a
:
f
_
b
a
_
= 0 g
_
b
a
_
+ R
which means
R = f
_
b
a
_
.
Example. Find whether 2x + 1 is a factor of 2x

3
3x + 2.
Substitute x =
1
2
into f(x) = 2x
3
3x + 2. We get
f
_
1
2
_
=
1
4
+
3
2
+ 2 =
1 + 6 + 8
4
=
13
4
.
The remainder on dividing 2x
3
3x + 2 by 2x + 1 is
13
4
. So
2x
3
3x + 2 is not divisible by 2x + 1.
Example. Find the remainder when x
3
2x
2
+1 is divided
by x 1.
Substitute x = 1 in f(x) = x
3
2x
2
+ 1. We get
f(1) = 1 2 + 1 = 0.
The remainder is 0. (So x
3
2x
2
+1 is divisible by x 1.)
Factor Theorem. x t is a factor of f(x) if and only if t
is a root of f(x) (i.e. f(t) = 0).
Proof. If x t is a factor of f(x) we can write
f(x) = (x t)h(x)
and so
f(t) = (t t)h(t) = 0 h(t) = 0,
so t is a root of f(x).
Conversely, suppose that t is a root of f(x). Then by
the Remainder theorem f(x) is divisible by x t since the
remainder is f(t) = 0. So x t is factor of f(x).
Example Find whether x +
1
2
is a factor of
2x
4
+ x
3
x
2
+
3
2
x + 1.
Substitute x =
1
2
in f(x) = 2x
4
+ x
3
x
2
+
3
2
x + 1.
f
_
1
2
_
=
1
8

1
8

1
4

3
4
+ 1 = 0.
Hence
1
2
is a root of f(x). So, by the Factor Theorem, x+
1
2
is a factor of f(x).
To nd the factorisation we need to do the long divi-
sion:
2x
3
0x
2
x +2
x +
1
2
| 2x
4
+x
3
x
2
+
3
2
x +x
2x
4
+x
3
0 x
2
+
3
2
x
x
2
1
2
x
2x +1
2x +1
0
Hence
2x
4
+ x
3
x
2
+
3
2
x + 1 =
_
x +
1
2
_
(2x
3
x + 2).
Repeated Factor Theorem. x t is a repeated factor of
f(x) if and only if t is a root of both f(x) and f
(x).
Proof. Proof is omitted.
Example. Prove that x + 1 is a repeated factor of
f(x) = x
5
+ 2x
4
3x
2
3x 1.
We compute
f(1) = 1 + 2 3 + 3 1 = 0.
Hence 1 is a root of f(x). Also
f
(x) = 5x
4
+ 8x
3
6x 3
f
(1) = 5 8 + 6 3 = 0,
so 1 is a root of f
(x). By the Repeated Factor Theorem,

x + 1 is a repeated factor of f(x).
In the last example if we want to factorise f(x) we have
to divide f(x) (x + 1)
2
as follows:
x
3
+0x
2
x 1
x
2
+2x +1 | x
5
+2x
4
+0x
3
3x
2
3x 1
x
5
+2x
4
x
3
0x
4
x
3
3x
2
0x
4
+0x
3
+0x
2
x
3
3x
2
3x
x
3
2x
2
x
x
2
2x 1
x
2
2x 1
0
Hence
x
5
+ 2x
4
3x
2
3x 1 = (x
2
+ 2x + 1)(x
3
x 1).
Notice that the root t corresponding to the repeated factor
(xt) of a polynomial f(x) is called a repeated root of f(x).
* Some textbook use the words a mul-

tiple root.

1. > Could you please help me with the following question:
>
> Prove that if a is a repeated root of the polynomial
> f(x) then a is a root of f(x).
Answer. So, we need this:
Prove that if is a repeated root of the polynomial f(x) then
is a root of f
(x).
Actually, it was in the lectures. Proof goes that way:
Assume that is a repeated root of the polynomial f(x). Then
f(x) = (x )
k
g(x)
for some polynomial g(x) and k 2. We can dierentiate this
equality using the rule
(p(x)q(x))
= p
(x)q(x) + p(x)q
(x) :
f(x)
= [(x )
k
g(x)]
= [(x )
k
]
g(x) + (x )
k
[g(x)]
= k(x )
k1
g(x) + (x )
k
g
(x)
= (x )
k1
[kg(x) + (x )g
(x)]
Since k 2, k 1 1 and x is a divisor of f
(x). Hence is
a root of f(x).
2. > I tried solving this particular question:
> "(c) [6 marks] Given that the polynomial
> x3 -3a2x+a has a repeated root,
> find the possible values of a (real or complex)."
> So far i have three answers to this question:
> zero, + and - under root of half.
> I wonder if i am even right!
> Could please send a detailed working,
> I shall highly appreciate it!
Answer. It is almost right. Here is the working:
Since the polynomial has a repeated root, its derivative has the
same root. The derivative of x
3
3a
2
x + a is 3x
2
3a
2
and
has roots a. Hence either a or a is a root of the original
polynomial, x
3
3a
2
x + a.
Assume rst that a is a root of x
3
3a
2
x + a, then
a
3
3a
2
a + a = 0
We can take the common factor a out of the left hand side of the
equation and get
a(2a
2
+ 1) = 0,
which gives us solutions a = 0 or a =
1
2
exactly as you stated.
We can turn now to the remaining case when a is a root of
x
3
3a
2
x + a and, correspondingly,
(a)
3
3a
2
(a) + a = 0.
After rearranging the left hand side we have
a
3
+ 3a
3
+ a = 0,
or
2a
3
+ a = 0,
which, in its turn, can be factorised as
a(2a
2
+ 1) = 0.
Hence we nd a = 0 or a =
i
2
an extra solution.
3. > i need help with this exercise
> "find the repeated roots of the polynomial 4x^4+x^2+3x+1"
This is is a tricky one; believe me, exam problems are easier.
We know that a repeated root x = t of a polynomial f(x) is also
a root of f
(x). Therefore a repeated root of

f(x) = 4x
4
+ x
2
+ 3x + 1
is also a root of
f
(x) = 16x
3
+ 2x + 3.
But a common root x = t of polynomials f(x) and f
(x) is also a
root of any polynomial of the form
g(x) = a(x)f(x) + b(x)f
(x);
indeed, substitute:
g(t) = a(t)f(t) + b(t)f
(t) = a(t) 0 + b(t) 0 = 0 + 0 = 0.

hence we can try to nd a smaller polynomial with the same root.
Take
4f(x)xf
(x) = 4(4x
4
+x
2
+3x+1)x(16x
3
+2x+3) = 2x
2
+9x+4.
This is already quadratic polynomial, and has roots x = 1/2
and x = 4. Of these two roots, we can check that x = 1/2 is
indeed a root of both f(x) and f
(x):
f
1
2
_
= 2 1 + 3 = 0,
f
_
1
2
_
=
1
4
+
1
4

3
2
+ 1 = 0.
and therefore x = 1/2 is a repeated root of f(x).
Thus
1
2
is a repeated root of f(x). Dividing f(x) by
(2x + 1)
2
= 4x
2
+ 4x + 1
gives
f(x) = (4x
2
+ 4x + 1)(x
2
x + 1).
But the roots of x
2
x + 1 are
1
2
+
1
2
3 and
1
2

1
2
3. Thus
1
2
is the only repeated root of f(x).
Lecture 17
Polynomials, continued
Fundamental Theorem of Algebra
Let
f(x) = a
n
x
n
+ a
n1
x
n1
+ + a
1
x + a
0
be any polynomial of degree n (that is, a
n
= 0), where
a
n
, . . . , a
0
are real or complex
numbers. Then f(x) factorises * By that time, you should know com-
plex numbers from a parallel lecture
course. A brief reminder: complex num-
bers are numbers of the form a + bi,
where i =
1 and a and b are real

numbers. The set of complex numbers
is denoted C.
Notice that in Electrical Engineering the
letter j is used for
1 in notation for
complex numbers, because i is reserved
for current.
as
f(x) = a
n
(x
1
)(x
2
) (x
n
)
where
1
, . . . ,
n
are real or complex numbers.
Of course,
* Some frequently used Greek letters:
alpha
beta
gamma
delta
epsilon
lambda
mu
nu
pi
rho
1
, . . . ,
n
are roots of f(x) but notice that some of the roots
may be equal. For example,
x
3
3x
2
+ 3x 1 = (x 1)
3
has three equal roots.
The polynomial
x
4
+ 2x
2
+ 1 = (x
2
+ 1)
2
= [(x i)(x + i)]
2
= (x i)
2
(x + i)
2
has two roots i and two roots i.
Polynomials with real coecients
Let
f(x) = a
n
x
n
+ a
n1
x
n1
+ + a
1
x + a
0
where a
n
= 0 and a
0
, . . . , a
n
are real. Then f(x) can be
written as a product of linear and quadratic polynomials with
real coecients.
Example. Let f(x) = x
3
+ x 2. Then
f(x) = (x 1)(x
2
+ x + 2)
= (x 1)(x
2
)(x
3
)
Notice that the polynomials (x 1) and (x
2
+ x + 2) (in the
rst factorisation) have real coecients, while the roots
1
and
2
of the quadratic polynomial x
2
+ x + 2 are
2,3
=
1
1
2
4 1 2
2
=
1
7
2
=
1
7i
2
, and therefore
2
=
1
2
+
7
2
i
3
=
1
2

7
2
i
Corollary. A polynomial
f(x) = a
n
x
n
+ a
n1
x
n1
+ + a
1
x + a
0
with real coecients and of odd degree n has at least one real
root.
In particular, a cubic polynomial
f(x) = a
3
x
3
+ a
2
x
2
+ a
1
x + a
0
with real coecients (a
3
= 0) has at least one real root.
Sample Test

1. What are the roots of the polynomial
f(x) = x
4
+ 2x
3
+ x
2
?
(A) 0, 0, 1, 1, (B) 1, 1, 0, 1 (C) 1, 1, 0, 0.
Answer: (C). Indeed,
x
4
+ 2x
3
+ x
2
= x
2
(x
2
+ 2x + 1) = (x 0)
2
(x + 1)
2
.
2. Let , and be roots of the equation
g(x) = x
3
3x
2
+ 2x.
Find .
(A) 0. (B) 2. (C) 3.
Answer: (A). Indeed,
x
3
3x
2
+ 2x = x(x
2
3x + 2)
and one of the roots is obviously 0. Hence the product of all roots
is 0. Notice that there is no need to nd all roots.
0N1 Mathematics Lecture 18 Roots and Coefficients 87
Lecture 18
Relations between the roots and
the coecients of a polynomial
Consider
ax
2
+ bx + c = a(x )(x )
The right hand side (RHS) is
* Typo corrected in the following

equation
a(x
2
x x + ) = a[x
2
( + )x + ()]
Hence
b
a
= ( + )
c
a
=
or
+ =
b
a
=
c
a
Therefore
the sum of the roots of ax
2
+ bx + c is
b
a
,
the product of the roots of ax
2
+ bx + c is
c
a
.
For a cubic polynomial, we have
ax
3
+ bx
2
+ cx + d = a(x )(x )(x )
Here,
* RHS = right hand side

LHS = left hand side
RHS = a(x
2
x x + )(x )
= a(x
3
x
2
x
2
x
2
+ x + x + x )
LHS = a
_
x
3
+
b
a
x
2
+
c
a
x +
d
a
_
Hence
+ + =
b
a
+ + =
c
a
=
d
a
For a polynomial of degree four
ax
4
+ bx
3
+ cx
2
+ dx + e = a(x )(x )(x )(x ),
and arguing as above we conclude that
+ + + =
b
a
+ + + + + =
c
a
+ + + =
d
a
=
e
a
Example. If the equation x
3
+ 2x
2
x + 5 has roots ,
and , nd
(i)
2
+
2
+
2
;
(ii)
2
2
+
2
2
+
2
2
.
We have
+ + = 2
+ + = 1
= 5
(i) Observe that
( + + )
2
=
2
+
2
+
2
+ 2 + 2 + 2.
Hence
2
+
2
+
2
= ( + + )
2
2( + + )
= 4 + 2
= 6.
(ii) Observe that
( + + )
2
=
2
2
+
2
2
+
2
2
+ 2
2
+ 2
2
+ 2
2
=
2
2
+
2
2
+
2
2
+ 2( + + )
Hence
2
+
2
2
+
2
2
= ( + + )
2
2( + + )
= 1 2 (5) (2)
= 1 20
= 19.
Example. If , and are the roots of the equation

x
3
2x
2
+ 2x + 1 = 0
nd the equations with the roots
(i)
1
,
1
,
1
(ii)
2
,
2
,
2
(iii) 1, 1, 1.
Note + + = 2, + + = 2, = 1.
(i) We nd
1
+
1
+
1
,
1
+
1
+
1
, and
1
.
Notice
1
+
1
+
1
=
+ +
= 2,
1
+
1
+
1
=
+ +
= 2,
1
=
1
= 1.
If the required equation with roots
1
,
1
,
1
is
x
3
+ bx
2
+ cx + d = 0
we have, by previous calculations,
b =
1
+
1
+
1
= 2
c =
1
+
1
+
1
= 2
d =
1
= 1.
Hence the equation is
x
3
+ 2x
2
2x + 1 = 0.
(ii) We nd
2
+
2
+
2
,
2
2
+
2
2
+
2
2
and
2
2
.
We already know that
2
+
2
+
2
= ( + + )
2
2( + + )
and
2
+
2
2
+
2
2
= ( + + )
2
2( + + )
Hence
2
+
2
+
2
= 2
2
2 2 = 0
and
2
+
2
2
+
2
2
= 2
2
2 (1) 2 = 8.
We can also compute
2
= ()
2
= (1)
2
= 1.
The required equation is x
3
+bx
2
+cx +d = 0 where b = 0,
c = 8, d = 1, that is,
x
3
+ 8x 1 = 0.
(iii) We nd
( 1) + ( 1) + ( 1),
( 1)( 1) + ( 1)( 1) + ( 1)( 1),
( 1)( 1)( 1).
The calculations are a bit messy and I switch to smaller font:
( 1) + ( 1) + ( 1) = ( + + ) 3
= 2 3
= 1,
( 1)( 1) + ( 1)( 1) + ( 1)( 1) = + + + 3
= + + 2( + + ) + 3
= 2 2 2 + 3
= 1,
( 1)( 1)( 1) = ( 1)( + 1)
= + + 1
= ( + + ) + ( + + ) 1
= 1 2 + 2 1
= 2.
The required equation is x
3
+bx
2
+cx+d = 0 where b = 1,
c = 1, d = 2, that is,
x
3
+ x
2
+ x + 2 = 0.
Sample test problems

1. Let and be roots of the polynomial
g(x) = x
2
+ 2x + 5.
Find a polynomial with roots 1 and 1.
(A) x
2
+ 4x + 8.
(B) x
2
4x + 8.
(C) x
2
+ 4x + 5.
Answer: (A)
Solution: The new polynomial is
x
2
[( 1) + ( 1)]x + ( 1)( 1)
which can be rearranged as
x
2
[( + ) 2]x + [ ( + ) + 1)
Since, by the root/coecients formula applied to g(x) we have
+ = 2, = 5,
the new polynomial is
x
2
+ 4x + 8.
2. Two roots of the polynomial
f(x) = x
3
2x
2
+ x + a
equal = 0 and = 1. Find the value of the coecient a and
the third root .
(A) a = 1, = 0
(B) a = 1, = 1
(C) a = 0, = 1.
Answer: (C)
Solution: The free term a of the polynomial is minus product
of roots, and, since one of the roots is 0, it has to be equal 0:
a = = 0 1 = 0
Hence the polynomial equals
f(x) = x
3
2x
2
+ x
= x(x
2
2x + 1)
= x(x 1)
2
and the third root = 1.

The following question refers to Problem 5(c) from January
2008 Examination:
If and , are the roots of the equation
x
2
2x + 3 = 0,
nd
1
2
+
1
2
.
> I have got a question about the solution to the task 5 C in 2008 exam paper.
> We should find what 1/A^2 + 1/B^2 is equal to. The problem I have
> encountered is that I cannot find the roots of the equation. A+B=2 and AB=3,
> when I try to rearrange this equation I get something like B^2 - 2B +3=0
> Which is the same as the original equation. In addition, when I use
> discriminant method, I get that discriminant is below zero, namely x= (2+-
> sqrt(4 - 12))/ 2 and discriminant=4-12=-8. Accordingly, there should be no
> roots. If my calculations are correct, then how is it possible to calculate
> the value of 1/A^2 + 1/B^2 if there are no roots?
Indeed, the solution can be found without evaluating the
roots and without even noticing that and are in eect
complex:
1
2
+
1
2
=

2
+
2
2
=
( + )
2
2
()
2
.
But + = 2 and = 3, hence
1
2
+
1
2
=
2
2
2 3
3
2
=
2
9
.
A straightforward solution which involves computation of roots
is more complex. First, we have to nd and . The roots
of
x
2
2x + 3 = 0
are
x
1,2
=
2
2
2
4 3
2
=
2
8
2
= 1
2i,
so we can take
= 1 +
2i and = 1
2i.
Then
1
2
+
1
2
=
1
(1 +
2i)
2
+
1
(1
2i)
2
=
1
1 + 2
2i 2
+
1
1 2
2i 2
=
1
1 + 2
2i
+
1
1 2
2i
=
(1 2
2i) + (1 + 2
2i)
(1 + 2
2i)(1 2
2i)
=
2
(1)
2
(2
2i)
2
=
2
1 (8)
=
2
9
.
0N1 Mathematics Lectures 1920 Mathematical Induction 95
Lectures 1920
Principle of Mathematical Induction
* Recommended reading: Richard

Hammack, Book of Proof, Chapter 10.
Let p
1
, p
2
, p
3
, . . . be an innite sequence of statements, one
statement p
n
for each positive integer n. For example,
p
1
is 9
1
1 is divisible by 8
p
2
is 9
2
1 is divisible by 8
p
3
is 9
3
1 is divisible by 8
so for each positive integer n, p
n
is the statement
p
n
is 9
n
1 is divisible by 8.
Suppose that we have the following information
(1) p
1
is true.
(2) The statements
p
1
p
2
, p
2
p
3
, p
3
p
4
, p
4
p
5
. . .
are all true, i.e.
p
k
p
k+1
is true for each positive integer k.
Then we can deduce
p
1
is true and p
1
p
2
is true implies p
2
is true,
p
2
is true and p
2
p
3
is true implies p
3
is true,
p
3
is true and p
3
p
4
is true implies p
4
is true,
that is,
p
1
, p
2
, p
3
, . . . are all true i.e.
p
n
is true for all n.
Example. Prove that 9
n
1 is divisible by 8 for every pos-
itive integer n.
Proof We use mathematical induction. For each integer n,
p
n
is the statement 9
n
1 is divisible by 8.
(1) p
1
is the statement 9
1
1 is divisible by 8. This is clearly
true.
(2) Suppose p
n
is true for n = k, that is, 9
k
1 is divisible
by 8. We want to deduce p
n
for n = k+1, i.e. 9
k+1
1 is
divisible by 8. Since 9
k
1 is divisible by 8, 9
k
1 = 8a
for an integer n. Hence 9
k
= 8a + 1 and
9
k+1
= 9 (8a + 1) = 9 8a + 9.
After subtracting 1 from both sides we have
9
k+1
1 = 9 8a + 8 = 8(9a + 1)
which means that 9
k+1
1 is divisible by 8. Hence p
n
is true for n = k + 1.
Hence, by mathematical induction, p
n
is true for all n.
Example. Prove that
1
2
+ 2
2
+ 3
2
+ + n
2
=
1
6
n(n + 1)(2n + 1)
for every positive integer n.
Proof For each positive integer n, p
n
denotes the statement
1
2
+ 2
2
+ 3
2
+ + n
2
=
1
6
n(n + 1)(2n + 1)
In particular,
p
1
is 1
2
=
1
6
1 2 3 T
p
2
is 1
2
+ 2
2
=
1
6
2 3 5 T
.
.
.
.
.
.
p
k
is 1
2
+ 2
2
+ + k
2
=
1
6
k(k + 1)(2k + 1)
p
k+1
is 1
2
+ 2
2
+ + k
2
+ (k + 1)
2
=
1
6
(k + 1)((k + 1) + 1)(2(k + 1) + 1)
(1) p
1
is the statement 1
2
=
1
6
1 2 3 which is
clearly true.
(1) Suppose the statement p
n
is true for n = k, i.e.
1
2
+ 2
2
+ + k
2
=
k(k + 1)(2k + 1)
6
.
Add (k + 1)
2
to both sides:
1
2
+ 2
2
+ + k
2
+ (k + 1)
2
=
k(k + 1)(2k + 1)
6
+ (k + 1)
2
=
k(k + 1)(2k + 1) + 6(k + 1)
2
6
and now, in the numerator,
I take the common factor
(k + 1) out:
=
(k + 1)[k(2k + 1) + 6(k + 1)]
6
and rearrange the expression
in the square brackets:
=
(k + 1)[2k
2
+ k + 6k + 6]
6
=
(k + 1)[2k
2
+ 7k + 6]
6
and factorise the expression
in the square brackets:
=
(k + 1)(k + 2)(2k + 3)
6
and nally rearrange:
=
(k + 1) [(k + 1) + 1] [2(k + 1) + 1]
6
.
But this is the statement p
n
for n = k + 1 as required.
n
is true for all n.
Example. Prove, by induction on n, that
1 + 3 + 5 + + (2n 1) = n
2
Solution. For each positive integer n, p
n
denotes the state-
ment
1 + 3 + 5 + + (2n 1) = n
2
In particular,
p
1
is 1 = 1
2
T
p
2
is 1 + 3 = 2
2
T
.
.
.
.
.
.
p
k
is 1 + 3 + + (2k 1) = k
2
p
k+1
is 1 + 3 + + (2k 1) + (2k + 1)
2
= (k + 1)
2
(1) p
1
is the statement 1 = 1
2
which is clearly true.
(1) Suppose the statement p
n
is true for n = k, i.e.
1 + 3 + + (2k 1) = k
2
.
Add (2(k + 1) 1) = 2k + 1 to both sides:
1 + 3 + + (2k 1) + (2k + 1) = k
2
+ (2k + 1)
= k
2
+ 2k + 1
= (k + 1)
2
.
But this is the statement p
n
for n = k as required.
n
is true for all n.
Example (Examination of January 2007). Let p
1
de-
note the statement
1
2
= 1
1
2
;
furthermore, for each positive integer n, let p
n
denote the
statement
1
2
+
1
2
2
+ +
1
2
n
= 1
1
2
n
.
Prove, by induction, that p
n
is true for all n.
Solution. Basis of induction is the statement p
1
,
1
2
= 1
1
2
;
it is obviously true.
Inductive step: We need to prove that p
k
p
k+1
for
all k. To do that, assume that p
k
is true, that is,
1
2
+
1
2
2
+ +
1
2
k
= 1
1
2
k
.
Form this identity, we need to get p
k+1
. This is achieved by
adding
1
2
k+1
to the both sides of the equality p
k
:
_
1
2
+
1
2
2
+ +
1
2
k
_
+
1
2
k+1
=
_
1
1
2
k
_
+
1
2
k+1
.
But the righthand side simplies as
* A typo was corrected in the following

calculation.
1
1
2
k
+
1
2
k+1
= 1
2
2 2
k
+
1
2
k+1
= 1
2
2
k+1
+
1
2
k+1
= 1
_
2
2
k+1

1
2
k+1
_
= 1
2 1
2
k+1
= 1
1
2
k+1
and the result of this rearrangement is
1
2
+
1
2
2
+ +
1
2
k
+
1
2
k+1
= 1
1
2
k+1
,
which is exactly the statement p
k+1
. This completes the proof
of the inductive step.
Examples with briefer solutions
Example. Prove by induction on n that
1 + 2 + 3 + + n =
1
2
n(n + 1)
Solution.
Let p
n
be the statement 1 + 2 + + n =
1
2
n(n + 1).
p
1
is the statement 1 =
1
2
1 2. This is clearly true.
Suppose p
n
is true for n = k, i.e. 1+2+ +k =
1
2
k(k+1).
Then
1+2+ +k+(k+1) = (1+2+ +k)+(k+1) =
1
2
k(k+1)+(k+1)
=
1
2
k(k + 1) +
1
2
2(k + 1) =
1
2
(k + 1)(k + 2).
Thus
1 + 2 + + k + (k + 1) =
1
2
(k + 1)((k + 1) + 1).
Therefore p
n
is true for n = k + 1. By induction, p
n
is
true for all n.
Example. Prove, by induction on n, that n < 2
n
for every
positive integer n.
Solution. Basis of induction, n = 1:
1 < 2
1
is obviously true.
Inductive step. Assume that, for some k > 1,
k < 2
k
is true. Since 1 < k by assumption, we also have
1 < 2
k
.
Add the two inequalities together:
k + 1 < 2
k
+ 2
k
= 2
k+1
.
This proves the inductive step.
Example. Prove that every natural number is a product of
prime numbers.
* See properties of prime numbers on

Page 69.
Solution. Regarding products, it is useful to think of them
as products of lists of numbers; for example, the product of
the list 2, 2, 3 is 12. But a list can consist of just one
number, say 5, in which case its product is that number: 5.
If the list is empty: , its product can be usefully thought
to be 1, because in that case appending an empty list to any
list of numbers does not change the product of the latter list.
Please notice that we distinguish between lists and sets: list
can include repeated entries, while set cannot.
So, upon agreeing on this terminology, we consider state-
ments
p
n
: Every natural number which is less or equal then n is a
product of a list of prime numbers.
Statement p
1
is the Basis of Induction: 1 is the only num-
ber which is less or equal than 1, and 1 is the product of an
empty list of prime numbers. Hence p
1
is T.
Inductive Step: Assume now that p
k
is true and con-
sider number k + 1. Every number which is strictly less than
k+1 is less or equal than k and is therefore a product of a list
of prime numbers by the inductive assumption. So we need to
look only at k +1. If k +1 is prime than k +1 is the product
of list k + 1 of prime numbers. If k + 1 is not a prime,
then, since it is bigger than 1, it is a product of two numbers,
k + 1 = a b, both of which are smaller than k + 1. Hence a
is the product of list q
1
, q
2
, . . . , q
l
and b is the product of
a list r
1
, r
2
, . . . , r
m
of prime numbers, and we only need to
put these two lists together, making
k + 1 = q
1
q
2
q
l
r
1
r
2
r
m
the product of the list q
1
, q
2
, . . . , q
l
, r
1
, r
2
, . . . , r
m
of prime
numbers. This proves the Inductive Step.

> I have been on the internet looking up and doing further examples of
> proof by induction, as practice, and have come across something i dont
> understand. i wonder if you can help me. it is not an example you have
> used i think, but maybe you will enjoy a challenge?
>
> here is the link it is the second example
> http://zimmer.csufresno.edu/~larryc/proofs/proofs.mathinduction.html
>
> this is the question:
Theorem. Consider the sequence
a
0
, a
1
, a
2
, . . .
dened by
a
0
=
1
4
and a
n+1
= 2a
n
(1 a
n
) for n > 0.
A formula for the sequence an dened above, is
a
n
=
1
1
2
2
n
2
for all n greater than or equal to 0.
Proof. (By Mathematical Induction.)
Initial Step. When n = 0, the formula gives us
1
1
2
2
n
2
=
1
1
2
2
=
1
4
= a
0
.
So the closed form formula gives us the correct answer when
n = 0.
Inductive Step. Our inductive assumption is: Assume
there is a k, greater than or equal to zero, such that
a
k
=
1
1
2
2
k
2
.
We must prove the formula is true for n = k + 1.
First we appeal to the recursive denition of
a
k+1
= 2a
k
(1 a
k
).
Next, we invoke the inductive assumption, for this k, to get
a
k+1
= 2
1
1
2
2
k
2

_
1
1
1
2
2
k
2
_
=
_
1
1
2
2
k
_
_
1 +
1
2
2
k
2
_
=
1
1
2
2
k+1
2
.
This completes the inductive step.
> The very last step, from
> (1 - 1/2^2^k)(1 + 1/2^2^k)/2 = (1 - 1/2^2^k+1)/2.
> i cant figure out how he got from one to the other. do you
> know?
> i am also not completely sure if the 1/2^2^k means 1/(2^2^k) or
> (1/2)^2^k. i assumed the latter.
I repeat the last calculation in more detail; watch what is
happening in the square brackets.
a
k+1
= 2
1
1
2
2
k
2

_
1
1
1
2
2
k
2
_
= 2
1
1
2
2
k
2

_
_
2
_
1
1
2
2
k
_
2
_
_
=
_
1
1
2
2
k
_
_
1 +
1
2
2
k
2
_
=
_
1
1
2
2
k
_
_
1 +
1
2
2
k
_
2
=
1
1
2
2
k+1
2
.
0N1 Mathematics Appendices 104
Appendix I: Laws of Boolean Algebra
A B = B A
A B = B A
_
A A = A
A A = A
_
idempotent laws (2)
A (B C) = (A B) C
A (B C) = (A B) C
_
associative laws
(3)
A (B C) = (A B) (A C)
A (B C) = (A B) (A C)
_
distributive laws
(4)
A (A B) = A
A (A B) = A
_
absorbtion laws (5)
A U = A A U = U
A = A A =
(6)
(A
= A A A
= U
=
A A
= U
= U
(7)
(A B)
= A
(A B)
= A
_
De Morgans laws (8)
0N1 Mathematics Appendices 105
Appendix II: Laws of Propositional
Logic
p q = q p
p q = q p
_
p p = p
p p = p
_
idempotent laws (2)
p (q r) = (p q) r
p (q r) = (p q) r
_
p (q r) = (p q) (p r)
p (q r) = (p q) (p r)
_
distributive laws
(4)
p (p q) p
p (p q) p
_
absorbtion laws (5)
p T = p p T = T
p F = p p F = F
(6)
(p) = p p p = F T = F
p p = T F = T
(7)
(p q) = p q
(p q) = p q
_
De Morgans laws (8)
p q p q (9)
(p q) (p q) (q p) (10)
Equivalences relating and :
(x)p(x) (x) p(x) (11)
(x)p(x) (x) p(x) (12)

Set Theory

Uploaded by

Copyright:

Available Formats

Set Theory

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Set Theory

Uploaded by

Copyright:

Available Formats

Alexandre Borovik

* As of 201314 academic year

* The date is for 201314 academic year

* marginal = written on the margin,

, that is * or descriptive form

description of the elements in the set). For * explicit = specic, denite

: is used to denote such * Typographical terms:

. It was said * Some books call it null set.

It is usually denoted by . Thus * Notice the use of denite article THE.

* This section contains no compulsory

* We say that is a transitive relation

If A is a set with n elements then A has 2

* indenitely = for ever, without end

is also innite: * There is no universal agreement about

* This section contains no compulsory

* The typographic symbol is some-

* The typographic symbol is some-

or is understood in the * Connective is a word like or, and,

(from each other). * Or that A and B do not intersect.

of A and pronounced A prime) denotes * Notice that the complement A

are as follows. (A, B and C denote * Or laws of Boolean algebra.

* Marking scheme: 2 marks for a cor-

* This section contains no compulsory

because * obvious = evident, self-evident

* unambiguous = unmistakable, de-

* A good example when the use of a

element of [1, 2], this means that * arbitrary = taken at random

method was * alternative = other, another, dierent

from this picture that * see = observe, notice

a = 18, (ii) gives e = 5, (vi) gives g = 3, * gives = yields

| (make a Venn diagram!),

Y | (this is what we have to nd),

|, see a Venn diagram below.

* This section contains no compulsory

something. It is either true or false. If true, we say * assert = state, claim

pq has truth * convention = custom, agreement

Here is the truth table: * The typographical symbol is called

* The typographical symbol is called

* In Computer Science, the exclusive

of p and is written p. * Symbols sometimes used to denote

* This example is expanded at the end

* This section contains no compulsory

is true * In the conditional statement H C ,

* Notice that, in mathematical litera-

statement may be built up from statements * compound = complex, composite

* Some typographic terminology: in the

* This section contains no compulsory

for every choice of * i.e. = that is,

* This section contains no compulsory

* This section contains no compulsory

Perhaps, the very rst line in this solution needs a com-

is true observe that both universal state- * I added a detailed explanation on a

* If you think that all people are

* This section contains no compulsory

* Recommended reading: Book of Proof

value of x, we can take * concrete = specic, existing in real-

0N1 Mathematics Lecture 11 Methods of Proof 63

for * shorthand = abbreviation

* Recommended reading: Book of Proof

* Indeed, give a counterexample!