An Introduction To Mathematical Logic, Notation, Quantifiers, Definitions, and Proofs
An Introduction To Mathematical Logic, Notation, Quantifiers, Definitions, and Proofs
An Introduction To Mathematical Logic, Notation, Quantifiers, Definitions, and Proofs
MAT 137
In MAT137 we emphasize rigorous definitions and proofs. Many students struggle with this
because they have never seen anything similar in high-school math. These notes and the references
in it are designed to smooth that transition.
The basics
First we need some basic objects to work with. Read sections 1.2 and 1.3 from the textbook.
Mathematical logic
As a first introduction to mathematical logic, read section 3 (titled Mathematical logic) on http://uoft.me/precalc
Note: Actually, all nine sections of this website are an excellent preparation for MAT137. The other eight
sections are review, things we count on you having already learned in high school. (If this is not the case,
you should study then as well!) The section on Logic is probably new, which is why we emphasize it here.
On page 4 of the textbook you were introduced to the standard sets of numbers. Here are their usual
mathematical names:
the naturals or positive integers: N
the integers: Z
the rationals: Q
the reals: R
On section 3 of http://uoft.me/precalc you were introduced to mathematical logic. Here are some
standard pieces of notation:
A = B means If A, then B.
A B or A iff B (read A if and only if B) means The statements A and B are equivalent.
= for all or for every.
= there exists (at least one)
We will try not to go too heavy with notation, but you will encounter these symbols either in this course or
in future courses, so it is good to be aware of them.
1
MAT 137
Definitions
We are going to show with an example how to write down a mathematically rigorous definition. We want
to define what it means for a function to be increasing. We assume that you already understand intuitively
what this means something like the larger x is, the larger f (x) is but that is not yet a rigorous definition.
We need to compare the values of f at two points. An idea is to say that increasing means the following:
x1 < x2 ,
This is not a correct or incorrect definition. It simply is not a definition. What is f ? What are x1 and x2 ?
In a definition, you have to introduce all the symbols you will use. Otherwise, it is nonsensical.
Okay then: f is a function and x1 and x2 are arbitrary real numbers. How arbitrary? When we talk about
a function being increasing, we have to refer to a specific interval. A function could be increasing in some
intervals and decreasing in some other intervals. For example, the function whose graph is below is increasing
on [1, 1] but decreasing on [1,3]:
3
2
1
-4
-3
-2
-1
-1
-2
-3
This suggests that our definition might look like this:
Definition 4.1 (Bad). Let f be a function defined on the real interval I. We say that f is increasing on I
when
x1 , x2 I,
x1 < x2 ,
f (x1 ) < f (x2 )
Unfortunately, this definition is still ambiguous (or wrong, or nonsensical, but either way not good enough).
Do we mean that the last two equations are true for every pair of numbers x1 , x2 I? Do we mean that
there is a pair of numbers x1 , x2 I for which the two equations are true? Do we mean something else?
So, lets try to make it precise. Consider the following attempts to a more precise definition:
MAT 137
Definition 4.2. (Bad) Let f be a function defined on the real interval I. We say that f is increasing on I
when
For every x1 , x2 I,
x1 < x2 ,
f (x1 ) < f (x2 )
Definition 4.3. (Bad) Let f be a function defined on the real interval I. We say that f is increasing on I
when
There exist x1 , x2 I such that x1 < x2 and f (x1 ) < f (x2 )
Definition 4.4. Let f be a function defined on the real interval I. We say that f is increasing on I when
For every x1 , x2 I,
if
x1 < x2
then
Definition 4.2 is wrong. It is unambiguous and it means something precise, but it does not mean f is
increasing. Specifically, there isnt any function f that satisfies the property in Definition 4.2. Do you see
why?
Definition 4.3 is wrong. It is unambiguous and it means something precise, but it does not mean f is
increasing. Specifically, there are lots of functions f that satisfy the property in Definition 4.3, but which
are not increasing. Do you see why?
Definition 4.4 is the correct one! It is unambiguous, it means something precise, and it means exactly what
we want f is increasing to mean. Do you see why?
Incidentally, there is often more than one equivalent, correct way to define a concept. In Definition 4.4,
instead of
For every x1 , x2 I,
if x1 < x2 then f (x1 ) < f (x2 )
we could have written:
If x1 , x2 I and x1 < x2 ,
then
or
For every x1 , x2 I,
or
x1 , x2 I,
or
x1 < x2 = f (x1 ) < f (x2 )
for every
x1 , x2 I
5
5.1
Proofs
Basic structure of a proof
Most mathematical theorems are of the form If P, then Q. For example, a theorem we will learn early in
MAT 137 is:
Theorem. Every differentiable function is continuous.
MAT 137
(It is okay if you do not know yet what these concepts mean.) It may not look of the form If P, then Q,
but what this theorem is actually saying is
Theorem. If a function is differentiable, then it is continuous.
Or, to be even more precise, the theorem should probably be restated as
Theorem. Let f b a function and let a be a real number inside the domain of f . If f is differentiable at a,
then f is continuous at a.
The basic structure of a proof for an If P, then Q will always be the same. We start by assuming P to
be true, then we use other results, properties, and algebraic manipulations that we know are true, and we
conclude that Q has to be true. Schematically:
Theorem 5.1. If P, then Q.
Proof. Assume P is true.
STUFF
Then we have concluded that Q is true.
The little symbol at the end of the proof means end of the proof. Let us illustrate this with an example.
First, here is an example of a bad theorem and a bad proof:
Theorem 5.2 (Bad).
xy
x+y
2
Bad proof.
xy
xy
4xy
x+y
2
(x + y)2
4
(x + y)2
4xy
x2 + 2xy + y 2
(x y)2
Theorem 5.2 is bad because it is not a theorem; instead, it is an equation. An equation by itself can never
be a theorem. A theorem could be an equation together with an explanation of what the variables mean
and when the equation is true. So, when is this equation true? For all real numbers x and y? No, it isnt!
For example, when x = 1 and y = 4 the equation is false. This needs to be fixed.
As for the proof of Theorem 5.2, it is terrible! It has no words; it is just a bunch of disconnected statements;
there is no explanation of what we are assuming and what we are trying to proof; we do not know what
follows from what. Worse, it appears to start with the conclusion of the theorem, instead of ending with it.
If you ever write a proof like this in a test, it will get you 0 points, and it will make your grader cry.
Here is how to rewrite this into a correct theorem and a correct proof.
4
MAT 137
xy
x+y
2
Proof. Let x, y be any two positive real numbers. We know that c2 0 for every real number c. In particular,
we know that (x y)2 0. Lets expand this:
0
(x y)2
= 0
x2 2xy + y 2
= 4xy
x2 + 2xy + y 2
= 4xy
(x + y)2
= xy
(x + y)2
=
4
x+y
2
2
Since both x and y are positive, so is xy, and hence we can take the square root of both sides.
s
2
x + y
x+y
xy
=
2
2
where we have used that for any real number c, c2 = |c|. However, in this case, x and y are positive, and
so is x + y, so |x + y| = x + y and we conclude that
xy
x+y
2
xy
x+y
2
x+y
2
(1)
(x + y)2
4
(2)
xy
MAT 137
Notice that (2) is obtained from (1) by squaring both sides. Since both sides are positive in Equation (1),
the two equations are equivalent. Hence, we only need to prove that (2) is true, and it will follow that (1)
is true.
Now we do some algebraic manipulations, all of which are equivalences:
xy
4xy
4xy
0
(x + y)2
4
(x + y)2
x + 2xy + y
(3)
(4)
2
(x y)2
(5)
(6)
All of Equations (3) through (6) are equivalent. Equation (3) is the one we want to prove. Equation (6) is
clearly true. Hence, we have completed the proof.
5.2
Imagine that you need to prove a theorem of the form P = Q. There are three standard ways to do it.
Direct proof. This is what we discussed in section 5.1
1. Assume P is true.
2. Do stuff.
3. Conclude that Q is true
Proof by contrapositive. Instead of proving that [P = Q] directly, we prove the equivalent statement
[(not Q) = (not P)].
1. Assume Q is false.
2. Do stuff.
3. Conclude that P is false
If you are wondering why this works, review the section on Mathematical Logic at http://uoft.me/precalc
once more.
Proof by contradiction.
1. Assume P is true and Q is false.
2. Do stuff.
3. Obtain a contradiction.
The contradiction means that it is impossible for P to be true and for Q to be false at the same time.
Hence if P is true, Q must be true as well, and this is a proof of [P = Q].
Proofs by contrapositive and by contradiction will rarely appear in MAT137. However, they are very common
in mathematics, and you will encounter them often in MAT237 and in Linear Algebra.
5.3
Proof by induction