This is an update with (unexpected) good news. My collaborator Miles
Wheeler and I were given the Paul R.
Halmos – Lester R. Ford Award
for our paper Perturbing the mean value theorem: implicit functions, the morse
lemma, and beyond^{1}
^{1}The
arxiv version of this paper is called *When
are there continuous choices for the mean value abscissa*, and is slightly
longer than the final published form.

This was an unexpected honor. Partly, this is due to the fact that we first
wrote this paper years ago and it was accepted in 2019. But the publication
backlog meant that it wasn't published until January of 2021, and thus eligible
for the 2022 award. But also it's always sort of a nice surprise to hear that
people *read* what I write.

I described this paper before, and wrote an additional note on how we chose our functions and made the figures.

When my wife learned that this paper won an award, she asked if this "was that paper you really liked that you wrote wtih Miles during grad school"? Yes! It is that paper. I really do like this paper.

## Origin story

The string of ideas leading to this paper began when I first began to TA
calculus at Brown. I was becoming aware of the fact that I would soon be
teaching calculus courses, and I began to really think about why we structured
the courses the way we do.^{2}
^{2}I don't have a completely satisfactory
explanation for everything, especially for the "integration bag of tricks" or
the "series convergence bag of tricks" portion. But upon reflection, I can
understand the purpose of *most* portions of the calculus sequence.

One of my least favorite questions was the *verify the mean value theorem for
the function $f(x)$ on the interval $[1, 4]$ by...* sort of question. The
problem is that this question is really just a way to check that one
understands the statement of the mean value theorem — and this statement
*feels* very unimportant.

But it turns out that the mean value theorem is *extremely* important. The mean
value theorem and intermediate value theorems are the two sneaky abstractions
that encapsulate underlying topological ideas that we typically brush aside in
introductory calculus courses.

### We don't do calculus on real valued functions over the rationals

We illustrate this with two examples in the analogous case of functions
\begin{equation*}
f: \mathbb{Q} \longrightarrow \mathbb{R}.
\end{equation*}
I would expect that many introductory students would think these functions
*feel intuitively about the same* as functions from $\mathbb{R}$ to
$\mathbb{R}$. But in fact both the intermediaet value theorem and mean value
theorem are false for real-valued functions defined on the rationals.

The intermediate value theorem is false here. Consider the function \begin{align*} f(x): \mathbb{Q} &\longrightarrow \mathbb{R}\\ x &\mapsto x^2 - 2. \end{align*} On the interval $[0, 2]$, for example, we see that $f(0) = -2$ and $f(2) = 2$. But there is no $x \in \mathbb{Q} \cap [0, 2]$ such that $f(x) = 0$.

The mean value theorem is false too. Consider the function
\begin{align*}
f(x): \mathbb{Q} &\longrightarrow \mathbb{R} \\
x &\mapsto \begin{cases} 0 & \text{if } x^2 > 2, \\
1 & \text{if } x^2 > 2.
\end{cases}
\end{align*}
For this function, $f'(x) = 0$ everywhere, even though the function isn't
constant. (But it is *locally constant*). We try to avoid pathological examples
initially.^{3}
^{3}It is a funny thing. Initially, we pretend that everything
is as nice as possible to build intuition. Then we see that there are
pathological counterexamples and use these to sharpen intuition. Seeing these
too early is potentially misleading. Implicit within the order is an idea of
what *usually* hapens and what behavior is *exceptional*.

It turns out that the topological space the functions are defined on *really
matter.*

### Mean value theorem as abstraction

I don't talk about this in a typical calculus class because topological concerns are almost entirely ignored. We work with the practical case of real-valued functions defined over the reals. And the key tools we use to hide the underlying topological details are the intermediate and mean value theorems.

I didn't fully realize this until I wondered whether we could teach calculus
*without* covering these two theorems.

It might be possible. But I decided that it was a bad idea.

Instead, I think it's a good idea to give students a better idea of how these two theorems are two of the most important ideas in calculus. I kept this idea in mind when I wrote An intuitive introduction to calculus for my students in 2013.

And more broadly, I began to investigate just how many things we could reduce, as directly as possible, to the mean value theorem. Miles realized we could investigate the continuity of the mean value abscissa with a low dimnsional implicit function theorem and baby version of Morse's lemma, and the rest is history.

I've collected so many random facts and questions tangentially related to the mean value theorem over the years. But I could always collect more!

### MathFest

I attended MathFest
for the first time to receive this award in person. MathFest was an
extraordinary and interesting experience. I was particularly impressed by how
many sessions and talks focused on actionable ways to improve math education
in the classroom *now*.

## Info on how to comment

To make a comment, please send an email using the button below. Your email address

won't be shared(unless you include it in the body of your comment). If you don't want your real name to be used next to your comment, please specify the name you would like to use. If you want your name to link to a particular url, include that as well.bold,italics, and plain text are allowed in comments. A reasonable subset of markdown is supported, including lists, links, and fenced code blocks. In addition, math can be formatted using`$(inline math)$`

or`$$(your display equation)$$`

.Please use plaintext emailwhen commenting. See Plaintext Email and Comments on this site for more. Note also thatcomments are expected to be open, considerate, and respectful.