0% found this document useful (0 votes)
267 views

Discrete Mathematics 1.6 - 1.7 Solution

The document discusses logic and rules of inference. It introduces existential and universal quantifiers, and defines common rules of inference like modus ponens, modus tollens, disjunctive syllogism and hypothetical syllogism. It also provides examples of applying these rules of inference to prove conclusions from given premises in logical arguments.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
267 views

Discrete Mathematics 1.6 - 1.7 Solution

The document discusses logic and rules of inference. It introduces existential and universal quantifiers, and defines common rules of inference like modus ponens, modus tollens, disjunctive syllogism and hypothetical syllogism. It also provides examples of applying these rules of inference to prove conclusions from given premises in logical arguments.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

26 Chapter 1 The Foundations: Logic and Proofs

52. We simply want to say that there exists an x such that P (x) holds, and that every y such that P (y) holds

must be this same x . Thus we write ∃x P (x) ∧ ∀y(P (y) → y = x) . Even more compactly, we can write
∃x∀y(P (y) ↔ y = x) .

SECTION 1.6 Rules of Inference


2. This is modus tollens. The first statement is p → q , where p is “George does not have eight legs” and q
is “George is not a spider.” The second statement is ¬q . The third is ¬p . Modus tollens is valid. We can
therefore conclude that the conclusion of the argument (third statement) is true, given that the hypotheses
(the first two statements) are true.

4. a) We have taken the conjunction of two propositions and asserted one of them. This is, according to Table 1,
simplification.
b) We have taken the disjunction of two propositions and the negation of one of them, and asserted the other.
This is, according to Table 1, disjunctive syllogism. See Table 1 for the other parts of this exercise as well.
c) modus ponens d) addition e) hypothetical syllogism

6. Let r be the proposition “It rains,” let f be the proposition “It is foggy,” let s be the proposition “The
sailing race will be held,” let l be the proposition “The life saving demonstration will go on,” and let t be the
proposition “The trophy will be awarded.” We are given premises (¬r ∨ ¬f ) → (s ∧ l) , s → t , and ¬t . We
want to conclude r . We set up the proof in two columns, with reasons, as in Example 6. Note that it is valid
to replace subexpressions by other expressions logically equivalent to them.

Step Reason
1. ¬t Hypothesis
2. s → t Hypothesis
3. ¬s Modus tollens using (1) and (2)
4. (¬r ∨ ¬f ) → (s ∧ l) Hypothesis
5. (¬(s ∧ l)) → ¬(¬r ∨ ¬f ) Contrapositive of (4)
6. (¬s ∨ ¬l) → (r ∧ f ) De Morgan’s law and double negative
7. ¬s ∨ ¬l Addition, using (3)
8. r ∧ f Modus ponens using (6) and (7)
9. r Simplification using (8)

8. First we use universal instantiation to conclude from “For all x , if x is a man, then x is not an island”
the special case of interest, “If Manhattan is a man, then Manhattan is not an island.” Then we form the
contrapositive (using also double negative): “If Manhattan is an island, then Manhattan is not a man.” Finally
we use modus ponens to conclude that Manhattan is not a man. Alternatively, we could apply modus tollens.

10. a) If we use modus tollens starting from the back, then we conclude that I am not sore. Another application
of modus tollens then tells us that I did not play hockey.
b) We really can’t conclude anything specific here.
c) By universal instantiation, we conclude from the first conditional statement by modus ponens that dragon-
flies have six legs, and we conclude by modus tollens that spiders are not insects. We could say using existential
generalization that, for example, there exists a non-six-legged creature that eats a six-legged creature, and
that there exists a non-insect that eats an insect.
d) We can apply universal instantiation to the conditional statement and conclude that if Homer (respectively,
Maggie) is a student, then he (she) has an Internet account. Now modus tollens tells us that Homer is not a
student. There are no conclusions to be drawn about Maggie.
Section 1.6 Rules of Inference 27

e) The first conditional statement is that if x is healthy to eat, then x does not taste good. Universal
instantiation and modus ponens therefore tell us that tofu does not taste good. The third sentence says that
if you eat x , then x tastes good. Therefore the fourth hypothesis already follows (by modus tollens) from the
first three. No conclusions can be drawn about cheeseburgers from these statements.
f ) By disjunctive syllogism, the first two hypotheses allow us to conclude that I am hallucinating. Therefore
by modus ponens we know that I see elephants running down the road.

12. Applying Exercise 11, we want to show that the conclusion r follows from the five premises (p ∧ t) → (r ∨ s) ,
q → (u ∧ t) , u → p, ¬s , and q . From q and q → (u ∧ t) we get u ∧ t by modus ponens. From there we get
both u and t by simplification (and the commutative law). From u and u → p we get p by modus ponens.
From p and t we get p ∧ t by conjunction. From that and (p ∧ t) → (r ∨ s) we get r ∨ s by modus ponens.
From that and ¬s we finally get r by disjunctive syllogism.

14. In each case we set up the proof in two columns, with reasons, as in Example 6.
a) Let c(x) be “x is in this class,” let r(x) be “x owns a red convertible,” and let t(x) be “x has gotten
a speeding ticket.” We are given premises c(Linda) , r(Linda) , ∀x(r(x) → t(x)) , and we want to conclude
∃x(c(x) ∧ t(x)) .
Step Reason
1. ∀x(r(x) → t(x)) Hypothesis
2. r(Linda) → t(Linda) Universal instantiation using (1)
3. r(Linda) Hypothesis
4. t(Linda) Modus ponens using (2) and (3)
5. c(Linda) Hypothesis
6. c(Linda) ∧ t(Linda) Conjunction using (4) and (5)
7. ∃x(c(x) ∧ t(x)) Existential generalization using (6)
b) Let r(x) be “r is one of the five roommates listed,” let d(x) be “x has taken a course in discrete
mathematics,” and let a(x) be “ x can take a course in algorithms.” We are given premises ∀x(r(x) → d(x))
and ∀x(d(x) → a(x)) , and we want to conclude ∀x(r(x) → a(x)) . In what follows y represents an arbitrary
person.
Step Reason
1. ∀x(r(x) → d(x)) Hypothesis
2. r(y) → d(y) Universal instantiation using (1)
3. ∀x(d(x) → a(x)) Hypothesis
4. d(y) → a(y) Universal instantiation using (3)
5. r(y) → a(y) Hypothetical syllogism using (2) and (4)
6. ∀x(r(x) → a(x)) Universal generalization using (5)
c) Let s(x) be “x is a movie produced by Sayles,” let c(x) be “x is a movie about coal miners,” and let
w(x) be “movie x is wonderful.” We are given premises ∀x(s(x) → w(x)) and ∃x(s(x) ∧ c(x)) , and we want
to conclude ∃x(c(x) ∧ w(x)) . In our proof, y represents an unspecified particular movie.
Step Reason
1. ∃x(s(x) ∧ c(x)) Hypothesis
2. s(y) ∧ c(y) Existential instantiation using (1)
3. s(y) Simplification using (2)
4. ∀x(s(x) → w(x)) Hypothesis
5. s(y) → w(y) Universal instantiation using (4)
6. w(y) Modus ponens using (3) and (5)
7. c(y) Simplification using (2)
8. w(y) ∧ c(y) Conjunction using (6) and (7)
9. ∃x(c(x) ∧ w(x)) Existential generalization using (8)
28 Chapter 1 The Foundations: Logic and Proofs

d) Let c(x) be “ x is in this class,” let f (x) be “ x has been to France,” and let l(x) be “ x has visited the
Louvre.” We are given premises ∃x(c(x) ∧ f (x)) , ∀x(f (x) → l(x)) , and we want to conclude ∃x(c(x) ∧ l(x)) .
In our proof, y represents an unspecified particular person.
Step Reason
1. ∃x(c(x) ∧ f (x)) Hypothesis
2. c(y) ∧ f (y) Existential instantiation using (1)
3. f (y) Simplification using (2)
4. c(y) Simplification using (2)
5. ∀x(f (x) → l(x)) Hypothesis
6. f (y) → l(y) Universal instantiation using (5)
7. l(y) Modus ponens using (3) and (6)
8. c(y) ∧ l(y) Conjunction using (4) and (7)
9. ∃x(c(x) ∧ l(x)) Existential generalization using (8)

16. a) This is correct, using universal instantiation and modus tollens.


b) This is not correct. After applying universal instantiation, it contains the fallacy of denying the hypothesis.
c) After applying universal instantiation, it contains the fallacy of affirming the conclusion.
d) This is correct, using universal instantiation and modus ponens.

18. We know that some s exists that makes S(s, Max) true, but we cannot conclude that Max is one such s .
Therefore this first step is invalid.

20. a) This is invalid. It is the fallacy of affirming the conclusion. Letting a = −2 provides a counterexample.
b) This is valid; it is modus ponens.

22. We will give an argument establishing the conclusion. We want to show that all hummingbirds are small. Let
Tweety be an arbitrary hummingbird. We must show that Tweety is small. The first premise implies that
if Tweety is a hummingbird, then Tweety is richly colored. Therefore by (universal) modus ponens we can
conclude that Tweety is richly colored. The third premise implies that if Tweety does not live on honey, then
Tweety is not richly colored. Therefore by (universal) modus tollens we can now conclude that Tweety does
live on honey. Finally, the second premise implies that if Tweety is a large bird, then Tweety does not live
on honey. Therefore again by (universal) modus tollens we can now conclude that Tweety is not a large bird,
i.e., that Tweety is small, as desired. Notice that we invoke universal generalization as the last step.

24. Steps 3 and 5 are incorrect; simplification applies to conjunctions, not disjunctions.

26. We want to show that the conditional statement P (a) → R(a) is true for all a in the domain; the desired
conclusion then follows by universal generalization. Thus we want to show that if P (a) is true for a particu-
lar a, then R(a) is also true. For such an a, by universal modus ponens from the first premise we have Q(a) ,
and then by universal modus ponens from the second premise we have R(a) , as desired.

28. We want to show that the conditional statement ¬R(a) → P (a) is true for all a in the domain; the desired
conclusion then follows by universal generalization. Thus we want to show that if ¬R(a) is true for a partic-
ular a, then P (a) is also true. For such an a, universal modus tollens applied to the second premise gives us
¬(¬P (a) ∧ Q(a)) . By rules from propositional logic, this gives us P (a) ∨ ¬Q(a) . By universal generalization
from the first premise, we have P (a)∨Q(a) . Now by resolution we can conclude P (a)∨P (a) , which is logically
equivalent to P (a) , as desired.

30. Let a be “Allen is a good boy,” let h be “Hillary is a good girl,” and let d be “David is happy.” Then our
assumptions are ¬a ∨ h and a ∨ d . Using resolution gives us h ∨ d, as desired.
Section 1.7 Introduction to Proofs 29

32. We apply resolution to give the tautology (p ∨ F) ∧ (¬p ∨ F) → (F ∨ F) . The left-hand side is equivalent to
p ∧ ¬p, since p ∨ F is equivalent to p , and ¬p ∨ F is equivalent to ¬p . The right-hand side is equivalent to F .
Since the conditional statement is true, and the conclusion is false, it follows that the hypothesis, p ∧ ¬p , is
false, as desired.

34. Let us use the following letters to stand for the relevant propositions: d for “logic is difficult,” s for “many
students like logic,” and e for “mathematics is easy.” Then the assumptions are d ∨ ¬s and e → ¬d. Note
that the first of these is equivalent to s → d , since both forms are false if and only if s is true and d is false.
In addition, let us note that the second assumption is equivalent to its contrapositive, d → ¬e . And finally,
by combining these two conditional statements, we see that s → ¬e also follows from our assumptions.
a) Here we are asked whether we can conclude that s → ¬e . As we noted above, the answer is yes, this
conclusion is valid.
b) The question concerns ¬e → ¬s. This is equivalent to its contrapositive, s → e . That doesn’t seem to
follow from our assumptions, so let’s find a case in which the assumptions hold but this conditional statement
does not. This conditional statement fails in the case in which s is true and e is false. If we take d to be true
as well, then both of our assumptions are true. Therefore this conclusion is not valid.
c) The issue is ¬e ∨ d, which is equivalent to the conditional statement e → d . This does not follow from
our assumptions. If we take d to be false, e to be true, and s to be false, then this proposition is false but
our assumptions are true.
d) The issue is ¬d ∨ ¬e , which is equivalent to the conditional statement d → ¬e . We noted above that this
validly follows from our assumptions.
e) This sentence says ¬s → (¬e ∨ ¬d) . The only case in which this is false is when s is false and both e
and d are true. But in this case, our assumption e → ¬d is also violated. Therefore, in all cases in which the
assumptions hold, this statement holds as well, so it is a valid conclusion.

SECTION 1.7 Introduction to Proofs


2. We must show that whenever we have two even integers, their sum is even. Suppose that a and b are
two even integers. Then there exist integers s and t such that a = 2s and b = 2t . Adding, we obtain
a + b = 2s + 2t = 2(s + t) . Since this represents a + b as 2 times the integer s + t , we conclude that a + b is
even, as desired.

4. We must show that whenever we have an even integer, its negative is even. Suppose that a is an even integer.
Then there exists an integer s such that a = 2s . Its additive inverse is −2s , which by rules of arithmetic and
algebra (see Appendix 1) equals 2(−s) . Since this is 2 times the integer −s, it is even, as desired.

6. An odd number is one of the form 2n + 1 , where n is an integer. We are given two odd numbers, say 2a + 1
and 2b + 1 . Their product is (2a + 1)(2b + 1) = 4ab + 2a + 2b + 1 = 2(2ab + a + b) + 1 . This last expression
shows that the product is odd, since it is of the form 2n + 1 , with n = 2ab + a + b.

8. Let n = m2 . If m = 0 , then n + 2 = 2 , which is not a perfect square, so we can assume that m ≥ 1 . The
smallest perfect square greater than n is (m + 1)2 , and we have (m + 1)2 = m2 + 2m + 1 = n + 2m + 1 >
n + 2 · 1 + 1 > n + 2 . Therefore n + 2 cannot be a perfect square.

10. A rational number is a number that can be written in the form x/y where x and y are integers and y 6= 0 .
Suppose that we have two rational numbers, say a/b and c/d . Then their product is, by the usual rules for
multiplication of fractions, (ac)/(bd) . Note that both the numerator and the denominator are integers, and
that bd 6= 0 since b and d were both nonzero. Therefore the product is, by definition, a rational number.
30 Chapter 1 The Foundations: Logic and Proofs

12. This is true. Suppose that a/b is a nonzero rational number and that x is an irrational number. We must
prove that the product xa/b is also irrational. We give a proof by contradiction. Suppose that xa/b were
rational. Since a/b 6= 0 , we know that a 6= 0 , so b/a is also a rational number. Let us multiply this rational
number b/a by the assumed rational number xa/b. By Exercise 10, the product is rational. But the product is
(b/a)(xa/b) = x , which is irrational by hypothesis. This is a contradiction, so in fact xa/b must be irrational,
as desired.

14. If x is rational and not zero, then by definition we can write x = p/q , where p and q are nonzero integers.
Since 1/x is then q/p and p 6= 0 , we can conclude that 1/x is rational.

16. Assume to the contrary that x , y , and z are all even. Then there exist integers a, b, and c such that x = 2a ,
y = 2b, and z = 2c. But then x + y + z = 2a + 2b + 2c = 2(a + b + c) is even by definition. This contradicts
the hypothesis that x + y + z is odd. Therefore the assumption was wrong, and at least one of x , y , and z
is odd.

18. We give a proof by contraposition. If it is not true than m is even or n is even, then m and n are both odd.
By Exercise 6, this tells us that mn is odd, and our proof is complete.

20. a) We must prove the contrapositive: If n is odd, then 3n + 2 is odd. Assume that n is odd. Then we can
write n = 2k + 1 for some integer k . Then 3n + 2 = 3(2k + 1) + 2 = 6k + 5 = 2(3k + 2) + 1 . Thus 3n + 2 is
two times some integer plus 1, so it is odd.
b) Suppose that 3n + 2 is even and that n is odd. Since 3n + 2 is even, so is 3n . If we add subtract an odd
number from an even number, we get an odd number, so 3n − n = 2n is odd. But this is obviously not true.
Therefore our supposition was wrong, and the proof by contradiction is complete.

22. We need to prove the proposition “If 1 is a positive integer, then 12 ≥ 1 .” The conclusion is the true statement
1 ≥ 1 . Therefore the conditional statement is true. This is an example of a trivial proof, since we merely
showed that the conclusion was true.

24. We give a proof by contradiction. Suppose that we don’t get a pair of blue socks or a pair of black socks.
Then we drew at most one of each color. This accounts for only two socks. But we are drawing three socks.
Therefore our supposition that we did not get a pair of blue socks or a pair of black socks is incorrect, and
our proof is complete.

26. We give a proof by contradiction. If there were at most two days falling in the same month, then we could
have at most 2 · 12 = 24 days, since there are 12 months. Since we have chosen 25 days, at least three of
them must fall in the same month.

28. We need to prove two things, since this is an “if and only if” statement. First let us prove directly that
if n is even then 7n + 4 is even. Since n is even, it can be written as 2k for some integer k . Then
7n + 4 = 14k + 4 = 2(7k + 2) . This is 2 times an integer, so it is even, as desired. Next we give a proof by
contraposition that if 7n + 4 is even then n is even. So suppose that n is not even, i.e., that n is odd. Then
n can be written as 2k + 1 for some integer k . Thus 7n + 4 = 14k + 11 = 2(7k + 5) + 1 . This is 1 more than
2 times an integer, so it is odd. That completes the proof by contraposition.

30. There are two things to prove. For the “if” part, there are two cases. If m = n , then of course m2 = n2 ;
if m = −n , then m2 = (−n)2 = (−1)2 n2 = n2 . For the “only if” part, we suppose that m2 = n2 . Putting
everything on the left and factoring, we have (m + n)(m − n) = 0 . Now the only way that a product of two
numbers can be zero is if one of them is zero. Therefore we conclude that either m + n = 0 (in which case
m = −n ), or else m − n = 0 (in which case m = n ), and our proof is complete.
Section 1.7 Introduction to Proofs 31

32. We write these in symbols: a < b , (a + b)/2 > a , and (a + b)/2 < b . The latter two are equivalent to
a + b > 2a and a + b < 2b, respectively, and these are in turn equivalent to b > a and a < b , respectively. It
is now clear that all three statements are equivalent.

34. We give direct proofs that (i ) implies (ii ), that (ii ) implies (iii ), and that (iii ) implies (i ). That will suffice.
For the first, suppose that x = p/q where p and q are integers with q 6= 0 . Then x/2 = p/(2q) , and this is
rational, since p and 2q are integers with 2q 6= 0 . For the second, suppose that x/2 = p/q where p and q
are integers with q 6= 0 . Then x = (2p)/q , so 3x − 1 = (6p)/q − 1 = (6p − q)/q and this is rational, since
6p − q and q are integers with q 6= 0 . For the last, suppose that 3x − 1 = p/q where p and q are integers
with q 6= 0 . Then x = (p/q + 1)/3 = (p + q)/(3q) , and this is rational, since p + q and 3q are integers with
3q 6= 0 .

36. No. This line of reasoning shows that if 2x2 − 1 = x , then we must have x = 1 or x = −1 . These are
therefore the only possible solutions, but we have no guarantee that they are solutions, since not all of our
steps were reversible (in particular, squaring both sides). Therefore we must substitute these values back into
the original equation to determine whether they do indeed satisfy it.

38. The only conditional statements not shown directly are p1 ↔ p2 , p2 ↔ p4 , and p3 ↔ p4 . But these each
follow with one or more intermediate steps: p1 ↔ p2 , since p1 ↔ p3 and p3 ↔ p2 ; p2 ↔ p4 , since p2 ↔ p1
(just established) and p1 ↔ p4 ; and p3 ↔ p4 , since p3 ↔ p1 and p1 ↔ p4 .

40. We must find a number that cannot be written as the sum of the squares of three integers. We claim that 7
is such a number (in fact, it is the smallest such number). The only squares that can be used to contribute
to the sum are 0 , 1 , and 4 . We cannot use two 4’s , because their sum exceeds 7 . Therefore we can use at
most one 4, which means that we must get 3 using just 0’s and 1’s . Clearly three 1’s are required for this,
bringing the total number of squares used to four. Thus 7 cannot be written as the sum of three squares.

42. Suppose that we look at the ten groups of integers in three consecutive locations around the circle (first-
second-third, second-third-fourth, . . . , eighth-ninth-tenth, ninth-tenth-first, and tenth-first-second). Since
each number from 1 to 10 gets used three times in these groups, the sum of the sums of the ten groups must
equal three times the sum of the numbers from 1 to 10 , namely 3 · 55 = 165 . Therefore the average sum is
165/10 = 16.5 . By Exercise 41, at least one of the sums must be greater than or equal to 16.5 , and since the
sums are whole numbers, this means that at least one of the sums must be greater than or equal to 17 .

44. We show that each of these is equivalent to the statement (v ) n is odd, say n = 2k +1 . Example 1 showed that
(v ) implies (i ), and Example 9 showed that (i ) implies (v ). For (v ) → (ii ) we see that 1 − n = 1 − (2k + 1) =
2(−k) is even. Conversely, if n were even, say n = 2m , then we would have 1 − n = 1 − 2m = 2(−m) + 1 , so
1 − n would be odd, and this completes the proof by contraposition that (ii ) → (v ). For (v ) → (iii ), we see
that n3 = (2k+1)3 = 8k 3 +12k 2 +6k+1 = 2(4k 3 +6k 2 +3k)+1 is odd. Conversely, if n were even, say n = 2m,
then we would have n3 = 2(4m3 ) , so n3 would be even, and this completes the proof by contraposition that
(iii ) → (v ). Finally, for (v ) → (iv ), we see that n2 + 1 = (2k + 1)2 + 1 = 4k 2 + 4k + 2 = 2(2k 2 + 2k + 1) is
even. Conversely, if n were even, say n = 2m , then we would have n2 + 1 = 2(2m2 ) + 1 , so n2 + 1 would be
odd, and this completes the proof by contraposition that (iv ) → (v ).

You might also like