You are on page 1of 3

Discrete Mathematics Rules of Inference

5-2

Previous Lecture
Logically equivalent statements Main logic equivalences

Rules of Inference Introduction

double negation DeMorgans laws commutative, associative, and distributive laws idempotent, identity, and domination laws the law of contradiction and the law of excluded middle absorption laws

Discrete Mathematics Andrei Bulatov

Discrete Mathematics Rules of Inference

5-3

Discrete Mathematics Rules of Inference

5-4

Logic Inference
ne of the main goals of logic is to distinguish valid and invalid arguments !hat can we say about the following arguments" ##$f you have a current password, then you can log onto the networ%& 'ou have a current password& (herefore, you can log onto the networ%& and ##$f you have a current password, then you can log onto the networ%& 'ou can log onto the networ%& (herefore, you have a current password&

Logic Inference (cntd)


!rite these arguments in symbolic form p ) you have a current password q ) you can log onto the networ% pq p q
#(herefore

pq q p

Discrete Mathematics Rules of Inference

5-5

Discrete Mathematics Rules of Inference

5-6

Inference and Tautologies


*hec% that + ,,p q- p- q is a tautology

Inference and Tautologies


Let us try + ,,p q- q- p p 0 0 1 1 q 0 1 0 1 pq 1 1 0 1
,p q- q

$f ,p q- p is false, that is if one of p q and p is false, then is true $f ,p q- p is true, then both p q and p are true& /ince the implication p q is true and p is true, q must also be true& (herefore is true& (herefore whatever values of p and q are, if ,p q- p is true, then q is also true& (he first example is a valid argument.

0 1 0 1

1 0 1 1

$n the case p + 0, q + 1 both conditions ,p q and q- are true, but p is false& (his is not a valid argument.

Discrete Mathematics Rules of Inference

5-7

Discrete Mathematics Rules of Inference

5-8

General Definition of Inference


(he general form of an argument in symbolic form is

Rules of Inference
2erifying if a complicated statement is a tautology is nearly impossible, even for computer& 3ortunately, general arguments can be replaced with small collection of simple ones, rules of inference&

(p1 p 2 p3 K p n ) q
premises conclusion

(he argument is valid if whenever each of the premises is true the conclusion is also true (he argument is valid if and only if the following compound statement is a tautology
(p1 p 2 p 3 K p n ) q

Modus ponens pq p q ##$f you have a current password, then you can log onto the networ%& 'ou have a current password& (herefore, you can log onto the networ%&

Discrete Mathematics Rules of Inference

5-9

Discrete Mathematics Rules of Inference

5-10

Rule of Syllogism
pq qr pr (he corresponding tautology ,,p q- ,q r-- ,p r##$f you send me an e)mail, then $ll finish writing the program& $f $ finish writing the program, then $ll go to sleep early& p ) #you will send me an e)mail q ) #$ will finish writing the program r ) #$ will go to sleep early ##(herefore, if you send me an e)mail, then $ll go to sleep early

Modus Tollens
pq q p (he corresponding tautology ,,p q- q- p ##$f today is 3riday, then tomorrow $ll go s%iing& ##$ will not go s%iing tomorrow& p ) #today is 3riday q ) #$ will go s%iing tomorrow ##(herefore, today is not 3riday

Discrete Mathematics Rules of Inference

5-11

Discrete Mathematics Rules of Inference

5-12

Rule of Disjunctive Syllogism


pq p q (he corresponding tautology ,,p q- p- q ##$ll go s%iing this wee%end& $ will not go s%iing on /aturday& p ) #$ will go s%iing on /aturday q ) #$ will go s%iing on /unday ##(herefore, $ will go s%iing on /unday

Rule for Proof by ases


pr qr ,p q- r (he corresponding tautology ,,p r- ,q r-- ,,p q- r##$f today is /aturday, then $ll go s%iing& $f today is /unday, then $ll go s%iing& p ) #today is /aturday q ) #today is /unday r ) #$ll go s%iing ##(herefore, if today is /aturday or /unday, then $ will go s%iing

Discrete Mathematics Rules of Inference

5-13

Discrete Mathematics Rules of Inference

5-14

Rules of ontradiction! Sim"lification! and #m"lification


4ule of *ontradiction p 3 p Reductio ad Absurdum

Logic Inference
(he goal of an argument is to infer the required conclusion from given premises

(he corresponding tautology ,p 3- p

4ule of /implification pq p p pq (he corresponding tautology ,p q- p

3ormally, an argument is a sequence of statements, each of which is either a premises, or obtained from preceding statements by means of a rule of inference

4ule of Amplification (he corresponding tautology p ,p q-

Discrete Mathematics Rules of Inference

5-15

Discrete Mathematics Rules of Inference

5-16

$%am"le
5remises" ##$t is not sunny this afternoon and it is colder than yesterday& !e will go swimming only if it is sunny& $f we do not go swimming, then we will ta%e a canoe trip& $f we ta%e a canoe trip, then we will be home by sunset& *onclusion" ##!e will be home by sunset& 6otation" p ) it is sunny this afternoon q ) it is colder than yesterday s ) we will ta%e a canoe trip r ) we will go swimming t ) we will be home by sunset 5remises" p q, r p, r s, and s t

$%am"le (cntd)
!e have p q, r p, r s, and s t /tep 1& p q 7& p 8& r p 9& r :& r s ;& s <& s t =& t 4eason premise simplification premise modus tollens premise modus ponens premise modus ponens

*onclusion" t

Discrete Mathematics Rules of Inference

5-17

&ome'or(
>xercises from the Boo%" 6o& 1a, 8c, 9a, :c, ?a ,page =9-