Chapter 1 Logic Induction and Reasoning Part 2
Chapter 1 Logic Induction and Reasoning Part 2
ANUJ GHIMIRE
An argument used to establish the truth of a
mathematical statement is called proof.
While establishing the truth different rules and
already proven facts are used.
Proof can be divided into
◦ Formal Proof
◦ Informal Proof
Formal proof is the technique where predefined
rules and theorems are used to show that the
given statement is true, but in Informal proof
such predefined rules and theorem may not be
used.
Informal proof can be further classified into:
◦ Direct Proof
◦ Indirect Proof
Direct Proof:
◦ Let p→q be an implication, in direct proof we
assume that hypothesis is true i.e p is true then by
using different theorem and already proven facts
we conclude that the conclusion is also true i.e. q is
true.
◦ The basic idea of direct proof is whenever the
hypothesis is true conclusion is true.
Show that if n is odd then n2 is odd using direct
proof.
◦ p: n is odd
◦ q: n2 is odd.
◦ p→q
In direct proof we assume that hypothesis is true
i.e n is odd is true.
By definition of odd number:
n=2k+1 [where k=0,1,2,3……]
Squaring both side we get
n2=(2k+1)2
n2=4k2+4k+1
Here 4k2 and 4k are divisible by 2, their sum is also divisible
by 2, the addition of 1 makes it odd i.e. n2 is odd
Here we can conclude that the assumption of n
to be odd derives n2 as odd.
So, whenever n is odd, n2 is odd is proved
using direct proof method.
Proof using direct method that the sum of two
rational number are rational.
Solution:
◦ Let r and s be two rational number
◦ p: r and s are rational.
◦ q: (r+s) is rational.
◦ p→q
In direct proof we assume that hypothesis is
true i.e r and s to be rational number is true.
By definition of rational number:
r = a/b for some integers a and b with b ≠ 0.
s = c/d for some integers c and d with d ≠ 0.
So
r + s = a/b + c/d
= (ad + bc)/bd
Now, let p = ad + bc and q = bd. Then, p and q are integers
[because products and sums of integers are integers and
because a, b, c and d are all integers.] Also, q ≠ 0.
Hence, r + s = p/q , where p and q are integers and q ≠ 0.
So, we can conclude that the assumption of r and s to be rational
derives (r+s) as rational number.
This concludes whenever two rational number are added the
result will also be a rational number using direct proof method.
Show that if (3n+2) is odd then n is odd using
direct proof.
Solution:
◦ p: (3n+2) is odd .
◦ q: n is odd.
◦ p→q
By definition of odd number,
3n+2=2k+1
n =(2k-1)/3
No further process can be applied to continue the
proof. This is called the dead end of the proof.
To make a conclusion we need indirect proof
approach.
For certain case, the direct proof may not be
appropriate. For eg. “If 3n+2 is odd then n is
odd.”
Here using direct proof we may not reach to
the conclusion, such problem is considered
as dead end of proof.
To overcome such problem we use indirect
proof approach.
Two types of indirect proof:
◦ Proof by Contrapositive.
◦ Proof by Contradiction.
Let p→q be an implication, in proof by contrapositive
we assume that negation of conclusion is true i.e ¬q
is true
Then by using different theorem and already proven
facts we conclude that the negation of hypothesis is
also true i.e. ¬p is true.
The idea behind the proof by contrapositive is the
negation of the conclusion leads to the negation of
hypothesis.
It is because implication and its contrapositive are
logically equivalent.
i.e. p→q≡¬q →¬p
Show that if (3n+2) is odd then n is odd using
proof by contrapositive.
Solution:
p: (3n+2) is odd.
q: n is odd.
p→q
¬p: (3n+2) is even.
¬q: n is even
In proof by contrapositive we assume the negation of
conclusion to be true. i.e. ¬q is true. So here we
consider n to be even.
By definition of even number:
n=2k
3n=6k
3n+2=6k+2
3n+2=2(3k+1)
Here 2(3k+1) is even i.e 3n+2 is even.
So ¬q leads to ¬p
¬q →¬p ≡ p→q
The negation of conclusion leads to the negation of
hypothesis so we can say that using logical equivalent
property the implication statement is also true.
Hence the statement “if (3n+2) is odd then n is odd.” is
true using proof by contrapositive.
Proof by contradiction is another approach of
indirect proof.
For proof by contradiction following situation
may arise:
◦ The given statement is implication.
◦ The given statement is not implication.
The given statement is implication:
For implication statement (p→q), we assume
that the negation of conclusion and
hypothesis is true. i.e p∧¬q is true.
Then by using different theorem and already
proven facts, we try to prove that p∧¬q is
false.
And if p∧¬q is false then its negation is true
i.e. ¬ (p∧¬q) is true.
This is the contradiction to our assumption.
Hence we can say that our assumption is
wrong and given statement is true.
Finally we can conclude (¬ p∨q) is true i.e.
P→Q is true.
The given statement is not implication:
For this we assume that the negation of given
statement is true
Then by using different theorem and already
proven facts, we reach to the point that
contradicts our assumption.
Hence we can say that our assumption is
wrong and given statement is true.
Show that √2 is irrational.
Exhaustive Proof
Show that 𝑝 = 𝑛2 + 2 is not a multiple of
4, where n is an integer as 2 ≤ 𝑛 ≤ 7
Soundness and Completeness of
System
Valid Formula
Valid formulae are those formulae which are valid under
all valuations
P∨¬P is valid formula.
Valid propositional formulae are also called tautologies.
Soundness and Completeness of
System
Satisfiable Formula
These can be made true under at least one valuation,
though not necessarily all valuations.
An example of satisfiable formula is P ∨ Q which is true
at P= T and Q=F.
Every valid formula is satisfiable.
Unsatisfiable Formula
The unsatisfiable formulas come out false under all
valuations.
Soundness and Completeness of
System
The axioms are all valid and each inference rule has
the following properties: If both premises are valid
then the conclusion is also valid.
It follows that, only valid formulae can be proved in a
sound logical system.
If φ can be proved in a sound logical system then φ is
valid.
If there is a proof of φ then φ is called a theorem.
Soundness and Completeness of
System
A proof system is sound if everything that is provable
is actually true.
Propositional logic is sound if when we use deduction
rules to prove that (P1, P2, ..., Pn) ⊢ (C) (that a set of
premises proves a conclusion) then we can also use a
truth table to show that P1, P2, ..., Pn ⊨ C (that a set of
premises semantically entails a conclusion).
Soundness and Completeness of
System
A proof system is complete if everything that is true
can be proved.
Propositional logic is complete if when we can use a
truth table to show that P1, P2, ..., Pn ⊨ C, then we can
also use deduction rules to prove that (P1, P2, ..., Pn) ⊢
(C).
Soundness and Completeness of
System
Propositional logic is also complete.
We assume that P1, P2, ..., Pn ⊨ C, and we consider the
truth table for (P1 ∧ P2 ∧ ... ∧ Pn) → C (since that will
be a tautology whenever P1, P2, ..., Pn ⊨ C).
In order to show propositional logic is complete, we
must show that we can use our deduction rules to
prove (P1, P2, ..., Pn) ⊢ (C).
Soundness and Completeness of
System
The soundness of a logical system is expressed as
If φ is a theorem (if φ can be proved), then φ is
valid.
The completeness of the system is expressed as
If φ is valid then φ is a theorem (φ can be proved)
Method of Tableaux
Tableaux is a graphical method to test the validity of
an argument.
It can also be used to prove the satisfiability
(consistency) or unsatisfiability (inconsistency) of a
set of a set of formula.
Suppose we have a finite set of formulas
1 , 2 , 3 ,........., m
Then, a tableau for 1 , 2 , 3 ,........., m is a tree
like structure formed by using the derivation rules for
each i
Method of Tableaux
This method in a propositional logic is the method in
which set of rules are applied symmetrically in a
formula or a set of formulae to check for its
consistency.
The main principle is to break down the formula or set
of formulas into its component up to smaller ones
until the complementary pair of literals is obtained or
no further expansion is possible.
Method of Tableaux
Let φ is a formula then we need to parse that φ in the
tree structure.
If a formula φ is not a literal, it has one of the form as:
φ = ¬¬B φ = (B → C)
φ=B∧C φ = ¬ (B → C)
φ = ¬ (B ^ C) φ =(B ↔ C)
φ =(B ∨ C) φ = ¬ (B ↔ C)
φ = ¬ (B ∨ C)
The parsing can be done in the following ways
according to propositional representation:
Method of Tableaux
φ = (B → C)
≡ ¬B ∨ C
φ = ¬ (B → C)
≡ ¬(¬B ∨ C)
≡ ¬¬B ∧ ¬C
≡ B ∧ ¬C
Method of Tableaux
φ =(B ↔ C)
≡ (B → C) ∧ (B → C)
≡ (¬B ∨ C) ∧ (¬C ∨ B)
≡ (¬B ∧¬C) ∨ (¬B ∧ B) ∨ (C ∧ ¬C )∨(C ∧ B)
≡ (¬B ∧¬C) ∨ (C ∧ B)
φ = ¬(B ↔ C)
≡ ¬{(B → C) ∧(B → C)}
≡ ¬{(¬B ∨ C)∧ (¬C ∨ B)}
≡ (¬¬B ∧ ¬C) ∨(¬¬C ∧ ¬B)
≡ (B ∧ ¬C) ∨ (C ∧ ¬B)
Method of Tableaux
¬¬B B∧C ¬ (B ∧ C) ≡ ¬B ∨ ¬ C
B ¬B ¬C
B
C
B∨C ¬ (B ∨ C) ≡ ¬B ∧ ¬ C
B C ¬B
¬C
Method of Tableaux
(B → C) ≡ ¬B ∨ C ¬ (B → C) ≡ B ∧ ¬ C
¬B C B
¬C
(B ↔ C)
¬(B ↔ C)
B ¬B
C ¬C B ¬B
¬C C
Method of Tableaux
Steps to draw tableaux of any set of formula:
Step_1: List the set of formulas in columnar form
Step_2: If φi is not the literal, apply the appropriate
rule.
Step_3: After each step, we check for satisfiability along
branches.
Step_4: If a literal p and ¬p appear on the same path of
the tableaux, then the branch not satisfiable i.e.
inconsistency is maintained and such path is said to be
contradictory or close
Step_5: Even if one path remains non-contradictory or
open then the formula φ at the root of tableau is
consistent.
Method of Tableaux
Note
If only set of formula is given (i.e. conclusion is not
given) and if any branch is open, then the set is
satisfiable sometimes called soundness.
If the argument (i.e. premises and conclusion) are given,
we negate the conclusion.
Then if the all branches of tree are closed (made up of
premises and negation of conclusion) then it is valid
called completeness.
Method of Tableaux (Example_1)
Draw the tableau for the following set of formula
{a c, (a b) (b c)}
and check for unsatisfiability or satisfiability.
Solution,
The tableaux for the given set of formula is:
a
c
( a b )
( b c ) Since, all the branches on tableau
closes, therefore the given set of
a b formula is unsatisfiable.
b c
Method of Tableaux (Example_2)
Draw the tableau for the following set of formula
( P Q) S , Q R, P T
and check for satisfiability.
Solution,
The tableaux for the given set of formula is:
( P Q) S , Q R, P T
Method of Tableaux (Example_2)
( P Q ) S
Q R
P T
P Q S
P Q Q R
P P
Q R Q R
T T
P P P
T T T
Since, some branch on tableau are open,
the given set of formula is satisfiable.
Method of Tableaux
Checking for validity of an argument using
tableaux
Suppose we have argument of the form:
p1
p2
.
.
pn
q
To check the validity of the above argument we first
negate the conclusion and draw the tableau including it
in the set of the formula. i.e. we draw a tableau for
Method of Tableaux
We then draw a tableau for
p1
p2
.
.
pn
q
If a single or some branch of the tableau is closed, the
argument is valid.
If all the branches of the tableau is open, the argument
is invalid.
Method of Tableaux (Question_1)
Check the validity of the following argument using
the method of tableaux:
pq
pr
qr
r
Solution:
First we negate the conclusion and prepare a tableaux
including it into the list of other formulas.
Method of Tableaux (Question_1)
pq
pr
qr
r
Method of Tableaux (Question_1)
pq
pr
qr
r
p q
r p r p
r s r s
r r s
s t
t
s t
Method of Tableaux (Question_3)
Check the validity of the following argument using
the method of tableaux:
P (Q R)
QS
RA
P B
(S A) B
Solution:
First we negate the conclusion and prepare a tableaux
including it into the list of other formulas.
Method of Tableaux (Question_3)
P (Q R)
QS
RA
P B
( S A) B
P (Q R)
QS
RA
P B
( S A) B
B
( S A)
P B
P Since, all the branches on tableau
P QR closes, therefore the given
argument is valid.
Q
R
Q S
R A
S A
End of Chapter 1
Thank You !!!!