On Almost Being Good at Math
A Lawyer Reflects on the Path not Taken
During the spring of my senior year in high school, I earnestly wanted to sharpen my mind. Perhaps because of this—or perhaps because of my pseudo-solipsism—I had never kissed a girl. I had already earned the grades colleges would see, and wanted to live a little. I found a vocation in tutoring my female classmates, and a jejune triumph in coming home late from doing so.
I decided not to take the AP exam in BC Calculus, to content myself with a B+ for my final grade, which meant I could get Bs on my last two quarter grades and rally for the final if my average dropped further than I planned. I had no chance of being top in my class, but I had a comfortable path to the average I needed to be one of a dozen “valedictorians” at my school in Fairfax County, Virginia.
I did not ditch AP Chemistry. It seemed useful in a way calculus just wasn’t. Once I found out my TI-85 could spit out definite integrals in seconds (and tables could handle indefinites if you knew differentiation, which was easy), the whole calculus thing seemed pretty stupid.
My newfound sociability had not rendered me oblivious to mathematical beauty. I could still appreciate the broadly applicable genius of making the partitions infinitely small. Limits struck me as officious, a way of formalizing an intuition. My thinking was more practical: just make the fucking partition smaller.
I knew I was slathered in computing power, and that Newton had spent years finding elegant workarounds for having less computing power than anyone with a calculator. Even with a calculator, you could just make the partitions really small.
In college, I saw that elegant models rested on cartoonish assumptions about human psychology and human-scale systems. The best models—the ones I could actually understand and improve—just iterated a set of interacting variables over time. Mathematical elegance began to look like a siren’s song in an age of promiscuous data and compute.
If applied math had been taught the way it is today—if there had been a department that showed me how to brute-force increasingly accurate models over time and use elegant math to benchmark and calibrate them—I would have given my mind to that pursuit. Instead, I became a lawyer.
As a 48-year-old lawyer, I regret that at my most intellectually vital and professionally flexible time, I strayed so far from math.
In the last twenty-seven years, I’ve learned spherical trigonometry to better understand flight paths, modeled presidential elections for fun, and solved Connect Four by coding in C++. If math were like pickleball, I’d win 4.0 competitions, have mixed results in 4.5, and get crushed in 5.0. But math isn’t like pickleball, and there’s no prestige in being the twelfth-best dude at math in Fayette County, Georgia. Anyway, to crack the top dozen, I would have had to work my butt off—and I’ve just been a lazy hobbyist. Math hobbyism would also be more fun, and probably more prestigious, if more women liked it—but the brute fact is most don’t, and it’s certainly not because I wanted to ban them from calculus class or from coding competitions.
Today, I was thinking about Italian high-speed trains. They make me happy because they are within reach but rare, so they have that aspirational glory that mass affluence has largely stolen from American consumption. Anyway, I wanted to know how many megawatts, constant load, it would take to get a train to 300 km/h over six minutes, because this is just a delicious acceleration path.
How long should I linger over the calculus? The fact that a v³ term comes up is interesting. Still, that v³ would be implicit in the output of a decent stepwise model. It’s an elegant piece of trivia that might come in handy if I ever need to engineer high-speed rail while camping without electronics—which is to say, basically never.
I had AI do it, and then checked its math anyway. If this is decadence, I want more.