DM Module2

MATH F213: Discrete Mathematics
First Semester 2021-22
BITS Pilani
§1.7: Methods of proof of an implication
Learning objectives
In this section, we will learn
1. nine common methods for proving implications

2. to justify the conclusions in every step of a proof
1
Method 1: Trivial proof of an implication
If q is true, then the implication p ! q is true irrespective of the truth

value of p. A trivial proof of the implication p ! q consists of proving
that the conclusion holds without using the premise.
2

Example
Prove:
If x is a positive real number, then x 2 + 1 > 0.
2

Example
Prove:
If x is a positive real number, then x 2 + 1 > 0.
Solution. Since x 2 0 for all real numbers x, we get
x2 + 1 > x2 0.
Hence x 2 + 1 > 0.
(Note that we did not use the fact that x is given to be positive.)
2
Method 2: Vacuous proof of an implication
If p is false, then the implication p ! q is true irrespective of the truth

value of q. In this case, we say that the implication p ! q is vacuously
true (the statement exists in vacuum).
3

Example
Prove:
If x is a real number with x 2 + 1  0, then x < 100.
3

Example
Prove:
If x is a real number with x 2 + 1  0, then x < 100.
Solution. Since x 2 + 1 > 0 for all real numbers x (as shown in the
previous example), we find that the given implication is vacuously true.
3
Method 3: Direct proof of an implication
A direct proof of the implication p ! q proceeds as follows.

1. Assume that p is true.
2. Using that assumption and other available information, prove that q
is true in a direct sequence of implications.
4

Example. (Ex. 2(h))

Prove:
Let a, b, c be integers. If a divides both b and c, then a2 divides bc.
4

Example. (Ex. 2(h))

Prove:
Solution. Since a divides both b and c, we get that

b = ak, c = a`
for some integers k, `.
4

Example. (Ex. 2(h))

Prove:

b = ak, c = a`
for some integers k, `. Now
bc = (ak)(a`) = a2 (k`).
4

Example. (Ex. 2(h))

Prove:

b = ak, c = a`
for some integers k, `. Now
bc = (ak)(a`) = a2 (k`).
Since k, ` are integers, k` is also an integer. Hence a2 divides bc.
4
Method 4: Indirect proof of an implication (Proof by Contra-
positive)
An indirect proof of the implication p ! q is a direct proof of its

contrapositive
⇠q!⇠p
.
5
positive)

contrapositive
⇠q!⇠p
.
Example.
Prove:
p
Let a, b, n be positive integers. If n = ab, then either a  n or,
p
b  n.
5
positive)

contrapositive
⇠q!⇠p
.
Example.
Prove:
p
p
b  n.
Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).)
5
positive)

contrapositive
⇠q!⇠p
.
Example.
Prove:
p
p
b  n.
Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).) Assume that

p p
a > n and b > n. Hence
p p
ab > ( n)( n) = n.
5
positive)

contrapositive
⇠q!⇠p
.
Example.
Prove:
p
p
b  n.
Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).) Assume that

p p
a > n and b > n. Hence
p p
ab > ( n)( n) = n.
p
Therefore n 6= ab. Hence if n = ab, then we must have either a  n
p
or, b  n.
5
positive)
Example.
Prove:
Let n be an integer and let p denote a factor of n exceeding 1. If p is
the smallest such factor, then p must be prime.
6
positive)
Example.
Prove:
Solution. Assume that p is composite.
6
positive)
Example.
Prove:
Solution. Assume that p is composite. Then, we can write
p = ab, 1 < a, b < p.
6
positive)
Example.
Prove:
p = ab, 1 < a, b < p.
Since a divides p and p divides n, we get that a divides n.
6
positive)
Example.
Prove:
p = ab, 1 < a, b < p.
Since a divides p and p divides n, we get that a divides n. But

1 < a < p. Therefore, p is not the smallest factor of n exceeding 1.
Hence the smallest factor of n exceeding 1 must be prime.
6
Algorithm to find the smallest prime factor
Given a positive integer n, our goal is to find the smallest prime factor of
n.
1. If n is even, its smallest prime factor is 2.
7
n.
2. If n is odd, let
p
s = b nc
p
(i.e., s is the greatest integer  n) and let P = {a1 , . . . , ak }
denote the set of all primes  s, with a1 < a2 < · · · < ak .
7
n.
2. If n is odd, let
p
s = b nc
p
3. Let i = 1. If ai divides n, return ai . Otherwise, increase i by 1 and
repeat Step 3. Return n if i exceeds k.
7
n.
2. If n is odd, let
p
s = b nc
p
Exercise. (Ex. 31(a))
Prove that 421 is prime.
7
n.
2. If n is odd, let
p
s = b nc
p
Exercise. (Ex. 31(a))
Prove that 421 is prime.
p
Hint. Use the above algorithm. Now 421 is odd and b 421c = 20. Next,
the primes  20 are 2, 3, 5, 7, 11, 13, 17, 19. Check whether any one of
these primes divides 421.
7
Method 5: Proof of an implication by contradiction
Recall that p ! q is true if and only if p ^ (⇠ q) is false. A proof of the

implication p ! q by contradiction proceeds as follows.
1. Assume that p ^ (⇠ q) is true.

2. Using that assumption, obtain a conclusion that is false, or
contradicts some other fact that has been established.
3. This shows that the assumption p ^ (⇠ q) must be false. Hence
implication p ! q is true.
8
Pigeonhole Principle
Let m1 , . . . , mn be positive integers. If m1 + m2 + . . . + mn n + 1
objects are put into n boxes, then either the first box contains at least
m1 objects, or the second box contains at least m2 objects,..., or the
nth box contains at least mn objects.
9
Proof. For i = 1, . . . , n,
let qi represent the statement ‘the ith box contains at least mi objects’
and
let p represent the statement ‘m1 + m2 + . . . + mn n + 1 objects are
put into n boxes’.
9
Proof. For i = 1, . . . , n,
and
To prove p ! (q1 _ q2 _ · · · _ qn ) by contradiction, we assume
p^ ⇠ (q1 _ q2 _ · · · _ qn ) or, p ^ (⇠ q1 ) ^ (⇠ q2 ) ^ · · · ^ (⇠ qn ) (by
DeMorgan’s law).
9
Proof. For i = 1, . . . , n,
and
p^ ⇠ (q1 _ q2 _ · · · _ qn ) or, p ^ (⇠ q1 ) ^ (⇠ q2 ) ^ · · · ^ (⇠ qn ) (by
DeMorgan’s law). Thus, for i = 1, . . . , n, the ith box contains at most
mi 1 objects. Hence the total number of objects in the n boxes is at
most (m1 1) + (m2 1) + · · · + (mn 1) = m1 + . . . + mn n.
9
Proof. For i = 1, . . . , n,
and
p^ ⇠ (q1 _ q2 _ · · · _ qn ) or, p ^ (⇠ q1 ) ^ (⇠ q2 ) ^ · · · ^ (⇠ qn ) (by
DeMorgan’s law). Thus, for i = 1, . . . , n, the ith box contains at most
mi 1 objects. Hence the total number of objects in the n boxes is at
most (m1 1) + (m2 1) + · · · + (mn 1) = m1 + . . . + mn n. This is
a contradiction, as the total number of objects in the n boxes is
m1 + m2 + . . . + mn n + 1. Hence our assumption is wrong. This
proves the Pigeonhole Principle.
9
Pigeonhole Principle (Second form)
If A denotes the average number of pigeons per hole, then some
pigeonhole contains at least dAe pigeons and some pigeonhole contains
at most bAc pigeons.
10
Proof. Let n denote the number of pigeonholes. For i = 1, . . . , n, let mi

denote the number of pigeons in the ith pigeonhole. We prove that either
m1  A or m2  A, . . . , or mn  A.
10

m1  A or m2  A, . . . , or mn  A. Assume that m1 > A and
m2 > A, . . . , and mn > A. Then, the total number of pigeons
m1 + . . . + mn > nA, implying that (m1 + . . . + mn )/n > A.
10

m1 + . . . + mn > nA, implying that (m1 + . . . + mn )/n > A. This is a
contradiction as A denotes the average number of pigeons per hole.
Hence our assumption is wrong. Therefore either m1  A or m2  A, . . . ,
or mn  A. Since mi ’s are integers, we get that some pigeonhole contains
10

m1 + . . . + mn > nA, implying that (m1 + . . . + mn )/n > A. This is a
contradiction as A denotes the average number of pigeons per hole.
Hence our assumption is wrong. Therefore either m1  A or m2  A, . . . ,
or mn  A. Since mi ’s are integers, we get that some pigeonhole contains
(The proof that some pigeonhole contains at least dAe pigeons is similar
and is left as an exercise or, see Example 1.7.13 of the textbook.)
10
Notation
• N denotes the set of positive integers:
N = {1, 2, 3, . . .}
• Z denotes the set of integers:
Z = {. . . , 3, 2, 1, 0, 1, 2, 3, . . .}
11
Applications of Pigeonhole Principle
• Let k 2 N. If kn + 1 pigeons are distributed in n pigeonholes, then

some pigeonhole contains at least k + 1 pigeons.
(Observe that dk + 1/ne = k + 1.)
12

Remark. In particular, taking k = 1, we get that if n + 1 pigeons are

distributed in n pigeonholes, then some pigeonhole contains at least 2
pigeons.
12

Remark. In particular, taking k = 1, we get that if n + 1 pigeons are

distributed in n pigeonholes, then some pigeonhole contains at least 2
pigeons.
Recall: The Division Algorithm

Let a, b 2 Z with b > 0. Then there exist unique integers q and r such
that
a = bq + r , 0  r < b.
12
Question
1. Let b 2 N. Prove that among any b + 1 numbers, it is possible to
find two numbers whose difference is divisible by b.
2. Prove that given any b 2 N, it is possible to find a non-zero number
which is divisible by b and whose digits are 0 and 5 only.
13
Question
Solution. (1) Let a1 , a2 , ..., ab+1 denote the numbers. By Division

algorithm, for each i = 1, . . . , b + 1, there exist unique integers qi and ri
such that
ai = bqi + ri , 0  ri  b 1.
13
Question

such that
ai = bqi + ri , 0  ri  b 1.
Since there are b possible values for each ri , but there are b + 1 ri ’s, by
the Pigeonhole Principle, there exist j, k, 1  j < k  b + 1 such that
rj = rk .
13
Question

such that
ai = bqi + ri , 0  ri  b 1.
rj = rk . Now
aj ak = (bqj + rj ) (bqk + rk ) = b(qj qk ) + 0.
Hence b divides aj ak . This completes the proof (note that j 6= k).
13
Question

such that
ai = bqi + ri , 0  ri  b 1.
rj = rk . Now
aj ak = (bqj + rj ) (bqk + rk ) = b(qj qk ) + 0.
Hence b divides aj ak . This completes the proof (note that j =
6 k).
13
Hint. (2) Consider numbers from the list 5, 55, 555, 5555, . . .
Method 6: Proof of an implication by cases
If p is of the form p1 _ p2 _ · · · _ pn , we may prove the implication p ! q

by proving separately each of the following implications
p1 ! q, p2 ! q, . . . , pn ! q.
14

p1 ! q, p2 ! q, . . . , pn ! q.
Example.
Prove:
If n is an integer, then n2 + 3n + 1 is an odd integer.
14

p1 ! q, p2 ! q, . . . , pn ! q.
Example.
Prove:
Solution. We make two cases, according to whether n is even or odd. If

n is even, write n = 2m, m 2 Z. Now
n2 + 3n + 1 = 4m2 + 6m + 1 = 2(2m2 + 3m) + 1,
which is an odd integer.
14

p1 ! q, p2 ! q, . . . , pn ! q.
Example.
Prove:

n2 + 3n + 1 = 4m2 + 6m + 1 = 2(2m2 + 3m) + 1,
which is an odd integer. If n is odd, write n = 2k + 1, k 2 Z. Now
n2 + 3n + 1 = (4k 2 + 4k + 1) + (6k + 3) + 1 = 2(2k 2 + 5k + 2) + 1,
which is an odd integer.
14

p1 ! q, p2 ! q, . . . , pn ! q.
Example.
Prove:

n2 + 3n + 1 = 4m2 + 6m + 1 = 2(2m2 + 3m) + 1,
which is an odd integer. If n is odd, write n = 2k + 1, k 2 Z. Now
n2 + 3n + 1 = (4k 2 + 4k + 1) + (6k + 3) + 1 = 2(2k 2 + 5k + 2) + 1,
which is an odd integer. Since n2 + 3n + 1 is shown to be an odd integer
in both the cases, the proof is complete. 14
Method 7: Proof of an implication by elimination of cases
We use [{(p1 _ p2 _ · · · _ pn ) _ q} ^ (⇠ p1 ) ^ (⇠ p2 ) ^ · · · ^ (⇠ pn )] ! q.
15
Example.
Prove:
If p is an odd prime, then either p = 3 or p has the form 6n + 1 or
6n + 5.
15
Example.
Prove:
6n + 5.
Solution. By Division Algorithm, there exist unique integers q, r such

that
p = 6q + r , 0  r  5.
15
Example.
Prove:
6n + 5.

that
p = 6q + r , 0  r  5.
If r = 0, we get 6 divides p, which is false as p is prime. Hence r 6= 0.
15
Example.
Prove:
6n + 5.

that
p = 6q + r , 0  r  5.
If r = 0, we get 6 divides p, which is false as p is prime. Hence r 6= 0. If
r = 2, 4, we get that 2 divides p, which is false as p is an odd prime.
Hence r 6= 2, 4.
15
Example.
Prove:
6n + 5.

that
p = 6q + r , 0  r  5.
Hence r 6= 2, 4. If r = 3, we get that 3 divides p, which is false as p is a
prime, unless q = 0 (and p = 3).
15
Example.
Prove:
6n + 5.

that
p = 6q + r , 0  r  5.
Hence r 6= 2, 4. If r = 3, we get that 3 divides p, which is false as p is a
prime, unless q = 0 (and p = 3). Thus, either p = 3 or p has the form
6n + 1 or 6n + 5.
15
1. One form:
[{(p1 _ p2 _ · · · _ pn ) _ q} ^ (⇠ p1 ) ^ (⇠ p2 ) ^ · · · ^ (⇠ pn )] ! q
2. Another form: {[p ! (q _ r )] ^ (⇠ q)} ! (p ! r )
16
Method 8: Conditional proof of an implication
We use that p ! (q ! r ) is equivalent to (p ^ q) ! r .
17

Example.
Prove:
If p is a prime, then if p divides the product of two integers a and b,
then either p divides a or p divides b.
17

Example.
Prove:
Solution. Suppose that p is a prime and p divides ab. (We are using
Conditional Proof.) Now we proceed by elimination of cases.
17

Example.
Prove:
Conditional Proof.) Now we proceed by elimination of cases. Suppose p
does not divide a. We will show that p has to divide b.
17

Example.
Prove:
does not divide a. We will show that p has to divide b. Since p is prime
not dividing a, we have gcd(p, a) = 1. Hence, by Bézout’s identity, there
exist integers x, y such that 1 = px + ay .
17

Example.
Prove:
does not divide a. We will show that p has to divide b. Since p is prime
not dividing a, we have gcd(p, a) = 1. Hence, by Bézout’s identity, there
exist integers x, y such that 1 = px + ay . This implies that
b = bpx + bay .
Since p divides bpx = p(bx) and p divides bay (as p divides ab), we get
that p divides bpx + bay i.e., p divides b. This completes the proof.
17
Proof of an equivalence
To prove p $ q, we prove each of p ! q and q ! p, using any valid

method of proof.
18

method of proof.
Example.
Prove: For any integer n, n is odd if and only if n2 is odd.
18

method of proof.
Example.
Solution. To prove the given statement, we need to prove the following

two implications.
1. If n is odd, then n2 is odd.
2. If n2 is odd, then n is odd.
18

method of proof.
Example.

two implications.
We first give a direct proof of (1). Assume that n is an odd integer.
Then n = 2k + 1 for some integer k. Hence
n2 = (2k + 1)2 = 2(2k 2 + 2k) + 1,
which is odd. This proves (1).
18

method of proof.
Example.

two implications.
n2 = (2k + 1)2 = 2(2k 2 + 2k) + 1,
which is odd. This proves (1). Next, we prove (2) by proving its
contrapositive, namely, ‘if n is even, then n2 is even’.
18

method of proof.
Example.

two implications.
n2 = (2k + 1)2 = 2(2k 2 + 2k) + 1,
which is odd. This proves (1). Next, we prove (2) by proving its
contrapositive, namely, ‘if n is even, then n2 is even’. Assume that n is an
even integer. Hence n = 2k for some integer k. Then n2 = 2(2k 2 ),
which is an even integer. This proves (2).
18
Question.
p
Prove that 2 is irrational.
19
Question.
p
Solution. We will prove the given statement by contradiction. Suppose

p
that 2 is rational. Write
p a
2 = , b > 0, gcd(a, b) = 1.
b
19
Question.
p

p
p a
2 = , b > 0, gcd(a, b) = 1.
b
Squaring and multiplying both sides by b 2 , we get 2b 2 = a2 . Hence a2 is
even, implying that a is even. (We have proved earlier that for any
integer n, n is odd if and only if n2 is odd. Why is that useful here?)
19
Question.
p

p
p a
2 = , b > 0, gcd(a, b) = 1.
b
Write a = 2k, for some integer k. Then
2b 2 = (2k)2 = 4k 2 .
Hence b 2 = 2k 2 . Therefore b 2 is even, implying that b is even.
19
Question.
p

p
p a
2 = , b > 0, gcd(a, b) = 1.
b
Write a = 2k, for some integer k. Then
2b 2 = (2k)2 = 4k 2 .
Hence b 2 = 2k 2 . Therefore b 2 is even, implying that b is even. Hence 2

divides both a and b. This contradicts the fact that gcd(a, b) = 1. Hence
p
our assumption must be wrong. Thus 2 is irrational.
19
§1.8: First order logic & other method of proof
Learning objectives
In this section, we will
1. learn some of the limits of propositional logic

2. introduce first order logic
20
Some limits of Propositional Logic
1. There is no way of saying that all objects of some type share certain
properties. Each fact requires a separate symbol.
21
2. Consider the following argument.
All humans are mortal.
Socrates is a human.
Therefore, Socrates is mortal.
Symbolizing this in propositional logic, we get an argument of the
form
p
q
—–
)r
21
2. Consider the following argument.
All humans are mortal.
Socrates is a human.
Therefore, Socrates is mortal.
Symbolizing this in propositional logic, we get an argument of the
form
p
q
—–
)r
This is an invalid argument in propositional logic as p, q, r have no
relation to each other.
21
First order logic
First-order logic is a formal language in which propositions are expressed

in terms of predicates, variables and quantifiers.
22
First order logic

Definition
An open proposition (or, predicate) in n variables from a set U is a
function f : U n ! {T , F }. Here U n denotes the n-fold Cartesian
product U ⇥ U ⇥ · · · ⇥ U and T , F represent True and False,
respectively. The set U is called the Universe of Discourse of f .
22
First order logic

Definition
Predicates are often abbreviated by single capital letters followed by a list

of arguments, the variables that appear in the predicate.
22
First order logic

Definition

of arguments, the variables that appear in the predicate. For example,
1. Z (x): x is an integer 4. I (x): x is an isosceles triangle
2. L(y ): y < 10 5. E (x): x is an equilateral triangle
3. P(x, y ): x is a parent of y
What is U in each case?
22
First order logic

Definition

of arguments, the variables that appear in the predicate. For example,
1. Z (x): x is an integer 4. I (x): x is an isosceles triangle
2. L(y ): y < 10 5. E (x): x is an equilateral triangle
3. P(x, y ): x is a parent of y
What is U in each case? By plugging in specific values for all the
variables, we obtain a proposition and we can discuss whether the
proposition is true or false. For instance, Z (1/2) is false, L(5) is true. 22
Quantifiers
Definition
The universal quantifier 8 states that a statement is true for all values
of a variable within some universe of permitted values.
23
Quantifiers
Definition
For example, ‘all equilateral triangles are isosceles’ can be expressed as

8x, [E (x) ! I (x)].
23
Quantifiers
Definition
For example, ‘all equilateral triangles are isosceles’ can be expressed as

8x, [E (x) ! I (x)].
Similarly, ‘all humans are mortal’ can be expressed as
8x, [H(x) ! M(x)], where
H(x) means x is human,
M(x) means x is mortal.
23
Quantifiers
If the Universe of Discourse U is the set of all integers, P(x) denotes ‘x

is prime’, G (x) denotes ‘x > 2’ and O(x) denotes ‘x is odd’, then we can
express ‘all primes > 2 are odd’ as
8x, {[P(x) ^ G (x)] ! O(x)}
24
Quantifiers
If the Universe of Discourse U is the set of all integers, P(x) denotes ‘x

is prime’, G (x) denotes ‘x > 2’ and O(x) denotes ‘x is odd’, then we can
express ‘all primes > 2 are odd’ as
8x, {[P(x) ^ G (x)] ! O(x)}
However, if the Universe of Discourse U is the set of primes > 2, then we

can express ‘all primes > 2 are odd’ as
8x, O(x)
24
Quantifiers
Definition
The existential quantifier 9 states that a statement must be true for at
least one value of the variable.
25
Quantifiers
Definition
The existential quantifier 9 states that a statement must be true for at
least one value of the variable.
For example, ‘some human is mortal’ can be expressed as

9x, [H(x) ^ M(x)].
(Be careful: why should we not write 9x, [H(x) ! M(x)]?)
25
Quantifiers
1. The set membership notation can be used to specify that the

universal and existential quantifiers are limited to some universe. For
example, 9x 2 Z, [x 2 = 4].
26
Quantifiers

example, 9x 2 Z, [x 2 = 4].
2. The symbol 9!x means ‘there is a unique x such that’. For example,
if P(x) means ‘x is an even prime’, then ‘9!x, P(x)’ means that
there is one and only one even prime.
26
Quantifiers

example, 9x 2 Z, [x 2 = 4].
2. The symbol 9!x means ‘there is a unique x such that’. For example,
if P(x) means ‘x is an even prime’, then ‘9!x, P(x)’ means that
there is one and only one even prime.
3. Negation:
⇠ 8x, P(x) ⌘ 9x, ⇠ P(x)
⇠ 9x, P(x) ⌘ 8x, ⇠ P(x)
26
Quantifiers
Sentence Abbreviated meaning

8x, F (x) all true
9x, F (x) at least one true
⇠ [9x, F (x)] none true
8x, [⇠ F (x)] all false
9x, [⇠ F (x)] at least one false
⇠ {9x, [⇠ F (x)]} none false
⇠ {8x, F (x)} not all true
⇠ {8x, [⇠ F (x)]} not all false
27
Quantifiers
Question (Ex 1 (b), (g), (o))

Express the following using first order logic.
1. Not all birds can fly.

2. Some men are not giants.
3. Not every actor is talented who is famous.
28
Quantifiers


Solution.
1. ⇠ {8x, [B(x) ! F (x)]} or, equivalently, 9x, (B(x) ^ [⇠ F (x)])
28
Quantifiers


Solution.

2. 9x, [M(x)^ ⇠ G (x)]
28
Quantifiers


Solution.

2. 9x, [M(x)^ ⇠ G (x)]
3. ⇠ {8x, [(A(x) ^ F (x)) ! T (x)]} or, equivalently,
9x, [A(x) ^ F (x)^ ⇠ T (x)]
28
Combining Quantifiers
1. The statement ‘the product of two real numbers is a real number’

can be written as
(8x) (8y ) [x 2 R ^ y 2 R ! xy 2 R].
29
1. The statement ‘the product of two real numbers is a real number’

can be written as
(8x) (8y ) [x 2 R ^ y 2 R ! xy 2 R].
2. Let U = R. The statement ‘for every real number there is some real
number greater than the number itself’ can be written as
(8x) (9y ) [y > x].
29
1. Note that (8x) (8y ) P(x, y ) and (8y ) (8x) P(x, y ) are logically
equivalent. Similarly, (9x) (9y ) P(x, y ) and (9y ) (9x) P(x, y ) are
logically equivalent.
30
1. Note that (8x) (8y ) P(x, y ) and (8y ) (8x) P(x, y ) are logically
equivalent. Similarly, (9x) (9y ) P(x, y ) and (9y ) (9x) P(x, y ) are
logically equivalent.
2. The quantifiers 8 and 9 should not be interchanged. For example, if
Rxy means x respects y , then
(9x) (8y ) [Ryx]
means that there is a person who is respected by everyone, but
(8x) (9y ) [Ryx]
means that for every person, there exists a person who respects
them, and
(8y ) (9x) [Ryx]
means that for every person, there exists a person who they respect.
30
Quantifiers
Question
1. Some humans own a house.

2. Every house has an owner.
31
Quantifiers
Question

Solution.
1. (9x) (9y )[Human(x) ^ House(y ) ^ Owns(x, y )]
31
Quantifiers
Question

Solution.
1. (9x) (9y )[Human(x) ^ House(y ) ^ Owns(x, y )]

2. 8x, {House(x) ! 9y , [Owns(y , x)]}
31
Proof techniques
1. Proof by example To prove a statement of the type ‘9x, F (x)’, it is

enough to show that F (c) is true for some c in the universe.
Such proofs are called existence proofs. An existence proof may be
constructive, where either a value of c as above is actually given or, an
algorithm is given for finding such a value. An existence proof may also
be non-constructive, where it is not indicated how such a c may be
found. The proof may be proof by contradiction or proof by cases.
32
Proof techniques
Question (Ex 4 (i))

Determine the truth or falsity of the statement
9x, [{x 2 > 10} ^ {x is even}],
when U = Z.
33
Proof techniques
Question (Ex 4 (i))

Determine the truth or falsity of the statement
9x, [{x 2 > 10} ^ {x is even}],
when U = Z.
Taking x = 4, we find that x is even and x 2 = 42 = 16 and therefore

x 2 > 10.
33
Proof techniques
Question
Prove that there exist irrational numbers x, y such that x y is a rational
number.
34
Proof techniques
Question
number.
Solution. We will prove this using a non-constructive proof. Consider

p p2
the real number z = 2 .
34
Proof techniques
Question
number.

p p2 p
the real number z = 2 . We know that 2 is irrational. If z is
rational, we find that the statement is true.
34
Proof techniques
Question
number.

p p2 p
the real number z = 2 . We know that 2 is irrational. If z is
rational, we find that the statement is true. If z is irrational, then we
p p 2
consider z 2 = 2 = 2 and therefore the statement is true in this case
as well.
34
Lagrange’s interpolation formula
The Lagrange interpolating polynomial is the polynomial P(x) of degree

 n 1 with specified values y1 , . . . , yn (respectively) at specified points
x1 , . . . , xn . It is given by
2 3
Xn Y (x xj )
P(x) = 4 5 yi
(xi xj )
i=1 i6=j
(x x2 )(x x3 ) · · · (x xn ) (x x1 )(x x3 ) · · · (x xn )
= y1 + y2
(x1 x2 )(x1 x3 ) · · · (x1 xn ) (x2 x1 )(x2 x3 ) · · · (x2 xn )
(x x1 )(x x2 ) · · · (x xn 1 )
+ ... + yn .
(xn x1 )(xn x2 ) · · · (xn xn 1 )
35
Question (Exercise 10)

Use Lagrange interpolation formula to construct a polynomial whose
graph passes through the points ( 1, 1), (0, 1), (1, 1), (2, 5).
36
Question (Exercise 10)

Use Lagrange interpolation formula to construct a polynomial whose
graph passes through the points ( 1, 1), (0, 1), (1, 1), (2, 5).
Solution. Plugging in the formula, we obtain
(x 0)(x 1)(x 2) (x ( 1))(x 1)(x 2)

P(x) = ·1+ ·1
( 1 0)( 1 1)( 1 2) (0 ( 1))(0 1)(0 2)
(x ( 1))(x 0)(x 2) (x ( 1))(x 0)(x 1)
+ ·1+ · ( 5)
(1 ( 1))(1 0)(1 2) (2 ( 1))(2 0)(2 1)
= x 3 + x + 1 (Check!)
36
Proof techniques
2. Proof by exhaustion To prove a statement of the type

‘8x, [⇠ F (x)]’, one may check for all objects in the universe and show
that none of them have the property F (x).
37
Proof techniques
Question
Prove that there are no rational roots to the polynomial
f (x) = x 6 6x 4 + 12x 2 13.
38
Proof techniques
Question
f (x) = x 6 6x 4 + 12x 2 13.
Solution. The Rational Roots Theorem states that

If P(x) = an x n + · · · + a1 x + a0 is a polynomial with integer coefficients,
then any rational root of P(x) has the form a/b where a, b are integers
such that a divides a0 and b divides an .
38
Proof techniques
Question
f (x) = x 6 6x 4 + 12x 2 13.

Using this, we find that the only possible rational roots of the given
polynomial are ±1, ±13.
38
Proof techniques
Question
f (x) = x 6 6x 4 + 12x 2 13.

polynomial are ±1, ±13. Now f (1) = f ( 1) = 6 6= 0.
38
Proof techniques
Question
f (x) = x 6 6x 4 + 12x 2 13.

polynomial are ±1, ±13. Now f (1) = f ( 1) = 6 6= 0. Next,
f (13) = f ( 13) = 13(135 6 · 133 + 12 · 13 1) 6= 0,
as the term within the bracket is 1 less than a multiple of 13 and hence is
non-zero.
38
Exercise
1. You may use the Rational Roots Theorem to give another proof of
p
the fact that 2 is irrational.
p p
2. Prove that 2 + 3 5 is irrational. (The previous question will come
in handy!)
39
Proof techniques
3. Proof by counterexample To disprove a statement of the type

‘8x, F (x)’, it is enough to provide a value c in the universe for which
F (c) is false.
40
Proof techniques
Question
Disprove that n2 n + 41 is a prime number for all positive integers n.
41
Proof techniques
Question
Disprove that n2 n + 41 is a prime number for all positive integers n.
Solution. Taking n = 42, we find that
n2 n + 41 = n(n 1) + 41 = 42 · 41 + 41 = 41 · 43,
which is not prime.
41
§1.9: Rules of inference for quantified propositions
Learning objectives
In this section, we will learn four additional rules of inference which

describe when universal and existential quantifiers may be added to or
removed from an assertion.
42
Fundamental Rule 5: Universal Specification
8x, P(x)
——————–
) P(c) for all c
43
Fundamental Rule 5: Universal Specification
8x, P(x)
——————–
) P(c) for all c
Fundamental Rule 6: Universal Generalization
P(c) for all c

——————–
) 8x, P(x)
43
Fundamental Rule 7: Existential Specification
9x, P(x)
——————–
) P(c) for some c
Note. The statements ‘9x, P(x)’ and ‘9x, Q(x)’ need not imply that
‘9x, [P(x) ^ Q(x)]’.
44
Fundamental Rule 7: Existential Specification
9x, P(x)
——————–
) P(c) for some c
Note. The statements ‘9x, P(x)’ and ‘9x, Q(x)’ need not imply that
‘9x, [P(x) ^ Q(x)]’.
Fundamental Rule 8: Existential Generalization
P(c) for some c

——————–
) 9x, P(x)
44
Example
Symolize the following argument and check for its validity:
Lions are dangerous animals.

There are lions.
Therefore, there are dangerous animals.
45
Example
Symolize the following argument and check for its validity:
Lions are dangerous animals.

There are lions.
Therefore, there are dangerous animals.
Let L(x) mean ‘x is a lion’ and D(x) mean ‘x is a dangerous animal’.

The above argument can be written as
8x, [L(x) ! D(x)]

9x, L(x)
——————–
) 9x, D(x)
45
Formal Proof
Assertion Reasoning
9x, L(x) Premise 2
L(a) Step 1 and Existential Specification
8x, [L(x) ! D(x)] Premise 1
L(a) ! D(a) Step 3 and Universal Specification
D(a) Steps 2 & 4, Fundamental Rule 1
9x, D(x) Step 5 and Existential Generalization
46
Venn Diagrams
In this section, we will learn about the English mathematician John

Venn’s way of pictorially representing certain statements and checking
the validity of certain arguments.
47
We will represent every category by a circle. We will draw one circle
inside another to represent inclusion. We will draw disjoint circles to
represent null intersection. We will draw intersecting circles to represent
non-null intersection.
48
Example
All mathematicians are logicians.
49
Example
No mathematicians are logicians.
50
Example
Some mathematicians are logicians.
M L
51
Example
Some mathematicians are not logicians.
M L
52
Checking validity using venn diagram: Example 1
Check the validity of the following argument using venn diagrams.

No logician is a cyclist.
) No mathematician is a cyclist.
53
54
The given argument is valid.
54
Example 2
Some mathematicians are logicians.

Some logicians are cyclists.
) Some mathematicians are cyclists.
55
56
The given argument is invalid.
56
Example 3

Some logicians are cyclists.
) No mathematicians are cyclists.
57
58
The given argument is invalid.
58
§1.10: Mathematical Induction
Learning objectives
In this section, we will learn the technique of Mathematical Induction,

used for proving certain universally quantified statements.
59
The Principle of Mathematical Induction (PMI)/ Weak Induc-
tion
Let P(n) be a statement that depends on n 2 N. If
• P(1) is true, and

• ‘P(k) is true implies P(k + 1) is true’ for all k 2 N,
then P(n) is true for all n 2 N.
60
Example
Prove that for every positive integer n, 3 divides n3 n.
61
Example
Let P(n) be the statement ‘3 divides n3 n.’
61
Example

Base case: We verify that P(1) is true. In this case,
n3 n = 13 1 = 0, which is divisible by 3 as 0 = 3 · 0.
61
Example

Inductive hypothesis: Let k 1. We assume that the statement
P(k) is true, i.e., 3 divides k 3 k. This means that k 3 k = 3` for
some integer `.
61
Example

some integer `.
Inductive conclusion: We’ll show that the statement P(k + 1) is true,
i.e., 3 divides (k + 1)3 (k + 1).
61
Example

some integer `.
i.e., 3 divides (k + 1)3 (k + 1).
Now
(k + 1)3 (k + 1) = (k 3 + 3k 2 + 3k + 1) k 1
= (k 3 k) + 3(k 2 + k) = 3` + 3(k 2 + k) (by Inductive hyp
= 3(` + k 2 + k).
61
Example

some integer `.
i.e., 3 divides (k + 1)3 (k + 1).
Now
(k + 1)3 (k + 1) = (k 3 + 3k 2 + 3k + 1) k 1
= (k 3 k) + 3(k 2 + k) = 3` + 3(k 2 + k) (by Inductive hyp
= 3(` + k 2 + k).
Since ` + k 2 + k 2 Z, we get that 3 divides (k + 1)3 (k + 1). Thus
P(k + 1) is true. Hence by PMI, P(n) is true for all n 2 N.
61
Example
Prove that for every positive integer n, there are more than n primes.
62
Example
Let P(n) be the statement ‘there are more than n primes.’
62
Example

Base case: We verify that P(1) is true. Since 2, 3 are primes, there is
more than one prime. Hence P(1) is true.
62
Example

Inductive hypothesis: Let k 1. We assume that the statement P(k)
is true, i.e., there are more than k primes.
62
Example

i.e., there are more than k + 1 primes.
62
Example

Using the Inductive hypothesis, we know that there exist k + 1 primes
p1 , . . . , pk+1 (say).
62
Example

Using the Inductive hypothesis, we know that there exist k + 1 primes
p1 , . . . , pk+1 (say). Let
N = p1 · · · pk+1 + 1.
Since N > 1, there exists a prime p which divides N (by Fundamental
Theorem of Arithmetic). If p = pi for some i = 1, . . . , k + 1, then pi
divides both N and p1 · · · pk+1 . Therefore pi must divide
N p1 · · · pk+1 = 1, implying that pi  1, which is false. Hence p is a
prime other than p1 , . . . , pk+1 . Hence there are at least k + 2 primes.
Thus P(k + 1) is true. Hence by PMI, P(n) is true for all n 2 N. 62
A different starting point
Let b 2 N. To prove the statement: P(n) is true for every n b, the

only changes we need to make are:
• Base case is P(b) rather than P(1).

• In our induction hypothesis, we assume P(k) is true for k b,
rather than k 1.
63
A different starting point
Let b 2 N. To prove the statement: P(n) is true for every n b, the

only changes we need to make are:
• Base case is P(b) rather than P(1).

• In our induction hypothesis, we assume P(k) is true for k b,
rather than k 1.
Question
Prove that for every n 5, we have 2n > n2 .
63
Let P(n) be the statement ‘2n > n20 .
64
Base case: We verify that P(5) is true. In this case, 2n = 25 = 32, and
n2 = 25 and 25 > 52 . Hence P(5) is true.
64
is true, i.e., 2k > k 2 .
64
i.e., 2k+1 > (k + 1)2 .
64
i.e., 2k+1 > (k + 1)2 .
Now
2k+1 = 2 · 2k > 2k 2 (by Inductive hypothesis)

= k 2 + k 2 > k 2 + 2k + 1,
as k 2 2k = k(k 2) > 1 for k 5.
64
i.e., 2k+1 > (k + 1)2 .
Now
2k+1 = 2 · 2k > 2k 2 (by Inductive hypothesis)

= k 2 + k 2 > k 2 + 2k + 1,
as k 2 2k = k(k 2) > 1 for k 5. Hence

2k+1 > k 2 + 2k + 1 = (k + 1)2 . Thus P(k + 1) is true. Hence by PMI,
P(n) is true for all n 5.
64
Notation
Pn
The notation i=m Xi is called sum-
mation notation and it represents the
sum Xm + . . . + Xn .
• ⌃: upper case Greek letter
sigma
• i: index of summation (dummy
variable)
• m, n: lower and upper bounds of
summation, respectively
65
Notation
Pn
sum Xm + . . . + Xn .
sigma
variable)
Eg.
P7 2
• i=3 i = 32 + 42 + 52 + 62 + 72
Pn 1 1 1
• i=1 i 2 =1+ 4 + ... + n2
65
Notation
Pn
sum Xm + . . . + Xn .
sigma
variable)
Eg.
P7 2
• i=3 i = 32 + 42 + 52 + 62 + 72
Pn 1 1 1
• i=1 i 2 =1+ 4 + ... + n2
• sum of first n odd positive
integers:
65
Notation
Pn
sum Xm + . . . + Xn .
sigma
variable)
Eg.
P7 2
• i=3 i = 32 + 42 + 52 + 62 + 72
Pn 1 1 1
• i=1 i 2 =1+ 4 + ... + n2
Pn
integers: i=1 (2i 1)
65
Notation
Pn
sum Xm + . . . + Xn . Properties
• ⌃: upper case Greek letter • Multiplication by a constant:
Pn Pn
sigma i=m cXi =c i=m Xi , where
• i: index of summation (dummy c is a constant

variable) • Addition and subtraction of
• m, n: lower and upper bounds of two sums:
Pn Pn
summation, respectively i=m Xi ±
P i=m Yi
n
= i=m (Xi ± Yi )
Eg.
P7 2
• Changing the bounds of the
• i=3 i = 32 + 42 + 52 + 62 + 72
Pn 1 index of summation:
• =1+ 1
+ ... + 1 Pn Pn+k
i=1 i 2 4 n2 i=m Xi = i=m+k Xi k
Pn
integers: i=1 (2i 1)
65
Notation
Qn
The notation i=m Xi is called product notation and it represents the
product Xm Xm+1 · · · Xn .
Q
: upper case Greek letter pi
66
Notation
Qn
Q
Eg.
n ✓
Y ◆ ✓ ◆✓ ◆ ✓ ◆
1 1 1 1
1 = 1 1 ··· 1 .
i2 4 9 n2
i=2
66
Notation
Qn
Q
Eg.
n ✓
Y ◆ ✓ ◆✓ ◆ ✓ ◆
1 1 1 1
1 = 1 1 ··· 1 .
i2 4 9 n2
i=2
Question
Guess a closed form for this product and prove it by induction.
66
Qn 1 n+1
Let P(n) be the statement ‘ i=2 1 i2 = 2n .’
67
Qn
Let P(n) be the statement ‘ i=2 1 i12 = n+1
2n .’
Base case: We verify that P(2) is true.
LHS= 1 14 = 34 = RHS
67
Qn
Let P(n) be the statement ‘ i=2 1 i12 = n+1
2n .’
LHS= 1 14 = 34 = RHS
Qk
is true, i.e., i=2 1 i12 = k+1
2k .
67
Qn
Let P(n) be the statement ‘ i=2 1 i12 = n+1 2n .’
LHS= 1 14 = 34 = RHS
Qk
is true, i.e., i=2 1 i12 = k+1
2k .
Qk+1
i.e., i=2 1 i12 = 2(k+1)
k+2
.
67
Qn
LHS= 1 14 = 34 = RHS
Qk
is true, i.e., i=2 1 i12 = k+1 2k .
Qk+1
i.e., i=2 1 i12 = 2(k+1) k+2
.
Now
Y✓
k+1 ◆ Yk ✓ ◆! ✓ ◆
1 1 1
1 = 1 · 1
i2 i2 (k + 1)2
i=2 i=2
✓ ◆
k +1 1 (k + 1)2 1 k +2
= 1 2
= = .
2k (k + 1) 2k(k + 1) 2(k + 1)
67
Qn
LHS= 1 14 = 34 = RHS
Qk
is true, i.e., i=2 1 i12 = k+1 2k .
Qk+1
i.e., i=2 1 i12 = 2(k+1) k+2
.
Now
Y✓
k+1 ◆ Yk ✓ ◆! ✓ ◆
1 1 1
1 = 1 · 1
i2 i2 (k + 1)2
i=2 i=2
✓ ◆
k +1 1 (k + 1)2 1 k +2
= 1 2
= = .
2k (k + 1) 2k(k + 1) 2(k + 1)
Thus P(k + 1) is true. Hence by PMI, P(n) is true for all n 2.

⇣ ⌘
1
7 P(k + 1) = P(k) 1 (k+1) 2
67
——————–
68
Recursive definition of a function
Sometimes it is possible to define a function by one or more initial values,

and expressions involving values at prior inputs.
68

Examples.
• (Fibonacci sequence): F0 = F1 = 1 and Fn+1 = Fn + Fn 1 for n 1.
68

Examples.

• (Factorial function): 0! = 1 and (n + 1)! = (n + 1)n! for n 0.
68

Examples.

• (Factorial function): 0! = 1 and (n + 1)! = (n + 1)n! for n 0.
• The function f (n) = 2n + 1, n 2 N can also be defined as
f (0) = 1 and f (n + 1) = f (n) + 2 for n 0.
68
We say that a function f from the set of non-negative integers is defined

recursively if f (0) is given and for every n 2 N, f (n) is given in terms of
f (k)’s, 0  k < n.
The following theorem can be used to verify that certain recursively
defined functions are well-defined.
69
We say that a function f from the set of non-negative integers is defined

recursively if f (0) is given and for every n 2 N, f (n) is given in terms of
f (k)’s, 0  k < n.
The following theorem can be used to verify that certain recursively
defined functions are well-defined.
The Recursion Theorem
Let S be a set and let F : S ! S. Let s0 2 S be fixed. Then there is a
unique function f : N [ {0} ! S such that
1. f (0) = s0 (Initial condition)

2. f (n + 1) = F (f (n)) 8 n 2 N [ {0} (Recurrence relation or
generating rule).
69
Example (Ex II 1.)

Apply the Recursion Theorem to verify that the following recursive
definition defines a function.
g (0) = 1
g (n + 1) = 3g (n)2 + 7, for n 0.
70
Example (Ex II 1.)

Apply the Recursion Theorem to verify that the following recursive
definition defines a function.
g (0) = 1
g (n + 1) = 3g (n)2 + 7, for n 0.
Hint. Let s0 = 1 and let F (k) = 3k 2 + 7.
70
When Weak Induction fails
Question
Define a sequence (xn ) by
x1 = 4,
x2 = 68,
xn = 2xn 1 + 15xn 2 if n 3.
Show that
1
xn = 2 · ( 3)n + 10 · 5n for n 1.
71
When Weak Induction fails
Question
x1 = 4,
x2 = 68,
xn = 2xn 1 + 15xn 2 if n 3.
Show that
1
xn = 2 · ( 3)n + 10 · 5n for n 1.
Where does PMI fail?
71
The Principle of Strong Induction (PSI)
• P(1), . . . , P(b) are true for some b 2 N, and

• ‘P(1), . . . , P(k) is true implies P(k + 1) is true’ for all k b,
72
The Principle of Strong Induction (PSI)
• P(1), . . . , P(b) are true for some b 2 N, and

• ‘P(1), . . . , P(k) is true implies P(k + 1) is true’ for all k b,
A different starting point works analogous to PMI.
72
Question
x1 = 4,
x2 = 68,
xn+1 = 2xn + 15xn 1 if n 2.
Show that
1
xn = 2 · ( 3)n + 10 · 5n for n 1.
73
1
Let P(n) be the statement ‘xn = 2 · ( 3)n + 10 · 5n .’
74
1
Let P(n) be the statement ‘xn = 2 · ( 3)n + 10 · 5n .’
RHS=2·( 3)1 + 10 · 50 = 6 + 10 = 4 = LHS
We verify that P(2) is true.
RHS=2·( 3)2 + 10 · 51 = 18 + 50 = 68 = LHS
74
(Contd.)
Strong inductive hypothesis: Let k 2. We assume that the

1
statement P(i) is true for all 1  i  k, i.e., xi = 2 · ( 3)i + 10 · 5i for
all 1  i  k.
75
(Contd.)

statement P(i) is true for all 1  i  k, i.e., xi = 2 · ( 3)i + 10 · 5i 1 for
all 1  i  k.
i.e., xk+1 = 2 · ( 3)k+1 + 10 · 5k .
75
(Contd.)

all 1  i  k.
i.e., xk+1 = 2 · ( 3)k+1 + 10 · 5k .
Now
xk+1 = 2xk + 15xk 1

1 1 2
= 2(2 · ( 3) + 10 · 5k
k
) + 15(2 · ( 3)k + 10 · 5k )
(by Strong inductive hypothesis)
1 2
= ( 3)k (2 · 2 · ( 3) + 15 · 2) + 5k (2 · 10 · 5 + 15 · 10)
1 2
= ( 3)k · 18 + 5k · 250
k+1 k
= 2 · ( 3) + 10 · 5 .
75
(Contd.)

all 1  i  k.
i.e., xk+1 = 2 · ( 3)k+1 + 10 · 5k .
Now
xk+1 = 2xk + 15xk 1

1 1 2
= 2(2 · ( 3) + 10 · 5k
k
) + 15(2 · ( 3)k + 10 · 5k )
(by Strong inductive hypothesis)
1 2
= ( 3)k (2 · 2 · ( 3) + 15 · 2) + 5k (2 · 10 · 5 + 15 · 10)
1 2
= ( 3)k · 18 + 5k · 250
k+1 k
= 2 · ( 3) + 10 · 5 .
Thus P(k + 1) is true. Hence by PSI, P(n) is true for all n 1.

75
Remark
Let b 2 N. Suppose we wish to prove a statement of the type
8n b, P(n). It may be the case that we are able to prove the
implications P(k) ! P(k + 1) for all k b 0 , where b 0 > b. Then we
will need to verify all of the cases P(b), P(b + 1), . . . , P(b 0 ) as ’base
cases’, instead of just P(b).
76

DM Module2

Uploaded by

Copyright:

Available Formats

DM Module2

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

DM Module2

Uploaded by

Copyright:

Available Formats

MATH F213: Discrete Mathematics

First Semester 2021-22

In this section, we will learn

1. nine common methods for proving implications

If q is true, then the implication p ! q is true irrespective of the truth

If q is true, then the implication p ! q is true irrespective of the truth

If q is true, then the implication p ! q is true irrespective of the truth

Solution. Since x 2 0 for all real numbers x, we get

If p is false, then the implication p ! q is true irrespective of the truth

If p is false, then the implication p ! q is true irrespective of the truth

If p is false, then the implication p ! q is true irrespective of the truth

A direct proof of the implication p ! q proceeds as follows.

A direct proof of the implication p ! q proceeds as follows.

Example. (Ex. 2(h))

A direct proof of the implication p ! q proceeds as follows.

Example. (Ex. 2(h))

Solution. Since a divides both b and c, we get that

A direct proof of the implication p ! q proceeds as follows.

Example. (Ex. 2(h))

Solution. Since a divides both b and c, we get that

A direct proof of the implication p ! q proceeds as follows.

Example. (Ex. 2(h))

Solution. Since a divides both b and c, we get that

An indirect proof of the implication p ! q is a direct proof of its

An indirect proof of the implication p ! q is a direct proof of its

An indirect proof of the implication p ! q is a direct proof of its

Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).)

An indirect proof of the implication p ! q is a direct proof of its

Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).) Assume that

An indirect proof of the implication p ! q is a direct proof of its

Solution. (Recall that ⇠ (q _ r ) ⌘ (⇠ q) ^ (⇠ r ).) Assume that

Solution. Assume that p is composite.

Solution. Assume that p is composite. Then, we can write

p = ab, 1 < a, b < p.

Solution. Assume that p is composite. Then, we can write

p = ab, 1 < a, b < p.

Since a divides p and p divides n, we get that a divides n.

Solution. Assume that p is composite. Then, we can write

p = ab, 1 < a, b < p.

Since a divides p and p divides n, we get that a divides n. But

Recall that p ! q is true if and only if p ^ (⇠ q) is false. A proof of the

1. Assume that p ^ (⇠ q) is true.

Proof. Let n denote the number of pigeonholes. For i = 1, . . . , n, let mi

Proof. Let n denote the number of pigeonholes. For i = 1, . . . , n, let mi

Proof. Let n denote the number of pigeonholes. For i = 1, . . . , n, let mi

Proof. Let n denote the number of pigeonholes. For i = 1, . . . , n, let mi

• N denotes the set of positive integers:

• Z denotes the set of integers:

• Let k 2 N. If kn + 1 pigeons are distributed in n pigeonholes, then

• Let k 2 N. If kn + 1 pigeons are distributed in n pigeonholes, then

Remark. In particular, taking k = 1, we get that if n + 1 pigeons are

• Let k 2 N. If kn + 1 pigeons are distributed in n pigeonholes, then

Remark. In particular, taking k = 1, we get that if n + 1 pigeons are

Recall: The Division Algorithm

Solution. (1) Let a1 , a2 , ..., ab+1 denote the numbers. By Division

Solution. (1) Let a1 , a2 , ..., ab+1 denote the numbers. By Division

Solution. (1) Let a1 , a2 , ..., ab+1 denote the numbers. By Division

Solution. (1) Let a1 , a2 , ..., ab+1 denote the numbers. By Division

If p is of the form p1 _ p2 _ · · · _ pn , we may prove the implication p ! q

If p is of the form p1 _ p2 _ · · · _ pn , we may prove the implication p ! q