What is mathematics?

Mathematics is not best learned passively; you don’t sop it up like

a romance novel. You’ve got to go out to it, aggressive, and alert,
like a chess master pursuing checkmate.
Robert Kanigel (1991).
No one would doubt that a mathematics book is not like a novel. It is full
of formulae using strange symbols and Greek letters, and contains words like
‘theorem’, ‘proposition’, ‘lemma’, ‘corollary’, ‘proof’, and ‘conjecture’. Many of
these words are themselves Greek in origin.
This is the legacy of Pythagoras, who was probably the first mathematician in
anything like the modern sense (as opposed so somebody who used mathematics,
such as a surveyor or an accountant). We know little about Pythagoras, and
what we do know is unreliable, but it is clear that he cared very deeply about
the subject:
2 Introduction
the word ‘theory’ . . . was originally an Orphic word, which
Cornford interprets as ‘passionate sympathetic contemplation’ . . .
For Pythagoras, the ‘passionate sympathetic contemplation’ was
intellectual, and issued in mathematical knowledge . . . To those
who have reluctantly learnt a little mathematics in school this
may seem strange; but to those who have experienced the intoxicating
delight of sudden understanding that mathematics gives,
from time to time, to those who love it, the Pythagorean view will
seem completely natural . . .
Bertrand Russell (1961).
1.1 Notations. The most important thing about mathematics is that the
assertions we make have to have proofs; in other words, we must be able to
produce a logical argument which cannot be attacked or refuted. We will see
many proofs; the next section contains two classics from the ancient Greeks.
The words ‘Theorem’, ‘Proposition’, ‘Lemma’, and ‘Corollary’ all have
the same meaning: a statement which has been proved, and has thereby
become part of the body of mathematics. There are shades of difference: a
theorem is an important statement; a proposition is one which is less important;
a lemma has no importance of its own but is a stepping stone on the
way to a theorem; and a corollary is something which follows easily from a
theorem.
The word ‘Proof’ indicates that the argument establishing a theorem (or other
statement) will follow. The end of the argument is marked by the special symbol
_. If an exercise asks you to ‘prove’, ‘show’, or ‘demonstrate’ some statement,
you are being asked to construct a proof yourself.
A ‘Conjecture’ is a statement which is believed to be true but for which
we do not yet have a proof. Much of what mathematicians do is working to
establish a conjecture (or, since not all conjectures turn out to be true, to refute
one). Another important part of our work is to make conjectures based on our
experience and intuition, for others to prove or disprove. (The great twentiethcentury
Hungarian mathematician Paul Erd˝os said, ‘The aim of life is to prove
and to conjecture.’)
Mathematicians have not always been consistent about applying these terms.
Sometimes it happens that a result which first appeared as a lemma came to be
regarded as more important than the theorem it was originally used to prove.
(See Gauss’ Lemma in Chapter 2 for an example. One result in Chapter 6, Zorn’s
Lemma, is really an axiom!) Also, one of the most famous conjectures (until
recently) was ‘Fermat’s Last Theorem’, which asserted that there cannot exist
natural numbers x, y, z, n with x, y, z > 0 and n > 2 such that xn + yn = zn.
Fermat asserted this theorem and claimed to have a proof, but no proof was
found among his papers and it is now believed that he was mistaken in thinking
he had one; but the name stuck. The conjecture was proved by Wiles in the
1990s, but we still call it ‘Fermat’s Last Theorem’ rather than ‘Wiles’ Theorem’.
Introduction 3
A ‘Definition’ is a precise way of saying what a word means in the mathematical
context. Here is Humpty Dumpty’s view (in the words of Lewis
Carroll):
When I use a word, it means exactly what I want it to mean,
neither more nor less.
In mathematics, we use a lot of words with very precise meanings, often quite
different from their usual meanings. When we introduce a word which is to have
a special meaning, we have to say precisely what that meaning is to be. Usually,
the word being defined is written in italics. For example, you may meet the
definition:
An m × n matrix is an array of numbers set out in m rows and
n columns.
From that point, whenever you come upon the word ‘matrix’, it has this
meaning, and has no relation to the meanings of the word in geology, in medicine,
and in science fiction.
Most of the specialised notation in mathematics will be introduced as we go
along. Because we use so many symbols in our arguments, one alphabet is not
enough, and letters from the Greek alphabet are often called on. Table 1.1 shows
the Greek alphabet.
Other alphabets including Hebrew and Chinese have been used on occasion
too.
Another specialised alphabet is ‘blackboard bold’:
ABCDEFGHIJKLMNOPQRSTUVWXYZ.
This alphabet originated because, in print, mathematicians can use bold type
for special purposes, but bold type is difficult to reproduce on the blackboard
with a piece of chalk. These letters are typically used for number systems:
• N for the natural numbers 1, 2, 3, . . .
• Z for the integers . . . ,−2,−1, 0, 1, 2, . . .
• Q for the rational numbers or fractions such as 3/2
• R for the real numbers, including
√
2 and π
• C for the complex numbers, including i (the square root of −1).
Most of these letters are self-explanatory, but why Z and Q? The German word
for numbers is Zahlen, which gives us the Z. The rational numbers cannot be R,
so remember Q for quotients.
1.2 Proofs. The real answer to our earlier question ‘What is mathematics?’
is: Mathematics is about proofs. A proof is nothing but an argument to
convince you of the truth of some assertion. Mathematical statements require
proofs, which should be completely convincing, though you might have to work
to understand the details. If, after a lot of effort, you are not convinced by an
4 Introduction
Table 1.1 The Greek alphabet
Name Capital Lowercase
alpha A α
beta B β
gamma Γ γ
delta Δ δ
epsilon E _
zeta Z ζ
eta H η
theta Θ θ
iota I ι
kappa K κ
lambda Λ λ
mu M μ
nu N ν
xi Ξ ξ
omicron O o
pi Π π
rho P ρ
sigma Σ σ
tau T τ
upsilon Υ υ
phi Φ φ
chi X χ
psi Ψ ψ
omega Ω ω
argument, then either the author has not made it clear, or the argument is not
correct.
The proofs should ultimately be founded on logic; but we will not be too
precise now about what constitutes a logically valid argument.
Here are two fine examples of proofs, from the time of ancient Greek
mathematics. In each case, the statement is not at all obvious, but the proof
persuades you that it must be true. In each case, the strategy is what we call
‘proof by contradiction’: that is, we show that assuming the opposite of what we
are trying to prove leads to an absurdity or contradiction. Also, in each case, the
proof has an ingenious twist.
The first theorem, probably due to Euclid, states that the series of prime
numbers goes on for ever; there is no largest prime number. (A prime number
is a natural number p greater than 1 which is not divisible by any natural numbers
except for itself and 1. Notice that this definition says that the number 1 is not
a prime number, even though it has no divisors except itself and 1. This makes
sense; we will see the reason later.)
Introduction 5
Theorem 1.1 There are infinitely many prime numbers.
Proof Our strategy is to show that the statement must be true because, if we
assume that it is false, then we are led to an impossibility.
So we suppose that there are only finitely many primes. Let there be n primes,
and let them be p1, p2, . . . , pn. Now consider the number N = p1p2 ・ ・ ・ pn+1. That
is, N is obtained by multiplying together all the prime numbers and adding 1.
Now N must have a prime factor (this is a property of natural numbers which
we will examine further later on). This prime factor must be one of p1, . . . , pn
(since by assumption, these are all the prime numbers). But this is impossible,
since N leaves a remainder of 1 when it is divided by any of p1, . . . , pn.
Thus our assumption that there are only finitely many primes leads to a
contradiction, so this assumption must be false; there must be infinitely many
primes.
The second theorem was proved by Pythagoras (or possibly one of his students).
This theorem is surrounded by legend: supposedly Hipparchos, a disciple
of Pythagoras, was killed (in a shipwreck) by the gods for revealing the disturbing
truth that there are ‘irrational’ numbers.
Theorem 1.2
√
2 is irrational; that is, there is no number x = p/q (where p
and q are whole numbers) such that x2 = 2.
Proof Again the proof is by contradiction. Thus, we assume that there is a
rational number p/q such that (p/q)2 = 2, where p and q are integers. We can
suppose that the fraction p/q is in its lowest terms; that is, p and q have no
common factor.
Now p2 = 2q2. Thus, the number p2 is even, from which it follows that p
must be even. (The square of any odd number is odd: for any odd number has
the form 2m + 1, and its square is (2m + 1)2 = 4m(m + 1) + 1, which is odd.)
Let us write p = 2r. Now our equation becomes 4r2 = 2q2, or 2r2 = q2. Thus,
just as before, q2 is even, and so q is even.
But if p and q are both even, then they have the common factor 2, which
contradicts our assumption that the fraction p/q is in its lowest terms.
Now we look at a few proof techniques, and introduce some new terms.
Proof by contradiction We have just seen two examples of this. In order to
prove a statement P, we assume that P is false, and derive a contradiction from
this assumption.
Proof by contrapositive The contrapositive of the statement ‘if P, then
Q’ is the statement ‘if not Q, then not P’. This is logically equivalent to the
original statement; so we can prove this instead if it is more convenient.
Converse Do not confuse the contrapositive of a statement with its converse.
The converse of ‘if P, then Q’ is ‘if Q, then P’. This is not logically equivalent to
6 Introduction
the original statement. For example, it can be shown that the statement ‘if 2n−1
is prime, then n is prime’ is true; but its converse, ‘if n is prime, then 2n − 1 is
prime’ is false: the number n = 11 is prime, but 211 − 1 = 2047 = 23 × 89.
This example by Lewis Carroll might help you remember the difference
between a statement and its converse.
‘Come, we shall have some fun now!’ thought Alice. ‘I’m glad
they’ve begun asking riddles.–I believe I can guess that,’ she added
aloud.
‘Do you mean that you think you can find out the answer to it?’
said the March Hare.
‘Exactly so,’ said Alice.
‘Then you should say what you mean,’ the March Hare went on.
‘I do,’ Alice hastily replied; ‘at least–at least I mean what I say–
that’s the same thing, you know.’
‘Not the same thing a bit!’ said the Hatter. ‘You might just as
well say that “I see what I eat” is the same thing as “I eat what
I see”!’ ‘You might just as well say,’ added the March Hare, ‘that
“I like what I get” is the same thing as “I get what I like”!’
‘You might just as well say,’ added the Dormouse, who seemed to
be talking in his sleep, ‘that “I breathe when I sleep” is the same
thing as “I sleep when I breathe”!’
‘It is the same thing with you,’ said the Hatter, and here the conversation
dropped, and the party sat silent for a minute, while
Alice thought over all she could remember about ravens and
writing-desks, which wasn’t much.
Counterexample Given a general statement P, to show that P is true it is
necessary to give a general proof; but to show that P is false, we have to give one
specific instance in which it fails. Such an instance is called a counterexample.
In the preceding paragraph, the number n = 11 is a counterexample to the
general statement ‘if n is prime, then 2n − 1 is prime’.
Sufficient condition, ‘if ’ We say that P is a sufficient condition for Q
if the truth of P implies the truth of Q; that is, P implies Q. Another way of
saying the same thing is ‘if P, then Q’, or ‘Q if P’. In symbols, we writeP ⇒Q.
Necessary condition, ‘only if ’ We say that P is a necessary condition
for Q if the truth of P is implied by the truth of Q, that is, Q implies P. (This
is the converse of the statement that P implies Q.) We also say ‘Q only if P’.
Necessary and sufficient condition, ‘if and only if ’ We say that P is
a necessary and sufficient condition for Q if both of the above hold, that
is, each of P and Q implies the other. We also say ‘P if and only if Q’. Note
that there are two things to prove: that P implies Q, and that Q implies P. In
symbols, we write P ⇔Q.
Introduction 7
Proof by induction This is a very important technique for proving things
about natural numbers. We discuss it later in this chapter.
1.3 Axioms. In the proofs in the last section, we used various properties of
numbers: every integer greater than 1 has a prime factor; any number is either
odd or even; and any fraction can be put into its lowest terms by cancelling
common factors. Later on in the book we will examine these assumptions.
The process of examining our hidden assumptions is very important in mathematics.
Each assumption should be proved, but the proof will probably involve
more basic assumptions. There is a sense in which everything can be built from
nothing using only the processes of logic. Usually this is much too long-winded;
so instead we start by making our basic assumptions explicit.
It used to be thought that the basic assumptions of mathematics were true
statements about the real world. Euclid’s geometry was the model for many
centuries. Euclid begins with axioms, which he regarded as ‘self-evident truths’,
and deduced a huge body of theorems from them. But one of his axioms, the
‘axiom of parallels’, is far from self-evident. Mathematicians tried hard to prove
it, but eventually were forced to admit that it was possible to construct a kind of
geometry in which this axiom is false. (This is now referred to as non-Euclidean
geometry.)
Now we regard the axioms as starting points which we choose, depending on
the branch of mathematics we are studying. The theorems we prove will be true
in any system (including any real-world system) which happens to satisfy the
axioms.
One of the advantages of this approach is that, instead of proving theorems
about, say, the integers, we can prove theorems about ‘principal ideal domains’;
as long as the integers satisfy the axioms for principal ideal domains, our theorems
will be true in the integers. This is how we shall justify the assumptions of
the last section about primes and common factors.
It is very important, however, not to bring in any hidden assumptions. For
example, if we are doing geometry, the axioms will probably refer to points and
lines; we must only use properties of points and lines specified in the axioms,
rather than our commonsense view of how points and lines behave.
The German mathematician David Hilbert put it like this:
One must be able at any time to replace ‘points, lines, and planes’
with ‘tables, chairs, and beer mugs’.
Here is a small example. Suppose that we are doing geometry with just the
following three of Euclid’s axioms:
(1) Any two points lie on a unique line.
(2) If the point P does not lie on the line L, then there is exactly one line L_
passing through P and parallel to L.
(3) There exist three non-collinear points.