CS 2742 (Logic in Computer Science) - Fall 2011: Antonina Kolokolova

CS 2742 (Logic in Computer Science) Fall 2011
Lecture 2
Antonina Kolokolova
September 9, 2011
1.1
Truth tables
Recall the puzzle about twin brothers from last class. It turns out that the question that
allows us to find out the name of the twin in front of us is Is John lying?. Let us consider
all possible situations:
1) Suppose that we met John and John tells the truth. Then he will truthfully reply
No.
2) Suppose that we met John and John lies. Then he will also say No: No, I am not
a liar, its the other guy!
3) Now suppose that we met Jim and Jim tells the truth. Then he will truthfully reply
Yes, since the other brother is the liar.
4) Finally, suppose that we met Jim, and Jim is the liar. Then he will also say Yes:
Yes, it is John who is the liar!
So as you see if the answer was No then it must have been John and if the answer was
Yes then it must have been Jim. Note that here we only learn what is the name of the
brother, not whether he lies or not. Also, you can check that Are you John? does not give
you an answer to this question.
We can write it as a table as follows. Let the propositional variable p denote this is John,
and the propositional variable q denote John tells the truth.
p: This is John q: John tells the truth Answer to Is John lying?

True
True
False
True
False
False
False
True
True
False
False
True
This kind of a table is called a truth table: it contains all possible truth values for the
pair of propositional variables p and q. This will be our tool for calculating truth values
of propositional formulas. For example, we can define the logical connectives from the
previous lecture using a truth table. We can also define formulas with values always true
(a tautology) and always false (a contradiction); for example
p
T
T
F
F
T
F
T
F
p q
(p implies q)
T
F
T
T
p q
(p and q)
T
F
F
F
p q
(p or q)
T
T
T
F
p
(not p)
F
F
T
T
p
(contradiction)
F
F
F
F
p p
(tautology)
T
T
T
T
So you can see that the answer to Is John lying? is the negation of This is John.
Definition 1 A truth assignment (to variables in a formulas) is an assignment of values
true/false to every variable.
When values of all variables are known, it is possible to calculate the truth value of the whole
formula, according to rules for logical connectives from the table above.
A truth assignment satisfies a formula if the formula evaluates to true under this assignment.
A formula is satisfiable if such a truth assignment exists. If a formula is not satisfiable by any
truth assignment, we call it a contradiction. If a formula is true under all truth assignments,
we call it a tautology.
To be more formal, we would define a propositional formula by structural induction, that is,
by saying that a proposition is a propositional formula, negation of a formula is a formula,
and two formulas connected by , , or other logical connective is also a formula. That
immediately gives us a way to evaluate propositional formulas starting from the propositions
and evaluating the truth values of the connectives according to the rules in the table above.
Example 1 (It is either not raining, or it is cloudy), and today is Monday.
Lets set p: it is raining, q: it is cloudy and r: today is Monday. Then (setting the
parentheses as above) the formula can be written as (p q) r. Today is Monday, it is
cloudy but not raining. So p is false, q is true and r is true. Then p is true, making the
true. And r is also true, making the true. The whole formula is, thus, true. Now suppose
it is raining, not cloudy and today is Monday (p = T, q = F, r = T ). In this case, both p
5
and q are false, so p q is false, and therefore the whole formula is false. Thus, this formula
is neither a tautology nor a contradiction.
How to evaluate logic formulas? Similar to arithmetic formulas, with being a as in
5, a and a +. Precedence rules: over , over , over . Otherwise, use
parentheses.
1.2
Showing Satisfiability vs. proving tautologies and contradictions.
Truth tables allow us to check if a given formula is a tautology, contradiction or is satisfiable,

by trying all possible truth assignments (how many truth assignments? 2number of variables ).
However, it is not a very efficient method, since checking a formula with even a thousand
variables (in software verification this is a reasonable number) would require a truth table
larger than the size of the universe!
Although there are some methods for checking if a formula is a tautology, none that we
know can give us an answer significantly faster then the truth tables. However, we dont
know if this is an inherent limitation or whether we havent come up with a smart enough
algorithm yet. This is a major open problem in mathematics, usually stated as P vs. NP
problem (one of 7 mathematical problems for the new millennium defined by Clay institute
which offers a million of dollars for solving each one). Checking if a given formula is not a
contradiction is a classical example of an NP complete problem.
Example 2 Lets look again at (p q) r and compute its truth table. We already
calculated two lines of the table: the p = F, q = T, r = T line and p = T, q = F, r = T line.
p
T
T
T
T
F
F
F
F
q
T
T
F
F
T
T
F
F
r p
T F
F F
T F
F F
T T
F T
T T
F T
(p q) (p q) r
T
T
T
F
F
F
F
F
T
T
T
F
T
T
T
F
Puzzle 2 (Knights and knaves 1) Some remote island is populated by two kinds of people: knights, who always tell the truth, and knaves, who always lie. Suppose you met two
islanders, call them A and B, and you hear A saying at least one of us is a knave. Can
you tell which of A and B is a knight and which is a knave?

CS 2742 (Logic in Computer Science) - Fall 2011: Antonina Kolokolova

Uploaded by

Copyright:

Available Formats

CS 2742 (Logic in Computer Science) - Fall 2011: Antonina Kolokolova

Uploaded by

Document Information

Original Description:

Original Title

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

CS 2742 (Logic in Computer Science) - Fall 2011: Antonina Kolokolova

Uploaded by

Copyright:

Available Formats

CS 2742 (Logic in Computer Science) Fall 2011

p: This is John q: John tells the truth Answer to Is John lying?

Showing Satisfiability vs. proving tautologies and contradictions.

Truth tables allow us to check if a given formula is a tautology, contradiction or is satisfiable,

You might also like