Proof Techniques

Discrete Mathematics
(Proof Techniques)
Pramod Ganapathi
Department of Computer Science
State University of New York at Stony Brook
March 21, 2022

What is a proof?
Definition
A proof is a method for establishing the truth of a statement.
Rigor Truth type Field Truth teller
0 Word of God Religion God/Priests
1 Authoritative truth Business/School Boss/Teacher
2 Legal truth Judiciary Law/Judge/Law makers
3 Philosophical truth Philosophy Plausible argument
4 Scientific truth Physical sciences Experiments/Observations
5 Statistical truth Statistics Data sampling
6 Mathematical truth Mathematics Logical deduction

What is a mathematical proof?
Definition
A mathematical proof is a verification for establishing the truth
of a proposition by a chain of logical deductions from a set of
axioms
Concepts
1. Proposition
Covered in sufficient depth in logic
2. Axiom
An axiom is a proposition that is assumed to be true
Example: For mathematical quantities a and b, if a = b, then
b = a
3. Logical deduction
We call this process – the axiomatic method
We will cover several proof techniques in this chapter

Why care for mathematical proofs?
The current world ceases to function without math proofs
(My belief) Reduction tree showing subjects that possibly could
be expressed or understood in terms of other subjects
Humanities
Psychology
Biology
Chemistry
Physics
Mathematics CS

Methods of mathematical proof
Statements Method of proof
Proving existential statements Constructive proof
(Disproving universal statements) Non-constructive proof
Proving universal statements Direct proof
(Disproving existential statements) Proof by mathematical induction
Well-ordering principle
Proof by exhaustion
Proof by cases
Proof by contradiction
Proof by contraposition
Computer-aided proofs

Introduction to number theory
Definition
Number theory is the branch of mathematics that deals with
the study of integers
Numbers Set
Natural numbers (N) {1, 2, 3, . . .}
Whole numbers (W) {0, 1, 2, . . .}
Integers (Z) {0, ±1, ±2, ±3, . . .}
Even numbers (E) {0, ±2, ±4, ±6, . . .}
Odd numbers (O) {±1, ±3, ±5, ±7, . . .}
Prime numbers (P) {2, 3, 5, 7, 11, . . .}
Composite numbers (C) {Natural numbers (> 1) that are not prime}
Rational numbers (Q) {Ratio of integers with non-zero denominator}
Real numbers (R) {Numbers with infinite decimal representation}
Irrational numbers (I) {Real numbers that are not rational}
Complex numbers (S) {real + i · real}

Even and odd numbers
Definitions
An integer n is even iff n equals twice some integer;
Formally, for any integer n,
n is even ⇔ n = 2k for some integer k
An integer n is odd iff n equals twice some integer plus 1;
Formally, for any integer n,
n is odd ⇔ n = 2k + 1 for some integer k
Examples
Even numbers:
0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, . . .
Odd numbers:
1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, . . .

Rational and irrational numbers
Definitions
A real number r is rational iff it can be expressed as a ratio of
two integers with a nonzero denominator;
Formally, if r is a real number, then
r is rational ⇔ ∃ integers a, b such that r =
a
b
and b 6= 0
A real number r is irrational iff it is not rational
Examples
Rational numbers:
10, −56.47, 10/13, 0, −17/9, 0.121212 . . . , −91, . . .
Irrational numbers:
√
2,
√
3,
√
2
√
2
, π, φ, e, π2, e2, 21/3, log2 3, . . .
Open problems:
It’s not known if π + e, πe, π/e, πe, π
√
2, and ln π are irrational

Divisibility
Definitions
If n and d are integers, then n is divisible by d, denoted by d|n,
iff n equals d times some integer and d 6= 0;
Formally, if n and d are integers
d|n ⇔ ∃ integer k such that n = dk and d 6= 0
Instead of “n is divisible by d,” we can say:
n is a multiple of d, or
d is a factor of n, or
d is a divisor of n, or
d divides n (denoted by d|n)
Note: d|n is different from d/n
Examples
Divides: 1|1, 10|10, 2|4, 3|24, 7| − 14, . . .
Does not divide: 2 - 1, 10 - 1, 10 - 2, 7 - 10, 10 - 7, 10 - −7, . . .

Quotient-Remainder theorem
Theorem
Given any integer n and a positive integer d, there exists an
integer q and a whole number r such that
n = qd + r and r ∈ [0, d − 1]
Examples
Let n = 6 and d ∈ [1, 7]
Num. (n) Divisor (d) Theorem Quotient (q) Rem. (r)
6 1 6 = 6 × 1 + 0 6 0
6 2 6 = 3 × 2 + 0 3 0
6 3 6 = 2 × 3 + 0 2 0
6 4 6 = 1 × 4 + 2 1 2
6 5 6 = 1 × 5 + 1 1 1
6 6 6 = 1 × 6 + 0 1 0
6 7 6 = 0 × 7 + 6 0 6

Prime numbers
Num. Factorization Prime?
2 2 = 1 × 2 = 2 × 1 3
3 3 = 1 × 3 = 3 × 1 3
4 4 = 1 × 4 = 4 × 1 = 2 × 2 7
5 5 = 1 × 5 = 5 × 1 3
6 6 = 1 × 6 = 6 × 1 = 2 × 3 = 3 × 2 7
7 7 = 1 × 7 = 7 × 1 3
8 8 = 1 × 8 = 8 × 1 = 2 × 4 = 4 × 2 7
9 9 = 1 × 9 = 9 × 1 = 3 × 3 7
10 10 = 1 × 10 = 10 × 1 = 2 × 5 = 5 × 2 7
11 11 = 1 × 11 = 11 × 1 3
12 12 = 1 × 12 = 12 × 1 = 2 × 6 = 6 × 2 = 3 × 4 = 4 × 3 7
13 13 = 1 × 13 = 13 × 1 3
14 14 = 1 × 14 = 14 × 1 = 2 × 7 = 7 × 2 7
15 15 = 1 × 15 = 15 × 1 = 3 × 5 = 5 × 3 7
16 16 = 1 × 16 = 16 × 1 = 2 × 8 = 8 × 2 = 4 × 4 7
17 17 = 1 × 17 = 17 × 1 3

Prime numbers
Definitions
A natural number n is prime iff n > 1 and it has exactly two
positive divisors: 1 and n
A natural number n is composite iff n > 1 and it has at least
three positive divisors, two of which are 1 and n
A natural number n is a perfect square iff it has an odd number
of divisors
A natural number n is not a perfect square iff it has an even
number of divisors
Examples
Perfect squares: 1, 4, 9, 16, 25, . . .
Not perfect squares: 2, 3, 5, 6, 7, 8, 10, . . .

Prime numbers
Definitions
A natural number n is prime iff n > 1 and for all natural
numbers r and s, if n = rs, then either r or s equals n;
Formally, for each natural number n with n > 1,
n is prime ⇔ ∀ natural numbers r and s, if n = rs
then n = r or n = s
A natural number n is composite iff n > 1 and n = rs for some
natural numbers r and s with 1 < r < n and 1 < s < n;
Formally, for each natural number n with n > 1,
n is composite ⇔ ∃ natural numbers r and s, if n = rs
and 1 < r < n and 1 < s < n

Unique prime factorization of natural numbers
n Unique prime
factorization
2 2
3 3
4 22
5 5
6 2 × 3
7 7
8 23
9 32
10 2 × 5
11 11
12 22 × 3
13 13
14 2 × 7
15 3 × 5
n Unique prime
factorization
16 24
17 17
18 2 × 32
19 19
20 22 × 5
21 3 × 7
22 2 × 11
23 23
24 23 × 3
25 52
26 2 × 13
27 33
28 22 × 7
29 29
n Unique prime
factorization
30 2 × 3 × 5
31 31
32 25
33 3 × 11
34 2 × 17
35 5 × 7
36 22 × 32
37 37
38 2 × 19
39 3 × 13
40 23 × 5
41 41
42 2 × 3 × 7
43 43
What is the pattern?

Unique prime factorization of natural numbers
Definition
Any natural number n > 1 can be uniquely represented as a
product of as follows:
n = pe1
1 × pe2
2 × · · · × pek
k
such that p1 < p2 < · · · < pk are primes in [2, n], e1, e2, . . . , ek
are whole number exponents, and k is a natural number.
The theorem is also called fundamental theorem of arithmetic
The form is called standard factored form

Some terms
Definitions
Absolute value of real number x, denoted by |x| is
|x| =
(
x if x ≥ 0
−x if x < 0
Triangle inequality. For all real numbers x and y,
|x + y| ≤ |x| + |y|
Floor of a real number x, denoted by bxc is
bxc = unique integer n such that n ≤ x < n + 1
bxc = n ⇔ n ≤ x < n + 1
Ceiling of a real number x, denoted by dxe is
dxe = unique integer n such that n − 1 < x ≤ n
dxe = n ⇔ n − 1 < x ≤ n

Some terms
Definitions
Given an integer n and a natural number d,
n div d = integer quotient obtained when n is divided by d,
n mod d = whole number remainder obtained when n is divided
by d.
Symbolically,
n div d = q and n mod d = r ⇔ n = dq + r
where q and r are integers and 0 ≤ r < d.

Properties of a proof
Properties
Concise (not unnecessarily long)
Clear (not ambiguous)
Complete (no missing intermediate steps)
Logical (every statement logically follows)
Rigorous (uses mathematical expressions)
Convincing (does not raise questions)
The way a proof is presented might be different from the way
the proof is discovered.

Even + odd = odd
Proposition
Sum of an even integer and an odd integer is odd.

Even + odd = odd
Proposition
Sum of an even integer and an odd integer is odd.
Proof
Suppose a is even and b is odd. Then
a + b
= (2m) + b (defn. of even, a = 2m for integer m)
= (2m) + (2n + 1) (defn. of odd, b = 2n + 1 for integer n)
= 2(m + n) + 1 (taking 2 as common factor)
= 2p + 1 (p = m + n and addition is closed on integers)
= odd (defn. of odd)

Problems for practice
Prove the following propositions:
Even + even = even
Even + odd = odd
Odd + odd = even
Even × integer = even
Odd × odd = odd

n is odd ⇒ n2
is odd
Proposition
The square of an odd integer is odd.

n is odd ⇒ n2
is odd
Proposition
The square of an odd integer is odd.
Proof
Prove: If n is odd, then n2 is odd.
n is odd
=⇒ n = (2k + 1) (defn. of odd, k is an integer)
=⇒ n2 = (2k + 1)2 (squaring on both sides)
=⇒ n2 = 4k2 + 4k + 1 (expanding the binomial)
=⇒ n2 = 2(2k2 + 2k) + 1 (factoring 2 from first two terms)
=⇒ n2 = 2j + 1 (let j = 2k2 + 2k)
(j is an integer as mult. and add. are closed on integers)
=⇒ n2 is odd (defn. of odd)

Odd = difference of squares
Proposition
Every odd integer is equal to the difference between the squares
of two integers

Proposition
of two integers
Workout
Write a formal statement.
∀ integer k, ∃ integers m, n such that
(2k + 1) = m2 − n2.
Try out a few examples.
1 = 12
− 02
− 1 = 02
− (−1)2
3 = 22
− 12
− 3 = (−1)2
− (−2)2
5 = 32
− 22
− 5 = (−2)2
− (−3)2
7 = 42
− 32
− 7 = (−3)2
− (−4)2
Find a pattern.
(k + 1)2 − k2 = (k2 + 2k + 1) − k2 = 2k + 1 = odd

Proposition
of two integers.
Proof
Any odd integer can be written as (2k + 1) for some integer k.
We rewrite the expression as follows.
2k + 1
= (k2 + 2k + 1) − k2 (adding and subtracting k2)
= (k + 1)2 − k2 (write the first term as sum)
= m2 − n2 (set m = k + 1 and n = k)
The term m is an integer as addition is closed on integers.
So, every odd integer can be written as the difference between
two squares.

k
k2
cells (k + 1)2
cells

Summation
Proposition
1 + 2 + 3 + · · · + n = n(n + 1)/2.

Summation
Proposition
1 + 2 + 3 + · · · + n = n(n + 1)/2.
Proof
Formal statement. ∀ natural number n, prove that
1 + 2 + 3 + · · · + n = n(n + 1)/2.
S = 1 + 2 + 3 + · · · + n
=⇒ S = n + (n − 1) + (n − 2) + · · · + 1
(addition on integers is commutative)
=⇒ 2S = (n + 1) + (n + 1) + (n + 1) + · · · + (n + 1)
| {z }
n terms
(adding the previous two equations)
=⇒ 2S = n(n + 1) (simplifying)
=⇒ S = n(n + 1)/2 (divide both sides by 2)

2999
+ 1
Proposition
2999 + 1 is prime.

2999
+ 1
Proposition
2999 + 1 is prime.
Workout
Trying out a few examples is not possible here.
When is a number prime?
A number that is not composite is prime.
When is a number composite?
A number is composite if we can factorize it.
How do you check if a number can be factorized?
Check whether the number satisfies an algebraic formula that
can be factored.
It seems like the given number can be represented as a3 + b3.

2999
+ 1
Proposition
2999 + 1 is prime.
Solution
False! 2999 + 1 is composite.
2999 + 1
= (2333)3 + 13 (terms represented as cubes)
= a3 + b3 (set a = 2333, b = 1)
= (a + b)(a2 − ab + b2) (factorize a3 + b3)
= (2333 + 1)(2666 − 2333 + 1) (substituting a and b values)
= composite

n2
+ 3n + 2
Proposition
There is a natural number n such that n2 + 3n + 2 is prime.

n2
+ 3n + 2
Proposition
Workout
∃ natural number n such that n2 + 3n + 2 is prime.
12
+ 3(1) + 2 = 6 composite
22
+ 3(2) + 2 = 12 composite
32
+ 3(3) + 2 = 20 composite
42
+ 3(4) + 2 = 30 composite
52
+ 3(5) + 2 = 42 composite
Find a pattern.
It seems like n2 + 3n + 2 is always composite.

n2
+ 3n + 2
Proposition
Solution
False!
Proving that the given statement is false is equivalent to proving
that its negation is true.
Negation. ∀ natural number n, n2 + 3n + 2 is composite.
n2 + 3n + 2
= n2 + n + 2n + 2 (split 3n)
= n(n + 1) + 2(n + 1) (taking common factors)
= (n + 1)(n + 2) (distributive law)
= composite (n + 1 > 1 and n + 2 > 1)

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x = 7.

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x = 7.
Proof
Substitute x = 7 in the expression to get 73 −7(72)+7−7 = 0.
As x satisfies the equation, x = 7.

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x = 7.
Proof
Substitute x = 7 in the expression to get 73 −7(72)+7−7 = 0.
As x satisfies the equation, x = 7.
Incorrect! What’s wrong?

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x = 7.
Proof
False!
A polynomial equation of degree n has n roots.
So, the polynomial equation x3 − 7x2 + x − 7 = 0 has 3 roots.
We factorize the expression.
x3 − 7x2 + x − 7
= x2(x − 7) + (x − 7) (taking x2 factor from first two terms)
= (x − 7)(x2 + 1) (taking (x − 7) factor)
= (x − 7)(x + i)(x − i) (factorizing (x2 + 1))
(this is because (x + i)(x − i) = (x2 − i2) = (x2 + 1))
So, the three roots to the equation x3 − 7x2 + x − 7 = 0 are
x = 7, x = −
√
−1, and x =
√
−1.

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x = 7.
Proof (continued)
Exactly one of the three roots is x = 7. Hence, we have
x = 7 =⇒ x3
− 7x2
+ x − 7 = 0
x3
− 7x2
+ x − 7 = 0 6
=⇒ x = 7

Polynomial root
Proposition
If x is a real number and x3 − 7x2 + x − 7 = 0, then x = 7.

Polynomial root
Proposition
If x is a real number and x3 − 7x2 + x − 7 = 0, then x = 7.
Proof
We factorize the expression.
x3 − 7x2 + x − 7
= x2(x − 7) + (x − 7) (taking x2 factor from first two terms)
= (x − 7)(x2 + 1) (taking (x − 7) factor)
= (x − 7)(x + i)(x − i) (factorizing (x2 + 1))
(this is because (x + i)(x − i) = (x2 − i2) = (x2 + 1))
So, the three roots to the equation x3 − 7x2 + x − 7 = 0 are
x = 7, x = −
√
−1, and x =
√
−1.
As x has to be a real number, x = 7.

a = b
Proposition
For all real numbers a and b, if a2 = b2, then a = b.

a = b
Proposition
Solution
False! Counterexample: a = 1 and b = −1.
In this example, a2 = b2 but a 6= b.

a = b
Proposition
Solution
Proposition
For all nonzero integers a and b, if a|b and b|a, then a = b.

a = b
Proposition
Solution
Proposition
For all nonzero integers a and b, if a|b and b|a, then a = b.
Solution
In this example, a|b and b|a, however, a 6= b.

2n
+ 1
Proposition
2n + 1 is prime for any natural number n.

2n
+ 1
Proposition
Workout
∀ natural number n, 2n + 1 is prime.
21
+ 1 = 3 prime
22
+ 1 = 5 prime
23
+ 1 = 9 = 32
composite
Find a pattern.
2n + 1 can be either prime or composite.

2n
+ 1
Proposition
Workout
∀ natural number n, 2n + 1 is prime.
21
+ 1 = 3 prime
22
+ 1 = 5 prime
23
+ 1 = 9 = 32
composite
Find a pattern.
2n + 1 can be either prime or composite.
Solution
False! Counterexample: n = 3
When n = 3, then 2n + 1 = 23 + 1 = 9 = 32 is composite.

n2
+ n + 41
Proposition
n2 + n + 41 is prime for any whole number n.

n2
+ n + 41
Proposition
Workout
∀ whole number n, n2 + n + 41 is prime.
02
+ 0 + 41 = 41 prime
12
+ 1 + 41 = 43 prime
22
+ 2 + 41 = 47 prime
32
+ 3 + 41 = 53 prime
42
+ 4 + 41 = 61 prime
52
+ 5 + 41 = 71 prime
Find a pattern.
It seems like n2 + n + 41 is always prime.

n2
+ n + 41
Proposition
Solution
False!
Formal statement. ∀ whole numbers n, n2 + n + 41 is prime.
Counterexample: 41.
(412 + 41 + 41 = 41(41 + 1 + 1) = 41 × 43)
Another counterexample: 40.
(402 +40+41 = 40(40+1)+41 = 40×41+41 = 41(40+1) =
41 × 41)

x/(y + z) + y/(x + z) + z/(x + y)
Proposition
x
y+z + y
x+z + z
x+y = 4 has no positive integer solutions.

x/(y + z) + y/(x + z) + z/(x + y)
Proposition
x
y+z + y
x+z + z
Workout
∀ x, y, z ∈ N, x/(y + z) + y/(x + z) + z/(x + y) 6= 4.
(x, y, z) x/(y + z) + y/(x + z) + z/(x + y) = 4 ?
(1, 1, 1) 1/2 + 1/2 + 1/2 = 1.5 6= 4
(1, 2, 1) 1/3 + 2/2 + 1/3 = 1.666 · · · 6= 4
(1, 2, 3) 1/5 + 2/4 + 3/3 = 1.7 6= 4
(1, 10, 100) 1/110 + 10/101 + 100/11 = 9.199 · · · 6= 4
Find a pattern.
It seems like there are no +ve integers satisfying the property.

x/(y + z) + y/(x + z) + z/(x + y)
Proposition
x
y+z + y
x+z + z
Solution
False!
Counterexample:
x = 15447680210874616644195131501991983748566432566
9565431700026634898253202035277999
y = 36875131794129999827197811565225474825492979968
971970996283137471637224634055579
z = 37361267792869725786125260237139015281653755816
1613618621437993378423467772036

121111111111111111111111111111111111111111
Proposition
For whole numbers n, 12 11 · · · 1
| {z }
n terms
is composite.

121111111111111111111111111111111111111111
Proposition
For whole numbers n, 12 11 · · · 1
| {z }
n terms
is composite.
Workout
(n, Number) Factorization
(0, 12) 3 × 4
(1, 121) 11 × 11
(2, 1211) 7 × 173
(3, 12111) 33 × 367
(4, 121111) 281 × 431
(5, 1211111) 253 × 4787
Find a pattern.
It seems like the sequence of numbers is composite.

121111111111111111111111111111111111111111
Proposition
For whole numbers n ≥ 0, 12 11 · · · 1
| {z }
n terms
is composite.
Solution
False!
Smallest counterexample: n = 136.
12,1111111111, 1111111111, 1111111111, 1111111111,
1111111111, 1111111111, 1111111111, 1111111111,
1111111111, 1111111111, 1111111111, 1111111111,
1111111111, 111111 is prime.

n2
is odd ⇒ n is odd
Proposition
If n2 is odd, then n is odd.

n2
is odd ⇒ n is odd
Proposition
If n2 is odd, then n is odd.
Proof
Seems very difficult to prove directly.
Contraposition: If n is even, then n2 is even.
n is even
=⇒ n = 2k (defn. of even, k is an integer)
=⇒ n2 = (2k)2 (squaring on both sides)
=⇒ n2 = 4k2 (simplifying)
=⇒ n2 = 2(2k2) (factoring 2)
=⇒ n2 = 2j (let j = 2k2)
(j is an integer as mult. is closed on integers)
=⇒ n2 is even (defn. of even)

n is odd ⇔ n2
is odd
Proposition
The square of an integer is odd if and only if the integer itself
is odd.

n is odd ⇔ n2
is odd
Proposition
is odd.
Workout
∀ integer n, n2 is odd ⇔ n is odd.
Odd numbers Even numbers
(1, 1) (0, 0)
(3, 9) (2, 4)
(5, 25) (4, 16)
(7, 49) (6, 36)
Pattern. It seems that the proposition is true.

n is odd ⇔ n2
is odd
Proposition
is odd.
Proof
There are two parts in the proof.
1. Prove that if n is odd, then n2 is odd.
Direct proof
2. Prove that if n2 is odd, then n is odd.
Proof by contraposition

n is odd ⇔ n2
is odd
Corollary
Prove that the fourth power of an integer is odd if and only if
the integer itself is odd.

n is odd ⇔ n2
is odd
Corollary
Proof
We have
n is odd ⇔ n2 is odd (previous theorem)
=⇒ n2 is odd ⇔ n4 is odd (previous theorem used on n2)
=⇒ n is odd ⇔ n4 is odd (transitivity of biconditional)

n is odd ⇔ n2
is odd
Corollary
Proof
We have
n is odd ⇔ n2 is odd (previous theorem)
=⇒ n2 is odd ⇔ n4 is odd (previous theorem used on n2)
=⇒ n is odd ⇔ n4 is odd (transitivity of biconditional)
Problem
Suppose k is a whole number. Prove that an integer n is odd
if and only if n2k
is odd.

n2
is even =⇒ n is even
Proposition
For all integers n, if n2 is even, then n is even.

n2
Proposition
Proof
Contrapositive. For all integers, if n is odd, then n2 is odd.
n = 2k + 1 (definition of odd number)
=⇒ n2 = (2k + 1)2 (squaring both sides)
=⇒ n2 = 4k2 + 4k + 1 (expand)
=⇒ n2 = 2(2k2 + 2k) + 1 (taking 2 out from two terms)
=⇒ n2 = 2m + 1 (set m = 2k2 + 2k)
(m is an integer as multiplication is closed on integers)
=⇒ n2 = odd (definition of odd number)
Hence, the proposition is true.

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x 6= 10.

Polynomial root
Proposition
If x3 − 7x2 + x − 7 = 0, then x 6= 10.
Proof
Contrapositive. If x = 10, then x3 − 7x2 + x − 7 6= 0
Substitute x = 10 in the expression.
We get 103 −7(102)+10−7 = 1000−700+10−7 = 303 6= 0.
That is, x = 10 does not satisfy x3 −7x2 +x−7 = 0 equation.
Hence, the contraposition is correct which implies that the orig-
inal statement is correct.

n - ab =⇒ n - a and n - b
Proposition
Let a, b, n ∈ Z. If n - ab, then n - a and n - b.

n - ab =⇒ n - a and n - b
Proposition
Let a, b, n ∈ Z. If n - ab, then n - a and n - b.
Proof
Contrapositive. Let a, b, n ∈ Z. If n|a or n|b, then n|ab.
n|a
=⇒ a = nc (for some c ∈ Z)
=⇒ ab = (nc)b = n(cb) (multiply by b)
=⇒ n|ab (definition of divisibility)
n|b
=⇒ b = nd (for some d ∈ Z)
=⇒ ab = a(nd) = n(ad) (multiply by a)
=⇒ n|ab (definition of divisibility)

n2
− 6n + 5 is even =⇒ n is odd
Proposition
Let n ∈ Z. If n2 − 6n + 5 is even, then n is odd.

n2
− 6n + 5 is even =⇒ n is odd
Proposition
Let n ∈ Z. If n2 − 6n + 5 is even, then n is odd.
Proof
Contrapositive. If n is even, then n2 − 6n + 5 is odd.
n is even
=⇒ n = 2a for some integer a (defn. of even)
=⇒ n2 − 6n + 5 = (2a)2 − 6(2a) + 5 (substitute n = 2a)
=⇒ n2 − 6n + 5 = 2(2a2) − 2(6a) + 2(2) + 1 (simplify)
=⇒ n2 − 6n + 5 = 2(2a2 − 6a + 2) + 1 (take 2 common)
=⇒ n2 − 6n + 5 is odd (defn. of odd)

xy > 9 =⇒ x > 3 or y > 3
Proposition
For reals x and y, if xy > 9, then either x > 3 or y > 3.

xy > 9 =⇒ x > 3 or y > 3
Proposition
Proof
Contrapositive. If x ≤ 3 and y ≤ 3, then xy ≤ 9.
Suppose x ≤ 3 and y ≤ 3.
=⇒ xy ≤ 9 (multiply the two inequalities)

xy > 9 =⇒ x > 3 or y > 3
Proposition
Proof
Contrapositive. If x ≤ 3 and y ≤ 3, then xy ≤ 9.
Suppose x ≤ 3 and y ≤ 3.
=⇒ xy ≤ 9 (multiply the two inequalities)
Incorrect! Why?

Irrationalirrational
can be rational
Proposition
An irrational raised to an irrational power may be rational.

Irrationalirrational
can be rational
Proposition
An irrational raised to an irrational power may be rational.
Proof
We make use of the fact that
√
2 is irrational.
Let x =
√
2
√
2
. Number x is either rational or irrational.
Case 1. If x is rational, then the proposition is true.
Irrational Irrational Rational
√
2
√
2
√
2
√
2
= x = rational
Case 2. If x is irrational, then the proposition is true.
Irrational Irrational Rational
x
√
2 x
√
2 =

√
2
√
2
√
2
=
√
2
√
2·
√
2
=
√
2
2
= 2

n2
Proposition
Proof
Negation. Suppose there is an integer n such that
n2 is even but n is odd.
n = 2k + 1 (definition of odd number)
=⇒ n2 = (2k + 1)2 (squaring both sides)
=⇒ n2 = 4k2 + 4k + 1 (expand)
=⇒ n2 = 2(2k2 + 2k) + 1 (taking 2 out from two terms)
=⇒ n2 = 2m + 1 (set m = 2k2 + 2k)
(m is an integer as multiplication is closed on integers)
=⇒ n2 = odd (definition of odd number)
Contradiction! Hence, the proposition is true.

Greatest integer
Proposition
There is no greatest integer.

Greatest integer
Proposition
There is no greatest integer.
Proof
Negation. Suppose there is a greatest integer N.
Then N ≥ n for every integer n.
Let M = N + 1.
M is an integer since addition is closed on integers.
M N since M = N + 1.
M is an integer that is greater than N.
So, N is not the greatest integer.

√
2 is irrational
Proposition
√
2 is irrational.

√
2 is irrational
Proposition
√
2 is irrational.
Proof
Suppose
√
2 is the simplest rational.
=⇒
√
2 = m/n (m, n have no common factors, n 6= 0)
=⇒ m2 = 2n2 (squaring and simplifying)
=⇒ m2 = even (definition of even)
=⇒ m = even (why?)
=⇒ m = 2k for some integer k (definition of even)
=⇒ (2k)2 = 2n2 (substitute m)
=⇒ n2 = 2k2 (simplify)
=⇒ n2 = even (definition of even)
=⇒ n = even (why?)
=⇒ m, n are even (previous results)
=⇒ m, n have a common factor of 2 (definition of even)

If p|n, then p - (n + 1).
Proposition
For any integer n and any prime p, if p|n, then p - (n + 1).

If p|n, then p - (n + 1).
Proposition
For any integer n and any prime p, if p|n, then p - (n + 1).
Proof
Negation. Suppose there exists integer n and prime p such that
p|n and p|(n + 1).
p|n implies pr = n for some integer r
p|(n + 1) implies ps = n + 1 for some integer s
Eliminate n to get:
1 = (n + 1) − n = ps − pr = p(s − r)
Hence, p|1, from the definition of divisibility.
As p|1, we have p ≤ 1. (why?)
As p is prime, p 1.

#Primes is infinite
Proposition
The set of prime numbers is infinite.

#Primes is infinite
Proposition
Proof
Negation. Assume that there are only finite number of primes.
Let the set of primes be {p1, p2, . . . , pn}
such that (p1 = 2) (p2 = 3) · · · pn.
Consider the number N = p1p2p3 . . . pn + 1. Clearly, N 1.

#Primes is infinite
Proposition
Proof
such that (p1 = 2) (p2 = 3) · · · pn.
(i) There is a prime that divides N.
Use unique prime factorization theorem.

#Primes is infinite
Proposition
Proof
such that (p1 = 2) (p2 = 3) · · · pn.
(i) There is a prime that divides N.
Use unique prime factorization theorem.
(ii) No prime divides N.
For all i ∈ [1, n], pi does not divide N as it leaves a remainder
of 1 when it divides N.
So, p1 6 | N, p2 6 | N, . . ., pn 6 | N.

Average
Proposition
If a1, a2, . . . , an are n real numbers for natural number n, then
at least one of these n numbers is greater than or equal to the
average of those n numbers.

Average
Proposition
Proof
Average A = (a1 + a2 + · · · + an)/n
Negation. ∀i ∈ {1, 2, . . . , n} ai A. That is
We have a1 A, a2 A, . . ., an A

Average
Proposition
Proof
Average A = (a1 + a2 + · · · + an)/n
Negation. ∀i ∈ {1, 2, . . . , n} ai A. That is
We have a1 A, a2 A, . . ., an A
Now add all these inequalities to get
(a1 + a2 + · · · + an) n × A
=⇒ A (a1 + a2 + · · · + an)/n on simplification
How is it possible that A is both equal to and greater than
(a1 + a2 + · · · + an)/n

Average
Proposition
Proof
Let amax represent the maximum among the n real numbers.
Let average A = (a1 + a2 + · · · + an)/n. Then

Average
Proposition
Proof
a1 = amax − b1 such that b1 ≥ 0
. . .
an = amax − bn such that bn ≥ 0

Average
Proposition
Proof
. . .
an = amax − bn such that bn ≥ 0
Adding the above equations, we get
(a1 + a2 + · · · + an) = n × amax − (b1 + b2 + · · · + bn)
=⇒ amax = [(a1 + a2 + · · · + an) + (b1 + b2 + · · · + bn)]/n
= ((a1 + a2 + · · · + an)/n) + ((b1 + b2 + · · · + bn)/n)
= A + ((b1 + b2 + · · · + bn)/n)
≥ A (∀i, bi ≥ 0)

2p
− 1 is prime =⇒ p is prime
Proposition
Suppose p ∈ N and p ≥ 2. If 2p − 1 is prime, then p is prime.

2p
Proposition
Proof
Negation. Suppose p is an integer at least 2 such that 2p − 1
is prime and p is composite.

2p
Proposition
Proof
Negation. Suppose p is an integer at least 2 such that 2p − 1
is prime and p is composite.
p is composite
=⇒ p = rs such that both r, s are in the range [2, p − 1]
Then, 2p − 1
= 2rs − 1 (substitute for p)
= (2r)s − 1 (abc = (ab)c)
= (2r − 1)

(2r)s−1
2r−1

(multiply and divide by (2r − 1) 0)
= (2r − 1) 1 + (2r)1 + (2r)2 + · · · + (2r)s−1

= m × n (m ≥ 2 and n ≥ 2)

Pythagorean triplets
Proposition
For integers a, b, c, if a2 + b2 = c2, then a is even or b is even.

Proposition
Proof
Negation. a and b are odd and a2 + b2 = c2.

Proposition
Proof
a = 2m + 1; b = 2n + 1 (definition of odd)
Consider a2 + b2
= (2m + 1)2 + (2n + 1)2
= 4m2 + 4n2 + 4m + 4n + 2 (expand)
= 4 × (m2 + n2 + m + n) + 2 (take common factor)
≡ 2 mod 4 (remainder is 2 when divided by 4)

Proposition
Proof
a = 2m + 1; b = 2n + 1 (definition of odd)
Consider a2 + b2
= (2m + 1)2 + (2n + 1)2
= 4m2 + 4n2 + 4m + 4n + 2 (expand)
= 4 × (m2 + n2 + m + n) + 2 (take common factor)
≡ 2 mod 4 (remainder is 2 when divided by 4)
c = 2k or c = 2k + 1 (quotient-remainder theorem)
Consider c2
= 4k2 or 4(k2 + k) + 1 (squaring)
6≡ 2 mod 4 (remainder is never 2 when divided by 4)

n2
+ 3n + 2
Proposition
Proof 2
False!
Negation. ∀ natural number n, n2 + 3n + 2 is composite.
We prove the negation in two cases:
1. n is even
2. n is odd

n2
+ 3n + 2
Proof 2 (continued)
1. Prove that n is even =⇒ n2 + 3n + 2 is composite.
n is even
=⇒ n2 is even and 3n is even (even × integer = even)
=⇒ n2 + 3n + 2 is even (even + even = even)
=⇒ n2 + 3n + 2 is composite (2 is a factor)
2. Prove that n is odd =⇒ n2 + 3n + 2 is composite.
n is odd
=⇒ n2 is odd and 3n is odd (odd × odd = odd)
=⇒ n2 + 3n is even (odd + odd = even)

n2
+ 3n + 2
Proof 2 (continued)
1. Prove that n is even =⇒ n2 + 3n + 2 is composite.
n is even
=⇒ n2 is even and 3n is even (even × integer = even)
2. Prove that n is odd =⇒ n2 + 3n + 2 is composite.
n is odd
=⇒ n2 is odd and 3n is odd (odd × odd = odd)
=⇒ n2 + 3n is even (odd + odd = even)
Proposition
Use this approach to prove that for all natural number n,
9n4 − 7n3 + 5n2 − 3n + 10 is composite.

Odd2
= 8m + 1
Proposition
The square of any odd integer has the form 8m + 1 for some
integer m.

Odd2
= 8m + 1
Proposition
The square of any odd integer has the form 8m + 1 for some
integer m.
Proof
n is odd
=⇒ n = 4q or n = 4q + 1 or n = 4q + 2 or n = 4q + 3
(n can be written in one of the four forms
using the quotient-remainder theorem)
But, n 6= 4q and n 6= 4q + 2 (as 4q and 4q + 2 are even)
Hence, n = 4q + 1 or n = 4q + 3.
Case 1. n = 4q + 1.
=⇒ n2 = (4q + 1)2 = 8(2q2 + q) + 1 = 8m + 1,
where m = 2q2 + q = integer.
Case 2. n = 4q + 3.
=⇒ n2 = (4q + 3)2 = 8(2q2 + 3q + 1) + 1 = 8m + 1,
where m = 2q2 + 3q + 1 = integer.

(x2
− y2
) mod 4 6= 2
Proposition
There is no solution in integers to: (x2 − y2) mod 4 = 2.

(x2
− y2
) mod 4 6= 2
Proposition
There is no solution in integers to: (x2 − y2) mod 4 = 2.
Proof
Case 1. x is even and y is even.
=⇒ x2 = 4m and y2 = 4n
=⇒ x2 − y2 = 4(m − n).
Case 2. x is even and y is odd.
=⇒ x2 = 4m and y2 = 4n + 1
=⇒ x2 − y2 = 4(m − n) − 1.
Case 3. x is odd and y is even.
=⇒ x2 = 4m + 1 and y2 = 4n
=⇒ x2 − y2 = 4(m − n) + 1.
Case 4. x is odd and y is odd.
=⇒ x2 = 4m + 1 and y2 = 4n + 1
=⇒ x2 − y2 = 4(m − n).
In all these four cases, (x2 − y2) mod 4 6= 2.

Prove or disprove the following propositions:
If more than n pigeons fly into n pigeon holes for natural
number n, then at least one pigeon hole will contain at least two
pigeons. [Hint: Contradiction.]
1/
√
2 is irrational. [Hint: Contradiction.]
√
√
log2 3 is irrational. [Hint: Contradiction.]
log2 7 is irrational. [Hint: Contradiction.]
For all integers a and b, if ab is a multiple of 6, then a is even
and b is a multiple of 3. [Hint: Counterexample.]
There are no integers a and b such that 752b = 4183 − 326a.
[Hint: Contradiction.]
an + bn = cn has no integral solutions for all natural numbers
n ≥ 1. [Hint: Counterexample.]
[Hint: Contraposition.]

[Hint: Contraposition + division into cases.]
There are 1000 consecutive natural numbers that are not perfect
squares. [Hint: Direct proof.]
Consider any ten prime numbers that are greater than or equal
to 15. Then the sum of these prime numbers can never be (1
trillion + 1). [Hint: Direct proof, contradiction.]
Let n be a positive integer. Prove that the closed interval [n, 2n]
contains a power of 2. [Hint: Division into cases (power of 2 and
not a power of 2).]

Rational + rational = rational. [Hint: Direct proof.]
Rational + irrational = irrational. [Hint: Contradiction.]
Irrational + irrational = rational or irrational. [Hint: Examples.
√
2 + (−
√
2) = 0 and 1
√
2
+ 1
√
2
=
√
2.]
Rational × rational = rational. [Hint: Direct proof.]
Rational × irrational = rational or irrational. [Hint: Examples
0 ×
√
2 = 0 and 1 ×
√
2 =
√
2.]
Nonzero rational × irrational = irrational. [Hint: Contradiction.]
Irrational × irrational = rational or irrational. [Hint: Examples
√
2 ×
√
2 = 2 and
√
2 ×
√
2 =
√
6.]
Rationalrational = rational or irrational. [Hint: Examples 11 = 1
and 21/2 =
√
2.]

Prove 1 = 2 using basic algebra
Proof
a 0, b 0 B Given
a = b B Given
ab = b2 B Multiply both sides by b
ab − a2 = b2 − a2 B Subtract a2 from both sides
a(b − a) = (b + a)(b − a) B Factoring
a = b + a B Divide both sides by (b − a)
0 = b B Subtract a from both sides
b = 2b B Add b to both sides
1 = 2 B Divide both sides by b
What is the problem with this proof?

Proof
a 0, b 0 B Given
a = b B Given
ab = b2 B Multiply both sides by b
ab − a2 = b2 − a2 B Subtract a2 from both sides
a(b − a) = (b + a)(b − a) B Factoring
a = b + a B Divide both sides by (b − a)
0 = b B Subtract a from both sides
b = 2b B Add b to both sides
1 = 2 B Divide both sides by b
Error
Cannot divide by 0 in mathematics
Cannot divide by (b − a) as a = b

Proof
n2 + 2n + 1 = (n + 1)2 B Expand
n2 = (n + 1)2 − (2n + 1) B Subtract
n2 − n(2n + 1) = (n + 1)2 − (2n + 1) − n(2n + 1) B Subtract
n2 − n(2n + 1) = (n + 1)2 − (n + 1)(2n + 1) B Factoring
n2 − n(2n + 1) + (2n + 1)2/4 =
(n + 1)2 − (n + 1)(2n + 1) + (2n + 1)2/4 B Add
(n − (2n + 1)/2)2 = ((n + 1) − (2n + 1)/2)2 B Simplify
n − (2n + 1)/2 = (n + 1) − (2n + 1)/2 B Square roots
n = n + 1 B Add
1 = 2 B Subtract

Proof
n2 + 2n + 1 = (n + 1)2 B Expand
n2 = (n + 1)2 − (2n + 1) B Subtract
n2 − n(2n + 1) = (n + 1)2 − (2n + 1) − n(2n + 1) B Subtract
n2 − n(2n + 1) = (n + 1)2 − (n + 1)(2n + 1) B Factoring
n2 − n(2n + 1) + (2n + 1)2/4 =
(n + 1)2 − (n + 1)(2n + 1) + (2n + 1)2/4 B Add
(n − (2n + 1)/2)2 = ((n + 1) − (2n + 1)/2)2 B Simplify
n − (2n + 1)/2 = (n + 1) − (2n + 1)/2 B Square roots
n = n + 1 B Add
1 = 2 B Subtract
Error
Cannot take square roots directly
a2 = b2 does not imply a = b
E.g.: 12 = (−1)2 does not imply 1 = −1

Prove 1 = 2 using calculus
Proof
R
udv = uv −
R
vdu B Product rule
Set u = 1
x and v = x; We get du = − 1
x2 dx and dv = dx
R 1
x dx = x · 1
x −
R
x ·

− 1
x2

dx B Substitute
R 1
x dx = 1 +
R 1
x dx B Simplify
0 = 1 B Subtract
1 = 2 B Add

Prove 1 = 2 using calculus
Proof
R
udv = uv −
R
vdu B Product rule
Set u = 1
x and v = x; We get du = − 1
x2 dx and dv = dx
R 1
x dx = x · 1
x −
R
x ·

− 1
x2

dx B Substitute
R 1
x dx = 1 +
R 1
x dx B Simplify
0 = 1 B Subtract
1 = 2 B Add
Error
Cannot subtract integrals from both sides
R
dx = x + const. B const. depends on conditions
E.g.: d
dx (x + 1) = d
dx (x + 2) does not imply
R d
dx (x + 1) =
R d
dx (x + 2)

Prove 1 = 2 using algebra and calculus
Proof
x 6= 0 B Given
x = x B Given
x + x = 2x B Add
x + x + · · · + x
| {z }
x times
= x2 B Repeatedly add x times
1 + 1 + · · · + 1
| {z }
x times
= 2x B Differentiate
x = 2x B Simplify
1 = 2 B Divide

Prove 1 = 2 using algebra and calculus
Proof
x 6= 0 B Given
x = x B Given
x + x = 2x B Add
x + x + · · · + x
| {z }
x times
= x2 B Repeatedly add x times
1 + 1 + · · · + 1
| {z }
x times
= 2x B Differentiate
x = 2x B Simplify
1 = 2 B Divide
Error
Cannot write x + x + · · · + x
| {z }
x times
= x2 for non-integers
E.g.: Cannot write 1.5 + 1.5 + · · · + 1.5
| {z }
1.5 times
= 1.52

Prove 1 = 2 using continued fractions
Proof
1 = 2
3−1 = 2
3− 2
3−1
= 2
3− 2
3− 2
3−1
= 2
3− 2
3− 2
3− 2
3−1
= 2
3− 2
3− 2
3− 2
3−···
2 = 2
3−2 = 2
3− 2
3−2
= 2
3− 2
3− 2
3−2
= 2
3− 2
3− 2
3− 2
3−2
= 2
3− 2
3− 2
3− 2
3−···
1 = 2 B Continued fractions are the same

Prove 1 = 2 using continued fractions
Proof
1 = 2
3−1 = 2
3− 2
3−1
= 2
3− 2
3− 2
3−1
= 2
3− 2
3− 2
3− 2
3−1
= 2
3− 2
3− 2
3− 2
3−···
2 = 2
3−2 = 2
3− 2
3−2
= 2
3− 2
3− 2
3−2
= 2
3− 2
3− 2
3− 2
3−2
= 2
3− 2
3− 2
3− 2
3−···
1 = 2 B Continued fractions are the same
Error
Cannot equate the values of the continued fractions
The given continued fraction is x = 2
3−x
Solving for x, we have x = 1 or x = 2
Beware of infinity!

Prove 1 = 2 using infinite series
Proof
Consider Grandi’s series S = 1 − 1 + 1 − 1 + · · ·
S = (1 − 1) + (1 − 1) + · · · = 0 + 0 + · · · = 0
S = 1 + (−1 + 1) + (−1 + 1) + · · · = 1 + 0 + 0 + · · · = 1
0 = 1 B S = 0 and S = 1
1 = 2 B Add

Prove 1 = 2 using infinite series
Proof
Consider Grandi’s series S = 1 − 1 + 1 − 1 + · · ·
S = (1 − 1) + (1 − 1) + · · · = 0 + 0 + · · · = 0
S = 1 + (−1 + 1) + (−1 + 1) + · · · = 1 + 0 + 0 + · · · = 1
0 = 1 B S = 0 and S = 1
1 = 2 B Add
Error
Cannot use several algebraic methods on a divergent series
Grandi’s series is divergent
Beware of infinity!

Prove 1 = 2 using set theory
Proof
Using Georg Cantor’s set theory and his idea of one-to-one
correspondence, we can show that the number of points on the
number line segment [0, 1] is same as the number of points on
the number line segment [0, 2]
1 = 2

Prove 1 = 2 using set theory
Proof
Using Georg Cantor’s set theory and his idea of one-to-one
correspondence, we can show that the number of points on the
number line segment [0, 1] is same as the number of points on
the number line segment [0, 2]
1 = 2
Error
Solution is out of scope
The problem is because the principles that apply in the world of
finite quantities do not apply in the world of infinite quantities
Beware of infinity!

Prove 1 = 2 using geometry
Proof
Banach-Tarski paradox states that a solid ball can be split into a
finite number of disjoint subsets, which can then be assembled
to create two identical copies of the original solid ball
1 = 2

Prove 1 = 2 using geometry
Proof
Banach-Tarski paradox states that a solid ball can be split into a
finite number of disjoint subsets, which can then be assembled
to create two identical copies of the original solid ball
1 = 2
Error
Solution is out of scope
The problem is because the principles that apply in the world of
finite quantities do not apply in the world of infinite quantities
Beware of infinity!

The Pythagorean theorem
History. The theorem first appeared in a Babylonian tablet dated
1900-1600 B.C.
Incorrect proofs. Alexander Bogomolny’s website Cut-The-Knot
https://www.cut-the-knot.org/pythagoras/FalseProofs.shtml
presents 9 incorrect proofs of the theorem
Correct proofs. Elisha Scott Loomis’ book “The Pythagorean
Proposition” presents 367 correct proofs of the theorem
(algebraic proofs + geometric proofs + trigonometric proofs)
More Proofs. An infinite number of algebraic and geometric
proofs exist for the theorem (Proof?)

Proof Techniques

More Related Content

What's hot (20)

Similar to Proof Techniques (20)

Recently uploaded (20)

Proof Techniques