He came in the dry season

December 1879. Ferdinand de Lesseps steps ashore at Colón with his young wife and his children, throws a ball for the locals, and spends a few weeks touring the isthmus. It is the driest month of the year. The rivers are polite. He is seventy-four, the man who cut Suez, and he announces that Panama will be easier: a sea-level channel, no locks, done in eight years for 658 million francs.

Then the rains came, and the Chagres River rose forty feet in a night, and the mosquitoes did what mosquitoes do to twenty thousand men.

The engineers on the ground knew within two years that locks were the only design that worked. De Lesseps held the sea-level plan for most of a decade. Not because he was stupid. Because switching meant that every meter already excavated became a monument to a wrong decision, and he could not make himself pay that price. By the time he relented it was 1887, and the collapse two years later wiped out the savings of roughly 800,000 French households.

Behavioral researchers Clayton Critcher and Christine Cho have a name for it, which Katy Milkman unpacked on her decision-making show Choiceology: doubling back aversion. We refuse a faster route to a goal when taking it requires undoing progress we've already made. Rooted in sunk cost is the story of having been right, the money down the drain is just the price.

Ethan Hawke, writing a book of instructions from a knight to his descendants, put the antidote in one line in Rules for a Knight: never make a big decision without first walking a mile. De Lesseps walked. He just walked in January.

Your inner voice has a co-author

A father, tired, says something off the cuff to a nine-year-old. I don't know what's the matter with you, you're always screwing things up. He means it as motivation. He forgets it by dinner.

Jim Loehr, a performance psychologist who spent decades working with athletes on what happens between their ears, told Shane Parrish on The Knowledge Project that he stopped on a dime when he understood the impact. A parent, for a window of years, has direct access to what he calls command central. Whatever the public voice says, the private voice will later repeat, in the child's own accent, for forty years. Loehr's test after that: is this what I want him saying to himself when he's fifty?

Most of us are running rules we never wrote. Some of the harshest sentences in your head were installed by someone who was in a bad mood on a Tuesday in 1994 and has no memory of it.

Which is not an argument for demolition. I wrote about the opposite failure recently, Why That Odd Process Exists: On Fences, Fail‑safes, and Change, because the reflex to rip out the weird rule is usually as lazy as the reflex to obey it. The self-critical voice that keeps you from shipping garbage is a fail-safe. The one that keeps you from asking for the raise is rust. They sound identical from the inside. The only way to tell them apart is to find out who put the fence there, and what was in the field.

Four houses away

Binghamton, New York, 1956. Dick Stack is twenty-seven. The store is gone, the house is gone, the car is gone, his wife is pregnant with their second child, and the creditors are real people in a small town who extended him goods on his word.

Bankruptcy was available. It was legal, it was designed precisely for him, and nobody in that town would have thought less of him for using it. Shane Parrish told this story on Outliers and made the point without decorating it: Stack wasn't educated, wasn't sophisticated, wasn't patient. He kept his word. He sold every object he owned worth a dollar and paid each creditor in full. His wife took their son and moved in with her parents. He moved in with his mother, four houses down the street, and took a job at Montgomery Ward selling sporting goods for a salary, in the same trade he'd just failed at.

We can think of his stance as a moral one, but that's not the point. Stack was buying an asset that does not appear on any statement and cannot be financed: the willingness of other people to hand him inventory on nothing but a handshake, a second time. In the Gulf, the old gold and hawala networks priced this for centuries with no contracts and no courts, and the price of losing it was total. You do not get a partial downgrade. You get a phone that stops ringing.

The insolvency was a cash flow event. The alternative was a permanent impairment of the only capital that compounds without a market to mark it. He chose the loss that ends.

His son later ran the sporting goods chain that came out of the second store. Which is a nice ending, and also beside the point, because Stack made the decision in 1956 with no idea any of that was coming. He made it standing in an empty garage owing money to men he'd see at church.

Most people are meticulous about the balance sheet nobody checks and careless about the one everybody remembers.

My Ledger — notes on investing, books, and things I'm still figuring out. The Library is where I keep the books that shaped how I think. The Journal is where I work through ideas in public.

Mandelbrot — the private toolbox behind the thinking.

Hit reply if something here sparked a thought. I read every response and always write back.