As we move into proving mathematical statements, we need to better understand the statements themselves. In your previous work with mathematics, youβve often used variables and equations.
Think of some examples of variables youβve seen in other math classes. What do these variables represent? Probably, the first example that comes to mind is something from an algebra class, such as \(y=x^2\) or \(f(x)=3x+2\text{.}\) You may also think of examples such as solving for \(x\) in an expression such as \(x^2=4\text{.}\) We use variables in mathematical statements to represent quantities that can vary. But we also use them to be more precise. Think of how much more confusing the first statement would be if we had to write it as "a number squared equals another number;" or the second statement, "the function which multiplies a number by 3 and adds 2." When you first learned varables, you probably WERE introduced to them in terms of sentences, but eventually, you got used to what the symbols mean.
In this class, we will rarely be interested in mathematical equations. We want to move to the common format of mathematical statements often found when describing mathematical defintions or theorems.
Many mathematical statements use variables. For example, \(x+3/2=4\) is not a statement since it has no truth value. However, if we say, βthere exists \(x\) such that \(x+3/2=4\)β or βfor all \(x\text{,}\)\(x+3/2=4\)β, and we know what set \(x\) belongs to, then we can decide whether these statements are true or false.
We want to be able to work with generic statements, like \(p\) and \(q\text{,}\) but also with variables. So instead of \(p\text{,}\) we will use \(P(x)\text{.}\) An expression, \(P(x)\text{,}\) represents some property or expression about \(x\text{.}\) We call \(P(x)\) the predicate.
Our properties donβt just need to be mathematical, though. For example we could have a predicate such as \(P(x)\) is β\(x\) is a math major.β In this case the set of possible values for \(x\) could be the set of students at Linfield, or the set of students in Introduction to Proofs. As we saw in ExampleΒ 2.2.1, it is important that we define our set of interest, as changing the set can change whether the statement is true or false.
To show a universal statement is true, you need to show all values in \(D\) make \(P(x)\) true. If your set is small, you can do this just by showing \(P(x)\) is true for each \(x\text{.}\) If \(D\) is infinite, however, we will need more general techniques.
To show an existential statement is false, you need to show all values in \(D\) make \(P(x)\) false, or no values make it true. If your set is small, you can do this just by showing \(P(x)\) is false for each \(x\text{.}\) If \(D\) is infinite, however, we will need more general techniques.
We can think of negation as switching the quantifier and negating \(P(x)\text{,}\) but it will be really helpful if we can understand WHY this is the negation. Thinking about negating a βfor allβ statement, we need the statement to not be true for all things, which means it must be false for something, Thus, there exists something making \(\neg P(x)\) true. Thinking about negating a βthere existsβ statement, we need there not to exist anything making \(P(x)\) true, which means \(P(x)\) must be false for everything. Thus, everything makes \(\neg P(x)\) true.
Many of our quantified statements may have predicates involving other logical connectives. So it is going to be important to remember how to negate "and"s, "or"s, and "if...then"s.
The relationship between βfor allβ and βthere existsβ can be used to show some surprising things. In particular, what happens if our domain, \(D\text{,}\) has nothing in it? Let \(D=\emptyset\text{,}\) the empty set. Is \(\forall x\in D, P(x)\) true or false? Well, letβs look at the negation: \(\exists x\in D, \neg P(x)\text{.}\) Now the negation MUST be false since \(D\) has nothing in it, so there canβt exist something in \(D\) making \(\neg P(x)\) true. Since the negation is false, the original statement is true! We say \(\forall x\in D, P(x)\) is vacuously true.
Consider the statement βFor all turtles, \(T\text{,}\) in Intro to Proofs, \(T\) is getting an A.β The negation is βThere exists a turtle, \(T\text{,}\) in Intro to Proofs, such that \(T\) is not getting an A.β Since no such turtle exists, the negation is false. Making the original true. So every turtle in Intro to Proofs is getting an A.
As one additional note, it can be helpful in deciding if your negation is correct to find the truth value of both the original and the negation. They should have opposite values. Similarly, if you need to determine the truth value of a complex statement, it might be easier to find the truth value of the negation.
Many statements in mathematics involve multple quantifiers. For example, βFor all real numbers \(x\text{,}\) there exists a real number \(y\) with \(x+y=0\text{.}\)β These statements, though frequent in math courses, represent some of the most complicated types of statements to understand. In this section we will try to understand the general structure of such statements.
Letβs look at the various ways we could have statements with two quantifiers. Since we have two quantifiers, we will have two variables. Thus, our predicate will now involve both variables. We can use the notation \(P(x, y)\) to mean a statement about \(x\) and \(y\text{.}\)
This means we can find a \(y\) in \(E\) that makes \(P(x, y)\) true for each \(x\) in \(D\text{.}\) In such statements \(y\) will often depend on \(x\text{.}\)
The order of the quantifiers matters. This means we can find a SINGLE \(x\) in \(D\) that makes \(P(x, y)\) true for every \(y\) in \(E\text{.}\) In such statements \(x\) does not depend on \(y\text{.}\) The same \(x\) must work for all the \(y\)βs.
Although we wonβt study the mathematical context of the following example in this class, it is a classic example in mathematics of a statement with multiple quantifiers and a conditional. It is so important that it is the subject of one set of math bike racks in front of Taylor Hall!
The definition of the limit of a function: For every \(\epsilon>0\text{,}\) there exists a \(\delta>0\) such that if \(|x-a|<\delta\) then \(|f(x)-L|<\epsilon\text{.}\) Or in symbols,
Now, we want to be able to negate statements with multiple quantifiers. There is nothing really new here, we just negate our quantified statements as we did for single quantifiers.
If you take Elementary Analysis, you will need to be able to negate the definition of the limit from ExampleΒ 2.2.6. Negate βFor every \(\epsilon>0\text{,}\) there exists a \(\delta>0\) such that if \(|x-a|<\delta\) then \(|f(x)-L|<\epsilon\)β. Our statement has the form \(\forall \epsilon, \exists \delta, P(\delta)\rightarrow Q(\epsilon)\text{,}\) so the negation has the form \(\exists \epsilon, \forall \delta, P(\delta)\wedge \neg Q(\epsilon).\)
Rewrite the statement in English without using the symbols \(\forall\) or \(\exists\text{.}\) Express your answer as simply as possible. Then write a negation for the statement. Determine which statement is true, the original or the negation.
For each statement write a new statement by interchanging the symbols \(\forall\) and \(\exists\text{.}\) Which is trueβ the given statement, the interchanged version, neither, or both?
Consider the following sequence of digits: 2300204. A person claims that all of the 1βs in the sequence are to the left of all of the 0βs in the sequence. Is this true? Justify your answer.