March 16th, 2018

There are two types of “Why”

The question “why?” can have two types of answers

  • What is the purpose of something
    • This assumes that things have a purpose
    • This is called “Teleology”
  • What is the mechanisms that causes something
    • This is the scientific approach
    • The mechanism has to follow some logic

Logic Arguments

Speaking the Truth

A brief introduction to Logic

  • The Universe contains Things

    • For example, “I”, “Istanbul”, “roses”, ”Turkish books”, “a letter”
  • Things have Attributes

    • For example, “large”, “red”, “old”, “which I received yesterday”
    • It is the same as to say that Things have Properties

Based on the book “Symbolic Logic” by Lewis Carroll

“Things” have “Attributes”

One Thing may have many Attributes, and one Attribute may belong to many Things

  • The Thing “a rose” may have the Attributes “red,” “scented,” “full-blown,”
  • The Attribute “red” may belong to the Things “a rose,” “a brick,” “a ribbon,”

Simple Predicates: one thing, one attribute

A phrase stating if a Thing has (or hasn’t) an Attribute

  • “The rose is red”
  • “Istanbul is a small city”

Predicates are either TRUE or FALSE

That is called the truth value of the predicate

We use symbols for “Things” and “Attributes”

Abstracting a little (to make it general), we can say that

  • If \(x\) is a Thing and \(A\) is an Attribute
  • then \(A(x)\) is a Predicate

When we see \(A(x)\) we say “\(x\) is \(A\)”

Example: \(x\): snow, \(A\) hot, \(A(x)\) is FALSE

Complex Predicates: several things, several attributes

We can combine simple predicates to make complex phrases that can be either TRUE or FALSE

To combine predicates, we can use only a few words

  • NOT: “The rose is NOT red”
  • AND: “The rose is red AND Istanbul is a small city”
  • OR: “The rose is red OR Istanbul is a small city”
  • IF…THEN: “IF the rose is red THEN Istanbul is a small city”

The meaning of “NOT”

This rule applies to a single predicate \(P(x)\)

The truth value of the complex predicate “NOT \(P(x)\)” depends on the truth value of \(P(x)\). We can see it in the truth table

\(P(x)\) NOT \(P(x)\)
TRUE FALSE
FALSE TRUE

Meaning of “AND”

The complex predicate \(P(x)\text{ AND }Q(y)\) is TRUE when all parts are true.

We can see it in a truth table

\(P(x)\) \(Q(y)\) \(P(x)\) AND \(Q(y)\)
TRUE TRUE TRUE
TRUE FALSE FALSE
FALSE TRUE FALSE
FALSE FALSE FALSE

Meaning of “OR”

The predicate \(P(x)\text{ OR }Q(y)\) is TRUE if any of the parts is TRUE

\(P(x)\) \(Q(y)\) \(P(x)\) OR \(Q(y)\)
TRUE TRUE TRUE
TRUE FALSE TRUE
FALSE TRUE TRUE
FALSE FALSE FALSE

Notice that this is an Inclusive OR

Exclusive OR rule

“XOR” is TRUE if any of \(P(x)\) and \(Q(y)\) is TRUE, but not both

\(P(x)\) \(Q(y)\) \(P(x)\) XOR \(Q(y)\)
TRUE TRUE FALSE
TRUE FALSE TRUE
FALSE TRUE TRUE
FALSE FALSE FALSE

Now the result if FALSE if \(P(x)\) and \(Q(y)\) are TRUE at the same time

Equivalence is a logic connector

Equivalence means “equal value”. \(P(x)\) is equivalent to \(Q(y)\) when \[P(x)\text{ is true IF AND ONLY IF }Q(y)\text{ is true}\]

\(P(x)\) \(Q(y)\) \(P(x)\) EQUIVALENT TO \(Q(y)\)
TRUE TRUE TRUE
TRUE FALSE FALSE
FALSE TRUE FALSE
FALSE FALSE TRUE

Many ways to say EQUIVALENT

\(P(x)\) is EQUIVALENT to \(Q(y)\)

\(P(x)\) is true IF AND ONLY IF \(Q(y)\) is true

(\(P(x)\) is true IF \(Q(y)\) is true) AND (\(P(x)\) is true ONLY IF \(Q(y)\) is true)

(IF \(Q(y)\) is true THEN \(P(x)\) is true) AND (IF \(P(x)\) is true THEN \(Q(y)\) is true)

(\(Q(y)\) IMPLIES \(P(x)\)) AND (\(P(x)\) IMPLIES \(Q(y)\))

\(P(x)\) is NECESSARY AND SUFFICIENT for \(Q(y)\)

Implication

The most important one

When it is true that “IF \(P(x)\) THEN \(Q(y)\)”?

\(P(x)\) \(Q(y)\) IF \(P(x)\) THEN \(Q(y)\)
TRUE TRUE TRUE
TRUE FALSE FALSE
FALSE TRUE TRUE
FALSE FALSE TRUE

All these are equivalent

\(P(x)\) IMPLIES \(Q(y)\)

IF \(P(x)\) is true THEN \(Q(y)\) is true

\(Q(y)\) is true IF \(P(x)\) is true

\(Q(y)\) is NECESSARY for \(P(x)\)

\(P(x)\) is SUFFICIENT for \(Q(y)\)

\(P(x)\) is true ONLY IF \(Q(y)\) is true

Bigger predicates

We can easily combine all the previous operations

  • \(P(x)\) AND \(Q(y)\) AND \(R(z)\)
  • \(P(x)\) OR NOT \(Q(x)\)

We use parenthesis to avoid ambiguity. For example

  • NOT \(P(x)\) AND \(Q(y)\) can be
    • NOT (\(P(x)\) AND \(Q(y)\))
    • (NOT \(P(x)\)) AND \(Q(y)\)

Avoid ambiguity

What do we mean when we say “NOT \(P(x)\) AND \(Q(y)\)”?

Let’s compare the two interpretations

\(P(x)\) \(Q(y)\) NOT (\(P(x)\) AND \(Q(y)\)) (NOT \(P(x)\)) AND \(Q(y)\)
TRUE TRUE FALSE FALSE
TRUE FALSE TRUE FALSE
FALSE TRUE TRUE TRUE
FALSE FALSE TRUE FALSE

They are different. Parenthesis are important

Are we saying “the same thing”?

Two predicates are equivalent when they have the same truth table

For example

\(P(x)\) \(Q(y)\) \(P(x)\) AND \(Q(y)\) \(Q(y)\) AND \(P(x)\)
TRUE TRUE TRUE TRUE
TRUE FALSE FALSE FALSE
FALSE TRUE FALSE FALSE
FALSE FALSE FALSE FALSE

For AND and OR, the order is not important

\(P(x)\) \(Q(y)\) \(P(x)\) OR \(Q(y)\) \(Q(y)\) OR \(P(x)\)
TRUE TRUE TRUE TRUE
TRUE FALSE TRUE TRUE
FALSE TRUE TRUE TRUE
FALSE FALSE FALSE FALSE

An important rule

De Morgan’s law

\(P(x)\) \(Q(y)\) NOT (\(P(x)\) AND \(Q(y)\)) (NOT \(P(x)\)) OR (NOT \(Q(y)\))
TRUE TRUE FALSE FALSE
TRUE FALSE TRUE TRUE
FALSE TRUE TRUE TRUE
FALSE FALSE TRUE TRUE

Negation of AND is the OR of negations

Another De Morgan’s law

\(P(x)\) \(Q(y)\) NOT (\(P(x)\) OR \(Q(y)\)) (NOT \(P(x)\)) AND (NOT \(Q(y)\))
TRUE TRUE FALSE FALSE
TRUE FALSE FALSE FALSE
FALSE TRUE FALSE FALSE
FALSE FALSE TRUE TRUE

Negation of OR is the AND of negations

Short notation of connectors

  • Good notation makes easy to write complex phrases in short space
  • Also helps to avoid ambiguity and errors

    Complex Phrase Notation
    \(P(x)\) AND \(Q(y)\) \(P(x) \wedge Q(y)\)
    \(P(x)\) OR \(Q(y)\) \(P(x) \vee Q(y)\)
    NOT \(P(x)\) \(\neg P(x)\)
    IF \(P(x)\) THEN \(Q(y)\) \(P(x) \Rightarrow Q(y)\)
    \(P(x)\) EQUIVALENT TO \(Q(y)\) \(P(x) \Leftrightarrow Q(y)\)

Predicates about attributes

To make phrases about Attributes we have to speak about the Things having these attributes

So far we saw how to evaluate the truth value of a logical phrase depending on the specific things (\(x\) and \(y\))

Now we care about the truth of the phrase in general

We have two key words, called quantifiers:

  • “All” (sometimes written as \(\forall\))
  • “Some”, or “Exists” (written as \(\exists\))

Examples of predicates about attributes

For example

  • “All things are natural”
  • “Some things are too expensive”
  • “All men are mortal”
  • “Nothing is impossible”

Writing the phrase in the correct form

The correct form of writing is using a Formal notation

Helps to be clear and precise

“All things are natural”

  • For all things \(x\), \(Natural(x)\)
  • \(\forall x, Natural(x)\)

“Some things are too expensive”

  • There exists a thing \(x\) such that \(TooExpensive(x)\)
  • \(\exists x,\, TooExpensive(x)\)

Negation of quantifiers

“Not all things are natural” means “There are some things that are not natural”

  • There exists a thing \(x\) such that NOT \(Natural(x)\)
  • \(\exists x,\, \neg Natural(x)\)

“Nothing is too expensive” means “All things are not too expensive”

  • For all things \(x\), NOT \(TooExpensive(x)\)
  • \(\forall x, \neg TooExpensive(x)\)

Rules for negations

These are the same

  • NOT (FOR ALL \(x\), \(P(x)\))
  • THERE EXISTS \(x\) SUCH THAT NOT \(P(x)\) \[\neg (\forall x, P(x)) \Leftrightarrow \exists x,\, \neg P(x)\]

These are the same

  • NOT (THERE EXISTS \(x\) SUCH THAT \(P(x)\)) is equivalent to
  • FOR ALL \(x\) NOT \(P(x)\) \[\neg (\exists x, P(x)) \Leftrightarrow \forall x,\, \neg P(x)\]