1.

5b The Intermediate Value Theorem

In this lesson we introduce an important theorem in calculus, the Intermediate Value Theorem. This is
important in itself, but also because it is an example of what we call an existence theorem in
mathematics. These tell us that under certain circumstances, a particular problem MUST have a solution.
But let’s start with a story.

There once was a monk who wished to spend a week meditating

at a temple on top of a mountain. He left one Friday morning AT 9
am from his monastery and walked up the lone path to the
temple. He arrived at 6 pm that evening.

He spent a week in deep thought, and left the following Friday

again at 9 am, arriving back at the monastery at 6 pm. When the
monk entered, his abbot, blind but wise, pronounced, “I know
that you were in the exact same place on the temple trail, exactly
one week apart.” The young monk was quizzical: “But wise father,
how could you know this? You could not see how I ascended or
descended on the path.” “Because of the intermediate value theorem, I know it must be true,” came the
reply.

What did the wise abbot know? We’ll see exactly in a bit, but let’s first examine this important theorem.
As we mentioned, the IVT is an existence theorem. It says that if f is a continuous function on a closed
interval [a,b] and if N is a number between f(a) and f(b). Then, there must be a c between a and b such
that f(c) = N. Notice the theorem does NOT tell you what c is exactly or how to find it, but just says that
such a number must exist.

Let’s use the graph of a continuous function to see why this must be true.
Remember that the IVT has some conditions that must hold, namely that f be continuous on [a,b] (that
includes the endpoints!). And N is a value between f(a) and f(b). With this set up, we draw the graph of a
continuous function from (a,f(a)) to (b,f(b)). Remember that for continuous graphs, we cannot pick up
our pencil when drawing them. The result is that no matter what f does, It cannot get to f(b) without
crossing N. The point where this happens is the c from the theorem.

Notice the theorem does not say how many times the function achieves the value N, just that it must AT
LEAST ONCE. See here, there are several such c’s.

It can’t be said enough that this theorem is about “continuous” functions. Suppose we didn’t have that?
Then, we might have a situation like this, where the function “jumps over” the value N.

As important as the statement of the Intermediate Value Theorem is how we use it. Let’s see two
examples. First we want to find consecutive integers (hence n and n+1) that bound a zero of a quintic
polynomial.
Well, we know that at 0, this function has the value -1, but that as x gets large, this polynomial yields
very large values, so it must have a zero, that is, an x-intercept, for some positive x values. We examine
its values at integer values. The polynomial’s output value change sign between 3 and 4.

Let’s check the requirements of IVT:

1) p is a polynomial, so it is continuous everywhere, in particular in the closed interval [3,4], where 3 and
4 here play the role of a and b in IVT

2) 0 is between p(3) and p(4) (0 is playing the role of N).

Thus, by IVT, we can conclude that there must be a number c between 3 and 4 where p(c) = 0. That is, c is
a zero of the polynomial.

Here’s another example. We want to show that this equation has a root (that is, a solution).

At first glance this problem looks like it is about 2 different functions, x and cos x. But, mathematicians
almost never just show two things are equal. Instead, they show their difference is 0.
Now we can apply the IVT to the function, cos x – x, which is continuous everywhere. At x=0, the output
it cos 0 = 1.

And let’s try the value of cos x - x at π, it is -1 – π which is –(1+ π) hence a negative number.

You might ask how one knows to choose π? We just picked a number at which we know how to evaluate
the function. So, like the polynomial example, the function changes sign. So it must be zero somewhere
in the interval (0, π).

Let’s look at it geometrically. You see the graphs of y=cos x and y=x. We just claimed that there is number
in the interval (0, π) such that these functions are equal. Geometrical that means that the two graphs
must intersect. So there must be a c in (0, π) where they intersect as we see from the graph.

All of which brings us to the wise abbot. How did he use the IVT? Well, the time interval, being the same,
is key. Let’s model the situation by letting f(t) = the position of the monk on the first Friday at hour t. g(t)
= the position of the monk on the second Friday at hour t. Then f and g are continuous because even the
most enlightened monk cannot teleport himself.
Assume the monastery is at position 0 and the temple is at position P> 0. Then the function f(t) – g(t) is
continuous on the interval [9,18] (that is, from 9 am to 6 pm). Furthermore, f(9) – g(9) = 0 – P<0
(because on the first Friday he started at the monastery and the second at the temple).

And f(18)-g(18) = P- 0 >0. So as before, there must have been a time c, between 9 am and 6pm where
f(c) = g(c) or we say there must be a c that is a root of f(c)-g(c)=0.
Graphically, we could think of the two function f and g as switching positions from 9 am to 6 pm. No
matter how they proceed, if they are continuous they must cross at least once!

Lastly, if you don’t like that, picture this:

On the way down the second Friday, the monk sees the ghost of his trip up the previous Friday. No
matter how he descends (fast, slow, two-steps-forward-one-step-back), he must pass his ghost self at
least once. This marks the point at which the time and position were identical exactly a week apart.
 Self-assessment 1.

Congrats on finishing this module. We have been introduced to a powerful theorem about continuous
functions, the Intermediate Value Theorem. This mathematically captures the intuitive idea that you
can’t get continuously from here to there without hitting each point in between. It has many uses,
including but not limited to proving the existence of solutions to equations and locating zeros of
polynomials.

Self-assessment answers.

1. A, D