4 METHODS OF MATHEMATICAL PROOF

4 METHODS OF MATHEMATICAL PROOF

Definition 4.1 (Proof ). A proof is a sequence of statements used to show that a proposition
is true.

Here are some important terminologies associated to proofs.

Definition 4.2 (Theorem). A theorem is a proposition that has been shown to be true.

Definition 4.3 (Lemma). A lemma is a proven proposition that is used as a stepping stone
to a larger result rather than as a statement of interest by itself.

Definition 4.4 (Corollary). A corollary is a proposition that can be established directly

from a theorem that has been proven.

Definition 4.5 (Conjecture). A conjecture is a conclusion or proposition based on incom-

plete information, for which no proof or disproof has yet been found.

Definition 4.6 (Axiom). Axioms are assumptions about a mathematical structure and are
assumed to be true.

Direct Proof: To prove the implication P =⇒ Q, we assume P is true. We then use P ,

axioms, definitions and previously derived results to show that Q is true. A direct proof has
the following form.
Suppose that P is true.
..
.
(argumentation)
..
.
Therefore Q is true.

Example 4.1. Let a, b and c be integers. Prove that if a|b and b|c, then a|c.

Solution. Let a, b and c be integers. Suppose that a|b and b|c. Then, by definition, there
are integers m and n such that b = am and c = bn. Thus, we have that

c = bn = (am)n = a(mn) = at, (4.1)

where t = mn ∈ Z. Therefore, a|c. ■

Example 4.2. Prove that b − a is odd whenever a and b are consecutive perfect squares.

Solution. Suppose that a and b are consecutive perfect squares. Then, a = k 2 and b = (k+1)2
for some k ∈ Z. It follows that

b − a = (k + 1)2 − k 2 = k 2 + 2k + 1 − k 2 = 2k + 1. (4.2)

Hence, b − a is odd. ■

[email protected] 1
 4 METHODS OF MATHEMATICAL PROOF
Example 4.3. Let A, B, M and N be sets. Show that A ⊆ B and M ⊆ N only if
A × M ⊆ B × N.

Solution. Let A, B, M and N be sets. Suppose that A ⊆ B and M ⊆ N . Now, let

(a, m) ∈ A × M . Then, by definition, a ∈ A and m ∈ M . Because A ⊆ B, a ∈ A implies
that a ∈ B. Also, m ∈ M implies that m ∈ N , since M ⊆ N . Consequently, we have that
(a, m) ∈ B × N . Hence, A × M ⊆ B × N . ■
Proof by Contrapositive:
The implication P → Q has been shown to be logically equivalent to its contrapositive
¬Q → ¬P . To prove P =⇒ Q, we can show that ¬Q =⇒ ¬P . A proof by contrapositive
has the following form.
Suppose that ¬Q is true.
..
.
(argumentation)
..
.
Therefore ¬P is true.

Example 4.4. Let n ∈ Z. Prove that if n2 is even, then n is even.

Solution. Let n ∈ Z. Suppose that n is odd. Then, n = 2d + 1 for some d ∈ Z. We get

n2 = (2d + 1)2 = 4d2 + 4d + 1 = 2 2d2 + 2d + 1 = 2r + 1,

(4.3)

where r = 2d2 + 2d ∈ Z. Therefore, n2 is odd. ■

Example 4.5. Let A and B be sets. Prove that A = { } or B = { } whenever A × B = { }.

Solution. Let A and B be sets. Suppose that A ̸= { } and B ̸= { }. Then, there exists
a and b such that a ∈ A and b ∈ B. It follows that (a, b) ∈ A × B, which means that
A × B ̸= { }. ■
Proof by contradiction:
To prove the implication P =⇒ Q, assume P and ¬Q. Using P and ¬Q, derive a contradic-
tion. A proof by contradiction has the following form.
Suppose that P and ¬Q are true.
..
.
(argumentation)
..
.
This is a contradiction. Therefore P implies Q.

Example 4.6. Prove that 2 does not divide n2 − 2 if n is an odd integer.

Solution. Suppose that n is odd and that 2 | n2 − 2. Then, by definition, we have that
n = 2k + 1 and n2 − 2 = 2t for some k, t ∈ Z.It follows that (2k + 1)2 − 2 = 2t, from which
we deduce that 2d = 1, where d = 2k 2 + 2k − t ∈ Z. This is a contradiction because 1 is not
a multiple of 2. Therefore, 2 does not divide n2 − 2 if n is an odd integer. ■

[email protected] 2
 4 METHODS OF MATHEMATICAL PROOF
Example 4.7. Prove that the only consecutive non-negative integers a, b and c that satisfy
a2 + b2 = c2 are 3, 4 and 5.
Solution. Suppose that a, b and c are non-negative consecutive integers such that a2 +b2 = c2
with a ̸= 3. This means that a2 + (a + 1)2 = (a + 2)2 , from which we obtain a2 − 2a − 3 = 0.
The equation a2 − 2a − 3 = 0 yields a = 3 or a = −1. Since a is non-negative, we must have
that a = 3. This contradicts our assumption that a ̸= 3. Therefore, the only consecutive
non-negative integers a, b and c that satisfy a2 + b2 = c2 are 3, 4 and 5. ■
Example 4.8. Let x, y ∈ R. Show that if x + y is an irrational number, then at least one
of x, y is irrational.
Solution. Let x, y ∈ R. Suppose that x + y is an irrational number and that x, y ∈ Q. Since
x, y ∈ Q, we have that x + y ∈ Q. This contradicts our assumption that x + y is an irrational
number. Therefore, if x + y is an irrational number, then at least one of x, y is irrational. ■
Existence proof:
Let P be a propositional function with domain of discourse D. Consider the existential
statement:
There exists x ∈ D such that P (x).
To prove that this statement is true, we must find an element x0 ∈ D for which P (x0 ) is
true. Such a proof is called an existence proof.
Example 4.9. Show that for every rational number q, there exists an integer n ̸= 0 such
that nq is an integer.
a
Solution. Let q ∈ Q. Then, there exists a, b ∈ Z with b ̸= 0 such that q = . Choose
b
b ∈ Z \ {0}. Then, we get bq = a ∈ Z. ■
Example 4.10. Show that there exists a, b ∈ Z such that b | (a2 + 1) but b ∤ (a4 + 1).
Solution. Choose 3, 5 ∈ Z. We find that 32 + 1 = 10, which is divisible by 5. However,
34 + 1 = 82 is not divisible by 5. ■
Proof by cases:
Let P be a propositional function with domain of discourse D. Suppose that
D = D1 ∪ D2 ∪ · · · ∪ Dn ,
where Di ∩ Dj = { } for any i, j ∈ {1, 2, 3, . . . , n} with i ̸= j, and each Di is nonempty. To
prove that P (x) is true for all x ∈ D, we can show that P (xi ) is true for all xi ∈ Di for all
i ∈ {1, 2, 3, . . . , n}.
Example 4.11. Prove that if n ∈ Z, then n2 + 5n + 3 is an odd integer.
Solution. Let n ∈ Z. We consider two cases based on the parity of n.
Case I: Let n be even. Then, n = 2k where k ∈ Z. It follows that
n2 + 5n + 3 = (2k)2 + 5(2k) + 3
= 4k 2 + 10k + 3
= 2(2k 2 + 5k + 1) + 1
= 2d + 1,

[email protected] 3
 4 METHODS OF MATHEMATICAL PROOF
2
where d = 2k + 5k + 1 ∈ Z.
Case II: Let n be odd. Then, n = 2k + 1 where k ∈ Z. Consequently, we get

n2 + 5n + 3 = (2k + 1)2 + 5(2k + 1) + 3

= 4k 2 + 4k + 1 + 10k + 5 + 3
= 4k 2 + 14k + 8 + 1
= 2(2k 2 + 7k + 4) + 1
= 2r + 1,

where r = 2k 2 + 7k + 4 ∈ Z.
In both cases n2 + 5n + 3 is odd. Therefore, if n ∈ Z, then n2 + 5n + 3 is an odd integer. ■
Vacuous proof:
From the definition of the conditional statement, the implication P −→ Q is true whenever
P is false. Thus, the proof of P =⇒ Q is done if we can establish that P is false.
Example 4.12. Show that for all x ∈ R, if x2 + 4x + 4 < 0, then x < 1.
Solution. Let x ∈ R. We deduce that x2 + 4x + 4 = (x + 2)2 ≥ 0. This means that
x2 + 4x + 4 < 0 is false. Hence, if x2 + 4x + 4 < 0, then x < 1. ■
Trivial proof:
From the definition of the conditional statement, the implication P −→ Q is true whenever
Q is true. Thus, the proof of P =⇒ Q is complete if we can establish that Q is true.
Example 4.13. Show that for all x ∈ R, if x ≤ 1 then x2 + 1 ≥ 1.
Solution. Let x ∈ R. Then, we have that x2 ≥ 0. It follows that x2 + 1 ≥ 1. Hence, if x ≤ 1
then x2 + 1 ≥ 1 for all x ∈ R. ■
Consider the terms am , am+1 , mm+2 , ..., an of a sequences. The sum of the terms of this
sequence is
am + am+1 + mm+2 + · · · + an . (4.4)
We can express this sum using the notation
n
X X
aj or aj , (4.5)
j=m m≤j≤n

called the sigma notation. The variable j is called the index of the summation and its
choice is arbitrary. The numbers m and n are called the lower limit and the upper limit,
respectively.
5
X
Example 4.14. What is the value of j 2.
j=1

Solution.
5
X
j 2 = 12 + 22 + 32 + 42 + 52 = 1 + 4 + 9 + 16 + 25 = 55. (4.6)
j=1

[email protected] 4
 4 METHODS OF MATHEMATICAL PROOF
NB: Let k = j − 1. Then j = k + 1 and 0 ≤ k ≤ 4. Consequently,
5
X 4
X
2
j = (k + 1)2 = 12 + 22 + 32 + 42 + 52 = 1 + 4 + 9 + 16 + 25 = 55. (4.7)
j=1 k=0

4 X
X 3
Example 4.15. Evaluate ij.
i=1 j=1

Solution. We obtain
4 X
X 3 4
X 4
X
ij = (i + 2i + 3i) = 6i = 60.
i=1 j=1 i=1 i=1

■
We take note of the following properties:
n
X n
X n
X
(Aj + Bj ) = Aj + Bj (4.8)
j=m j=m j=m

n
X n
X n
X
(Aj − Bj ) = Aj − Bj (4.9)
j=m j=m j=m
n
X n
X
kAj = k Aj , (4.10)
j=m j=m

where k is a constant.
n
" n
#
X X
k = (n − m + 1)k In particular k = nk , (4.11)
j=m j=1

where k is a constant.
n
X k
X n
X
Ai = Ai + Ai (4.12)
i=m i=m i=k+1

where m < k < n.

The principle of mathematical induction:

Let P be a propositional function with domain of discourse Z. Suppose that the following
two conditions hold:
(1) There exists n0 ∈ Z such that P (n0 ) is true.
(2) If P (k) is true for some k ≥ n0 , then P (k + 1) is true.
Then, P (n) is true for all integers n ≥ n0 .

In (2), the assumption “P (k) is true for some k ≥ n0 ” is known as the induction
hypothesis.

[email protected] 5
 4 METHODS OF MATHEMATICAL PROOF
n
Example 4.16. Prove that n! > 2 for all n ≥ 4.
Solution. Let P (n) be n! > 2n .
Now, we notice that 4! = 24 > 16 = 24 . Thus, P (4) is true.
Next, suppose that P (k) is true for some k ≥ 4. That is, k! > 2k . Since k ≥ 4, we deduce
that k + 1 ≥ 5 > 2. The inequality k + 1 > 2, together with the induction hypothesis k! > 2k ,
produces
(k + 1)! = (k + 1)k! > 2 × 2k = 2k+1 .
Hence P (k + 1) is true. Therefore, n! > 2n for all n ≥ 4. ■
Example 4.17. Prove that, for any integer n ≥ 1
n(n + 1)(2n + 1)
12 + 22 + 32 + · · · + n2 = .
6
Solution. Let P (n) be
n
X n(n + 1)(2n + 1)
i2 = .
i=1
6
We observe that
1
X 1(1 + 1)(2 · 1 + 1) 1·2·3
i2 = 12 = 1 and = = 1.
i=1
6 6

Consequently, we get
1
X 1(1 + 1)(2 · 1 + 1)
i2 = ,
i=1
6
which means that P (1) is true.
Next, suppose that P (k) is true for some k ≥ 1. That is,
k
X k(k + 1)(2k + 1)
i2 = .
i=1
6

Now, we get
k+1
X k
X
2
i = i2 + (k + 1)2
i=1 i=1
k(k + 1)(2k + 1)
= + (k + 1)2
6
(k + 1)
= [k(2k + 1) + 6(k + 1)]
6
(k + 1)
= (2k 2 + 7k + 6)
6
(k + 1)(k + 2)(2k + 3)
= ,
6
n(n + 1)(2n + 1)
which implies that P (k + 1) is true. Therefore, 12 + 22 + 32 + · · · + n2 =
6
for all integers n ≥ 1. ■

[email protected] 6
 4 METHODS OF MATHEMATICAL PROOF
Example 4.18. Prove that if n ∈ N, then 8n − 3n is divisible by 5.
Solution. Let P (n) be 5 | 8n − 3n .
Let n = 1. Then, 8n − 3n = 81 − 31 = 5, which is divisible by 5. Hence, P (1) is true.
Next, suppose that P (k) is true for some k ∈ N. That is, 5 | 8k − 3k . We have that

8k+1 − 3k+1 = 8(8k ) − 3(3k )

= 8(8k ) − 8(3k ) + 8(3k ) − 3(3k )
= 8(8k − 3k ) + 5(3k )
= 8(5m) + 5(3k ), for some m ∈ Z
= 5d,

where d = 8m + 3k ∈ Z. This means that 5 | 8k+1 − 3k+1 , and so P (k + 1) is true. Therefore,

if n ∈ N, then 8n − 3n is divisible by 5. ■

EXERCISE FOUR

1. Let a and b be integers. Prove that if a|b, then an |bn for all positive integers n.

2. Let a, b, c, m and n be integers. Prove that if a|b and a|c, then a|(bm + cn)

3. Prove that if m + n and n + p are even integers, where m, n, and p are integers, then
m + p is even.

4. Let a, b, c, n ∈ Z. Prove that n | (a − c), if n | (a − b) and n | (b − c).

5. Prove that if n is odd, then 3n is odd.

6. Prove that if n is an integer and n3 + 5 is odd, then n is even.

7. If m, n ∈ Z such that mn2 is even, then at least one of m, n is even.

8. Let n ∈ Z. Prove that if n2 is even, then n is even.

9. There is no rational number x such that x2 = 2.

1
10. Suppose a is an irrational number. Prove that a
is an irrational number.

11. Suppose that a is a rational number and that b is an irrational number. Prove that
a + b is an irrational number.

12. Prove that for any integer n ≥ 1 the sum of the odd integers from 1 to 2n − 1 is n2 .

13. Prove or disprove:

(a) If 3 divides the product mn of positive integers m and n, then 3 divides m or 3

divides n.
(b) If 4 divides the product mn of positive integers m and n, then 4 divides m or 4
divides n.

14. Let a ≥ 0 be a real number. Show that if for every ε ∈ R+ we have 0 ≤ a < ε, then
a = 0.

[email protected] 7
 4 METHODS OF MATHEMATICAL PROOF
15. Prove that, for any integer n ≥ 1

n(n + 1)
1 + 2 + 3 + ··· + n = .
2

16. Show that n2 < 2n for all n ∈ N with n > 4.

17. Prove that, for any integer ≥ 1

3 3 3 n2 (n + 1)2
3
1 + 2 + 3 + ··· + n = .
4

18. Show that n(n + 1) is divisible by 2 for all n ∈ N.

19. Show that n(n2 − 1) is divisible by 6 for all n ∈ N.

20. Determine whether the following assertions are true or false. If true, prove the result,
and if false, give a counterexample.

(a) 2x2 + 3y 2 > 0 for all real numbers x and y.

(b) If a is an even number then a 21 is an even number.
(c) For every real number x, there exists a real number y such that xy = 1.
(d) x2 ≥ x for all real number x.
(e) n2 ≥ n for all positive integers n.
(f) If n is an even integer then n2 + 3n is an even integer.
(g) There is an integer n ̸= 0 such that nq is an integer for every rational number q.

[email protected] 8