This summer I spent an unreasonable amount of time scratching an itch that probably affects eleven other people on Earth. I wanted email without Gmail, by which I do not mean that I wanted to abandon Google Workspace, operate my own mail server, maintain sender reputation, fight spam, or become the unpaid postmaster of a tiny and unreliable principality. Google can continue carrying the mail and accepting responsibility for the plumbing. What I wanted to escape was the browser. I wanted to sit down at any one of my old ThinkPads, open a terminal, type a short command, and return to an email environment that was already running in exactly the state in which I had left it. Same inbox, same folders, same open message, same cursor. I wanted email to feel less like visiting a commercial property and more like entering a room.

I bought the smallest useful DigitalOcean droplet I could find: Debian 13, one shared virtual CPU, 512 megabytes of RAM, ten gigabytes of storage, a New York IP address, and a monthly cost of four dollars. I added a one-gigabyte swap file because half a gigabyte of memory can occasionally find itself contemplating mortality. I protected it with SSH keys, disabled password and root login, configured the firewall to permit SSH and nothing else, and enabled automatic security updates. This is not hyperscale infrastructure. It is a digital broom closet with good locks.

I named it Pinebox. Its Linux hostname is the properly lowercase pinebox, but in my head it is Pinebox because it is a box containing Pine, except that it technically contains Alpine, the maintained descendant of Pine. Alpine is a terminal email client descended from the software many people encountered through university Unix accounts in the 1980s and 1990s. It provides folders, messages, an address book, a composer, and keyboard commands without surrounding the mail with news, advertisements, chat panels, productivity dashboards, nudges, badges, or whatever Google has decided belongs beside a letter this week.

Alpine alone was not enough because merely running a terminal email client on a remote server would still mean starting it again every time I connected. The piece that made Pinebox into the room I wanted was tmux. A tmux session continues running on the server when the SSH connection disappears. I can open Alpine from one ThinkPad, read some mail, leave a message open, close the laptop, and walk away. Alpine remains alive on the droplet. Hours later I can use another computer, connect to Pinebox, reattach to the same tmux session, and return to the same place. The laptop is no longer where the application lives. It is a door into the machine where the application has been waiting.

On my local machines, the command pine establishes the SSH connection, asks tmux to attach to the existing mail session or create it if necessary, and places me directly inside Alpine. There is no separate ritual of logging in, finding the session, remembering what I called it, and launching the client. I type pine and arrive. The whole arrangement is made of ordinary Unix components that are individually old, legible, and boring. SSH provides the secure connection. Tmux provides persistence. Alpine handles the mail. Debian sits underneath them and tries not to become interesting. The result does something enormous modern applications never manage: it preserves continuity without making me think about continuity.

Once that worked, another itch presented itself. I have several old ThinkPads, partly from thrift, partly from affection, and partly because I believe the correct number of ThinkPads is determined by the same mathematics used to calculate the correct number of cast-iron pans. The nuisance is remembering which computer contains the thing I wrote at 11:40 on a Tuesday night. Synchronization services are supposed to solve this, but synchronization creates its own little bureaucracy of accounts, agents, local databases, conflict files, desktop clients, browser interfaces, and opportunities to discover that the empty copy of a document has somehow become authoritative.

I did not want my computers to synchronize copies of my writing. I wanted there to be only one copy.

Pinebox was already online, secured, backed by a real Linux filesystem, and accessible from every computer I use. It already contained the communications system through which I might eventually send anything worth sending. There was no reason to rent a second server merely for conceptual neatness, so Pinebox acquired a second identity: Picobox. Pico was the small text editor associated with the original Pine environment, while Nano is the free Pico-like editor installed on my Debian droplet. Pinebox is the mail personality and Picobox is the writing personality, but both inhabit the same four-dollar machine.

I wanted the writing interface reduced to three letters: txt. When I type txt at the LinBay command line, LinBay does not launch an editor or create a file. It establishes an SSH connection to Pinebox and tells Pinebox to run a small script named picobox. That script examines whether I supplied a filename. If I typed only txt, it uses the current Eastern date and opens a journal entry such as /home/chris/text/journal/2026-09-02.txt. If the file does not exist, Nano creates it when I save. If it already exists, Nano reopens it and returns me to where I was working. When the date changes, the same command opens the next day’s file.

The server kept UTC by default, which would have caused Picobox to decide that tomorrow had arrived at eight o’clock in the evening during daylight-saving time. I changed Debian’s configured timezone to America/New_York. The underlying clock remains properly synchronized to UTC, while the operating system interprets the date in Eastern time and moves automatically between EST and EDT. This is the sort of detail that seems pedantic until September 3 begins annexing the final four hours of September 2.

The same command can open named documents. Typing txt poem opens or creates /home/chris/text/poem.txt. Typing txt poem.txt opens the same file because the script removes an existing .txt suffix before adding the canonical one. There will be no poem.txt.txt on my watch. Quoted filenames with spaces work as well, and I created text as an alias for the rare occasion when I feel extravagant enough to type a fourth letter. I expect most permanent filenames will use simple hyphenated names because I am building a writing appliance, not testing the outer limits of Bash quoting.

The crucial feature of the local launcher is what it cannot do. It contains no command for opening a local copy of Nano and no local writing path. Its only possible action is to invoke SSH and ask Pinebox to run Picobox. If the droplet is unavailable, txt fails and returns me to the LinBay prompt. It cannot quietly fall back to a local file. A system that sometimes saves remotely and sometimes saves locally is not a system; it is the opening chapter of a detective novel about missing documents.

The illusion is convincing because a terminal is only a rectangular grid of characters. When remote Nano appears inside the LinBay terminal, it looks and responds like a local application. Both machines have a user named Chris, so both have paths beginning with /home/chris, and the terminal window itself naturally continues to identify LinBay as the local host. The first time I saw the editor, I became suspicious that I had somehow built a complicated path back to my own hard drive. Nano can execute a shell command and insert its output into the current document, so I told it to run hostname. It inserted pinebox. The window belonged to LinBay, but the Nano process and the file belonged to the droplet. I discarded the test file because a journal entry consisting only of the hostname of a Debian server seemed a little too modernist even for me.

I configured Nano for prose rather than programming. Soft wrapping allows paragraphs to flow according to the width of the terminal without inserting hard line breaks into the file, and wrapping at blanks keeps words intact on the screen. Position logging returns me to the previous cursor position when I reopen a document. Search history persists between sessions. Tabs become spaces, the status display remains visible, and Nano writes backup versions into a private hidden directory on Pinebox rather than scattering tilde-suffixed files throughout the writing directory. The documents themselves remain ordinary text files with no proprietary structure, document database, export procedure, synchronization metadata, or application-specific container.

Nano is often treated as the editor people use only because Vim has trapped them inside a file and they need somewhere less demanding to recover. That is unfair. For prose, Nano is almost ideal. It opens instantly, puts the essential commands on the screen, works entirely from the keyboard, and does not confuse difficulty with power. It has no themes marketplace, account system, subscription tier, writing streak, daily goal, inspirational prompt, ambient rainfall, artificial intelligence offering to make my sentences more professional, or animated fountain pen congratulating me for reaching five hundred words. I do not call this a “distraction-free writing environment” because that phrase has been used to sell too many applications containing distractions. It is an editor containing the text and the commands required to edit the text.

Pinebox and Picobox appeal to me because they are made from small parts with understandable responsibilities. There is no bespoke platform holding the whole thing together. If I stop liking Nano, I can install another editor. If DigitalOcean annoys me, I can copy the directory to another Linux server. If every shell script disappears, the writing remains readable by tools older than some sovereign nations. Pinebox consists of Debian, SSH, tmux, Alpine, and configuration files. Picobox consists of Nano, a directory of text files, and two short shell scripts. Nothing is trapped inside either system because neither system is substantial enough to become a trap.

Technology companies have spent decades periodically rediscovering the network computer, always as a way of replacing a capable personal machine with a dependent window into somebody else’s enormous advertising or subscription apparatus. This is my preferred network computer. The clients can be almost anything capable of running SSH. The server is tiny, cheap, boring, and controlled by me. My ThinkPads do not need to agree about the current version of a document because they do not possess competing versions. They only need to know how to reach Pinebox.

Pinebox began as an attempt to make email feel like a persistent place rather than a succession of browser sessions. Picobox applies the same idea to writing. From LinBay, and eventually from every other box I use, pine enters the mail room and txt opens the current page. The applications run remotely, but the distance disappears inside the terminal. Everything is stored in one place, written in formats that require no rescue plan, and accessible through commands short enough to become muscle memory.

If I eventually rent a separate droplet and install an AI on it, that machine will obviously be called Blackbox. For now it has not earned the four-dollar monthly expenditure. Pinebox and Picobox are enough: one microscopic patch of rented silicon in New York where my mail remains open, my writing remains where I put it, and today’s journal entry is always three keystrokes away.

Philan Le’s contribution to the street furniture of Washington was a ten-foot wooden frame with a metal blade in the bed of a Toyota Tacoma. He had driven it from Julian, California. The Tacoma is the respectable American cousin of the Hilux, the pickup that has spent forty years as the unofficial staff car of men who dislike the government they are currently driving toward. In the Horn of Africa and the Levant the same silhouette is called a technical once you bolt something uncivil onto the bed. Le bolted a guillotine. The brand is doing the work it always does. It looks like errands. It photographs like an insurgency.

He parked on East Capitol Street, illegally, beside the Supreme Court and a short walk from the Capitol. An officer tapped the blade, found it sharpened, and asked whether the thing would cut. Le said it was the real deal. Capitol Police booked him for attempted possession of a prohibited weapon. A judge released him the next day, observing that he had threatened no one and had not fled. The Washington Post, a week later, framed the problem as threat versus protected speech.

That is a fairly ordinary sequence if you leave the adjectives out. A man built an execution machine, hauled it across the country, and left it in a no-parking zone outside the two buildings most likely to make the photograph mean something. The charge is modest. The commentary is already doing more work than the statute.

The comments under the Post are the more revealing document. A good number of readers are sure January 6 featured a gallow’s upgrade that never existed: a guillotine. Others file Le under art, Halloween, show-and-tell, a coffee run, a woodchipper, a sugar-cane chopper, a crucifix worn as jewelry. One suggests he was simply late for the last riot. Another says that if members of Congress take the blade personally, that is their conscience talking. The jokes are not stupid. They are a sorting mechanism. They decide in advance whether the object is allowed to be what it looks like.

January 6 did not involve a guillotine. It involved a small wooden platform on the Union Square lawn, put up before dawn by five men who unloaded lumber from a van around 6:30 a.m., left the crossbeam off, came back near 1 p.m., and added a bright orange rope. They have never been publicly identified or charged for the construction. The structure was crude. It would not have hanged Mike Pence. It would not have hanged a sandbag. The chant directed at the Vice President was ugly and specific. The rope was visible in a thousand photographs. Neither fact supplies a drop, a trapdoor, or an engineering diagram. For several years the national story treated the platform as if those things were incidental.

Le’s machine is the opposite problem. It looks like the job it was named for. His explanations do not. His brother called it a symbol of disapproval aimed at both parties. Le has described it as conceptual art, as self-defense, as a comment on forever wars, and as a trip undertaken with no particular member of Congress in mind. Prosecutors will have to prove intent to use it unlawfully against a person. That is why lawyers are already sounding doubtful. Motive is where a column can hide. Steel is not.

The January 6 file keeps being thickened for the same reason this new file is being thinned. The first object was not dangerous enough on its own, so the surrounding day had to carry it. The second object is too plainly an execution device, so the surrounding language has to promote it into speech.

The surrounding day, stripped of the liturgy, is still a riot and still not a hostage-taking. No senator was seized. No House member was seized. No staffer was kidnapped, shot, or carried off. They sheltered. Police took the beating in the west terrace tunnel—spray, poles, crush injuries, on the order of 140 officers hurt. That is the violence that actually occurred. It does not require a working gallows to have happened, and a working gallows would not have been proved by it.

Four civilians died on the grounds on the 6th. Ashli Babbitt was shot by a Capitol Police officer while climbing through the broken Speaker’s Lobby window. Kevin Greeson died of a heart attack. Benjamin Phillips died of a stroke. Rosanne Boyland collapsed in the tunnel; the medical examiner called it accidental amphetamine intoxication. Officer Brian Sicknick died the following day of strokes, manner natural. The story that he had been beaten to death with a fire extinguisher did not survive the autopsy. Four officers who had worked the site later killed themselves, weeks and months on, in a job that already loses more people off duty than on it. Those deaths belong in a discussion of the aftermath. They do not belong in a sentence that begins “killed at the riot.”

The only person killed by another person on the Capitol grounds that day was Babbitt. The officer was cleared. Dislike the clearance if you want. Do not rent cardiac events and delayed suicides to make a toy scaffold look load-bearing.

The racial soundtrack attached to that scaffold had the same problem of fit. A noose in American pictures arrives with a lynching history. The name in the chant was Mike Pence, which is an in-house punishment fantasy aimed at a very white, very churchy institutional Republican who would not halt the count. The rope’s résumé and the victim’s résumé were different documents. They were stapled together because the résumé was more useful than the victim.

Le’s blade is being given the other résumé: 1793, cake, aristocrats, a lesson for Congress, protected expression. Some of that may even be how he talks about it. It does not decide what the officer felt when he put a finger on the edge.

The useful comparison is not moral. It is taxonomic. One structure could not have done the thing it was said to threaten and was covered as if it could. The other looks as if it could do the thing its name announces and is being covered as a First Amendment puzzle. The comment section has already solved the inconvenience by remembering a guillotine on January 6 that was never there. Once the machines are the same in memory, the adjectives can move freely between them.

Staff went home on January 6. Members went home. The men who built the first structure walked off the lawn and stayed gone. The man who built the second sat on it until the police asked whether it would cut. Keep them in one paragraph. The rest is wardrobe.

Notes

Philan-Tam-Duy Le, 35, of Julian, California, arrested 25 August 2026; misdemeanor attempted possession of a prohibited weapon; released 26 August; trial listed 24 September.

January 6 structure: predawn assembly on Union Square, orange rope added about 1 p.m., builders not charged for the construction.

Congressional members and staff: not abducted or physically harmed. Injuries were among police.

Deaths on 6 January among civilians: Babbitt (gunshot), Greeson (heart), Phillips (stroke), Boyland (ruled accidental amphetamine intoxication). Sicknick died 7 January, strokes, manner natural.

Season 1, episode 3 of HBO’s Lanterns is “OutKast.” It stops the Nebraska mystery and spends the hour on John Stewart’s origin. The ring reaches his father first. One question: are you afraid? John Stewart Sr. says yes. Hal Jordan gets the job. That is a selection event. The show refuses to treat it like one.

Senior does not miss the cut and remain a man. He drinks. He shoots the television when the other candidate goes public. He stops being a father. Bernadette empties the gun and holds the house up. Then she sits a Guardian at the kitchen table and refinances the drop. Train the boy. Make him fearless. The universe owes the family because the father said he was scared and then came apart. The episode wants the wreckage to prove the question was illegitimate. The wreckage is the file.

Elite selection is not impressed by that reading.

Physical capacity gets you to the start line. What the course is built to measure is what remains when the body is used up. Sleep gone. Calories gone. Ruck heavier every day. In Delta selection the cadre go quiet. No times, no distances, no “good job.” A map point and a walk. Men who need a scoreboard panic and invent a reason to stop. Special Forces adds the misery of other miserable men and watches who stays useful to the team. Ground Branch wants the same software with the lights off and no unit around you. DEVGRU and the rest of that world are not hunting a perfect first try. They are hunting a mind that does not require a crowd, a speech, or a grievance to keep moving.

Dr. Eric Potterat, who spent a decade as the psychologist for the Navy SEALs and helped put mental toughness into that pipeline, has been making the same unromantic point for years: at the elite edge the hardware is close to even. Lungs, legs, strength. The split is software. Fatigue tells you that quitting is the rational, self-preserving choice. The people who pass treat that voice as a biological lie and take the next step anyway. Segment the trial. Stay in the present ugly minute. Do not negotiate with the part of the brain that wants a story. Potterat is not the pope of Delta or Ground Branch. He is useful here because he said out loud what every selection hut already knows and what this episode cannot afford to admit.

The people who do not pass are not all cowards. Some are injured. Some are timed out. Some DOR. The second test is the wake. Did he rehab and come back? Did he stay a professional in the unit that still had a job for him? Or did he disappear into a bottle and spend the rest of his useful years explaining why the course was unfair?

That is the column Senior occupies. He did not take an ambiguous no, put it in a box, and go back to work. He made the no his personality. Alcoholism after selection is not proof the standard was rigged. It is proof the man could not hold his identity when the external prize vanished. Intrinsic motivation is exactly that test. When there is no one cheering and no ring on the finger, what is left? In his case, a ruined household and a wife negotiating with aliens for a makeup slot.

SERE makes the distinction even cleaner, and the show fails that one too. Selection still lets you quit. You can ring the bell. You can drop the helmet. SERE Level C removes the exit. Isolation, noise, interrogation, no control over the room. The point is not whether you can pursue a goal while suffering. The point is whether you can keep your name when you cannot leave. Anger, panic, arguing with the interrogator—those are fails. Stoicism is not a lecture your mother gives you about the world. It is the ability to put fear in a box so the thinking part of the brain can still run. Senior never got to a resistance phase. He failed the part where you still have a house, a job, and a choice, and he chose collapse.

So the kitchen-table deal is the opposite of how you build a Lantern. You do not promote the son because the father shattered. You do not treat addiction as evidence the question was racist. You look at a man who could not metabolize rejection and you believe the data. John Stewart the Marine sniper can still be the real thing. Plenty of hard men come out of wrecked houses. What the episode will not admit is that forging a child as compensation for Dad’s washout is not grit. It is succession planning for a grievance.

Fearlessness is not inherited by speech. Durability after “no” is the tell. The schools that actually send people into the dark already know how to read it. They watch who tries again, who stays useful, and who falls apart and calls it theft. Senior told them which column he was in. OutKast just won’t read the column.

A man on Threads wore a shirt bragging that everyone eats from the hands of undocumented workers, then said America would let them all in no problem if they worked for free. That’s slavery, printed on cotton, with a hashtag. He got dragged for it. Good.

The mechanism, in numbers

Undocumented workers make up 40 to 60 percent of U.S. crop farmworkers depending on the year and the survey, and roughly 35 percent of all hired agricultural labor as of 2024, per the Congressional Research Service. In construction, 13 to 14 percent of the workforce is undocumented, and in the specialty trades—plastering, drywall, roofing—that figure climbs above a third. Hospitality runs about 7 percent. None of that labor can unionize, report wage theft, or call OSHA without risking removal, and the enforcement side of that arrangement is nearly nonexistent by design: from 2005 to 2023, worker arrests for immigration violations outnumbered employer arrests more than three to one, and fewer than 15 employers per year have faced prosecution for hiring undocumented labor since 1986. That is not an accident of under-resourcing. That is the enforcement ratio you build when you intend to punish the supply and never touch the demand.

Cesar Chavez ran his own border patrol in the 1970s to keep strikebreakers out, because an unorganizable workforce sets the wage floor for every legal worker standing next to it. He had the mechanism right fifty years before either party would say it out loud.

The Democratic number

E-Verify—the system that checks whether a new hire is authorized to work—is used by just 14 percent of U.S. employers nationally as of June 2025, per the Migration Policy Institute, even though MPI’s own estimate puts more than 9 million unauthorized workers in the labor force right now. It has been federal law on the books since 1997. Every Congress since has introduced a bill to make it mandatory nationwide. Every one has failed. That’s not a policy gap that opened recently—it’s a twenty-nine-year-old choice, renewed on a schedule, by both parties, in Congresses controlled by both parties.

DACA covers roughly 535,000 people. It has never addressed the other 9-plus million, and it was never structured to.

The Republican number

Self-deportation, the current administration’s cash-for-departure program, has paid out in three separate tranches: $1,000 at launch in May 2025, tripled to $3,000 as a year-end bonus in December, then set at $2,600 as the standing rate by January 2026. DHS’s own justification is explicit and numeric: the average cost to arrest, detain, and deport someone through normal channels is $18,245; buying a plane ticket and paying the stipend costs roughly $5,100, a claimed 70 percent savings. That arithmetic only works if you already know exactly who you’re targeting—the population with no criminal record, the only group DHS extends the offer to. Nobody buys out a dangerous person. You only buy out someone whose labor you were extracting a year ago and now need to be somewhere else without the transaction looking like what it is.

The domestic version, in numbers

You don’t need a border to run the same math. Roughly 800,000 of the 1.2 million people in state and federal prisons work, and the ACLU’s 2022 national survey found the average pay for non-industry prison labor runs 13 to 52 cents an hour—against a federal minimum wage of $7.25. In Louisiana, cellblock and field labor pays two cents an hour. Alabama, Arkansas, Florida, Georgia, Mississippi, South Carolina, and Texas pay nothing at all for most of it. Seventy percent of incarcerated workers surveyed said they couldn’t afford basic necessities on what they were paid. That labor generates at least $11 billion a year in goods and services for state governments and private contractors, according to the same ACLU report—which means the system isn’t failing to pay a living wage. It’s succeeding at not paying one, at a rate of $11 billion a year.

Where the shirt guy actually lands

He collapsed three systems into one sentence: we’ll take them if they’re free. Forty-thousandths of a dollar an hour in a Louisiana cellblock is free. Fourteen percent E-Verify compliance against 9 million unauthorized workers is a policy choice to keep it nearly free. A $2,600 buyout against an $18,245 removal cost is the state doing the arithmetic on how much the free labor was worth in the first place. He said it as a joke. The three systems above have been running the same math for decades without saying it at all.

The number nobody sits with

Mexico received $64.7 billion in remittances in 2024—more than its oil exports, tourism revenue, and foreign direct investment combined, according to Banco de México. The average individual transaction is around $400, sent by workers whose legal status gives them zero standing to negotiate their own wage. Ninety-five percent of that money comes from people working informal or low-wage jobs in agriculture, construction, domestic service, and manufacturing in the United States. Polite company treats that $64.7 billion as an unambiguous act of familial love, and it is. It’s also the receipt. Money that scarce, sent that far, by someone with that little legal leverage, is a direct measurement of how cheap the labor underneath it was bought for. Nobody prices a remittance as a virtue and the wage that produced it as an outrage in the same paragraph without noticing they’ve just filed two receipts for the same sale.

That’s the actual scandal, and it has a dollar figure. Not the shirt. The $11 billion in uncompensated prison labor, the $64.7 billion in remittances, the 86 percent of employers who never have to check a single hire’s status, and the 3-to-1 ratio of workers arrested to employers ever charged—four numbers, four different systems, one mechanism, and neither party has touched it because the number only gets worse for whichever side has to explain it first.

APPENDIX

Counter-arguments

Undocumented workers enter contracts voluntarily. That’s not slavery, that’s a job. True, and worth taking seriously rather than waving off. Nobody is kidnapped into a strawberry field. But “voluntary” is doing a lot of work in a labor market where the alternative to accepting sub-minimum wages under threat of deportation is deportation itself. A contract signed under the threat of the one leverage point the other party controls isn’t the same as a contract signed by someone with an exit. The 3-to-1 worker-to-employer arrest ratio is the actual measure of who bears the cost of that “voluntary” arrangement when it breaks down—it’s never the employer.

Compensated emancipation happened in D.C.—doesn’t that undercut the “the state never pays” framing? It happened exactly once, for about 3,000 people, and never scaled. The Confederacy rejected every broader compensated-emancipation offer Lincoln made, and the 13th Amendment freed everyone else without compensating anyone. One city-sized pilot program that never repeated isn’t evidence of a national buyout. It’s evidence the idea was tried, worked at a tiny scale, and was abandoned—which is closer to the piece’s point than against it.

Prison labor pays for room and board. It’s not comparable to chattel slavery. The 13th Amendment’s penal clause is the specific legal mechanism that makes prison labor exempt from the amendment that otherwise bans forced labor. That’s not rhetorical framing—it’s the literal text. Whether $11 billion a year in state and corporate revenue, extracted at 13 to 52 cents an hour from people who can be disciplined for refusing to work, constitutes “payment for room and board” or something closer to what the amendment was written to prohibit is exactly the question the penal clause was designed to make unaskable.

This is just “both sides” as a dodge to avoid picking a side. The piece isn’t claiming both parties want the same outcome—one wants the labor retained, one wants it removed. It’s claiming both outcomes get produced by the identical mechanism: a population kept legally powerless enough that its price can be set by someone else. That’s not equivalence of intent. It’s identity of method, and the method is the actual subject.

FAQ

Didn’t Democrats try with the 2013 Gang of Eight bill? Yes, and it passed the Senate 68-32 before dying in a Republican-controlled House. That’s a real attempt and a real failure that isn’t on Democrats alone. It also didn’t touch E-Verify enforcement, which stayed at roughly the same weak participation rate before and after the bill died, under administrations of both parties, for a reason that predates and outlasts that one bill.

Isn’t the self-deportation program actually more humane than a raid? By DHS’s own numbers, it’s cheaper—$5,100 against $18,245 for a standard removal—and it avoids the violence of a raid. It is also only offered to the population with no criminal record, which is the same population that was valuable specifically because it had no legal standing to negotiate a wage. Cheaper and more humane than the alternative isn’t the same claim as clean.

Is 40 to 60 percent undocumented agricultural labor even a stable number? No, and that instability is itself informative. USDA’s own data runs from 2020-2022 with nothing more recent officially published; CRS’s 2024 estimate lands at 35 percent of all hired agricultural labor; other estimates go as high as 60 percent depending on whether livestock and dairy are included. The range exists because nobody with the authority to fix the number has an incentive to pin it down precisely.

Numbers cited

• 35-60%—undocumented share of U.S. agricultural labor, depending on measure (USDA/NAWS 2020-2022; CRS 2024)

• 13-14%—undocumented share of the U.S. construction workforce; above one-third in specialty trades like drywall and roofing (American Immigration Council, 2022 data)

• 7.1%—undocumented share of the hospitality workforce (American Immigration Council)

• 3-to-1—ratio of worker to employer arrests for immigration-status violations, 2005-2023 (Migration Policy Institute)

• <15/year—employers prosecuted nationally for hiring undocumented workers in most years since 1986 (MPI)

• 14%—U.S. employers enrolled in E-Verify as of June 2025, against an estimated 9+ million unauthorized workers in the labor force (MPI)

• ~535,000—people covered by DACA, out of 9+ million unauthorized workers total

• $1,000 → $3,000 → $2,600—self-deportation stipend, May 2025 launch, December 2025 holiday rate, January 2026 standing rate (DHS)

• $18,245 vs. ~$5,100—DHS’s own stated cost of standard removal versus a self-deportation payout (DHS press releases)

• 13-52¢/hour—average U.S. prison labor wage for non-industry jobs, against a $7.25 federal minimum (ACLU, 2022)

• $0.02/hour—Louisiana cellblock and field-line pay (ACLU of Louisiana)

• $11 billion/year—value of goods and services produced by U.S. prison labor (ACLU)

• $64.7 billion—remittances Mexico received in 2024, exceeding oil exports, tourism, and FDI combined (Banco de México / BBVA Research)

• ~$400—average individual remittance transaction size (Banco de México)

• 95%—share of Mexico’s remittances sent by workers in informal or low-wage U.S. sectors (Wilson Center / El Colegio de la Frontera Norte)

Timeline

1865-1877—The 13th Amendment’s penal-labor clause becomes convict leasing almost immediately; by 1890, 90 percent of Alabama’s leased convict population is Black.

1962-1975—Chavez and the UFW run the wet line, physically blocking undocumented strikebreakers, because captive labor sets the wage floor for everyone.

1986—Reagan signs amnesty for 2.7 million people and adds employer sanctions that produce fewer than 15 prosecutions a year for the next four decades.

1996—Clinton signs the law that still defines mandatory detention and deportable offenses today.

1997—E-Verify launches. Every subsequent Congress introduces a mandatory-use bill. All have failed.

2001-2016—Bush builds the post-9/11 enforcement architecture; Obama runs it harder, deports a modern-era record, and creates DACA by executive order.

2013—Gang of Eight passes the Senate 68-32, dies in the House, fixes nothing on either side of the aisle.

May 2025—Self-deportation program launches at a $1,000 stipend.

December 2025—Stipend tripled to $3,000 for the remainder of the year.

January 2026—Standing rate set at $2,600.

Glossary

Deportability as discipline—a worker’s legal vulnerability doing the job a wage floor’s absence used to do, measurable in a 3-to-1 worker-to-employer arrest ratio.

Self-deportation—a payment DHS itself prices against an $18,245 alternative, offered only to the population with no criminal record.

The wet line—Chavez’s own 1970s border patrol against strikebreakers, the labor-left’s version of this argument, forty years before either party’s current framing.

The penal clause—the words “except as a punishment” in the 13th Amendment, currently worth $11 billion a year in uncompensated labor.

The remittance receipt—$64.7 billion a year, sent $400 at a time, by people whose legal powerlessness set the price of the labor that produced it.

A hypothetical can reveal a moral intuition. It cannot establish a prevalence, a cause or an unlimited public obligation.

A stranger on TikTok told me to try my argument on a starving child whose single mother works three jobs. He did not identify a child, a mother, a city, an employer, a rent bill, a missing father, a school, a church, a relative, a food bank, a benefits office or any reason every available form of assistance had failed. The mother and child entered the conversation fully assembled as a moral emergency. The mother had perfect industry, the child had perfect innocence, and society had perfect culpability. Their purpose was not to be found or fed. Their purpose was to make further questions indecent.

This family has circulated through American political speech for most of my 56 years. The details change with the decade, but the structure remains stable: one immaculate victim is placed at the farthest edge of suffering and then made to represent an enormous and varied population. The technique works because the underlying hardships are real. Single mothers do work multiple jobs. Children do miss meals. Families lose housing, fall through eligibility rules, flee abusive homes, lose documents, endure illness and encounter offices too understaffed or incompetent to help them. Yet the truth of each component does not authenticate every composite assembled from them. A hypothetical can reveal a moral intuition, but it cannot establish the prevalence, cause or remedy of the condition it depicts.

The anonymous starving child allows several claims to slide together. A household's anxiety about its grocery budget becomes starvation. The existence of sufficient food somewhere in America becomes proof that delivery is effortless. The existence of a public program becomes proof that every eligible person can use it. A moral conviction that government should provide a benefit becomes a claim that democracy has already settled the obligation. Those are separate propositions, and each requires its own evidence.

What food insecurity measures

Food insecurity is a real federal measurement, not a comparison with the richest child in town. The Department of Agriculture asks households whether they worried food would run out before they had money for more, whether purchased food failed to last, whether they could afford balanced meals, whether adults cut portions or skipped meals, whether anyone was hungry but did not eat, whether an adult lost weight for lack of money, and, in households with children, whether children ate too little, had meals reduced, skipped meals, felt hungry or went an entire day without eating. A household generally must affirm at least three conditions to be classified as food insecure, and progressively more severe answers place it farther along the scale.

The scale measures a continuum rather than a single physiological condition. USDA recognized this explicitly in 2006 when it replaced the label “food insecurity with hunger” with “very low food security.” Hunger is an individual physical experience; food insecurity is a household-level economic and social condition involving limited access to food. The National Academies emphasized that a person can remain food secure inside a household classified as food insecure because deprivation is not distributed evenly within every home. Parents commonly restrict their own diets before allowing their children's intake to fall.

USDA estimated that 47.9 million people lived in food-insecure households in 2024. It also estimated that 751,000 children, about one percent of American children, lived in households in which at least one child experienced very low food security. The severe category includes reports such as hunger, skipped meals or a child not eating for a whole day at some point during the year. Those children matter, and one percent of American children is not a trivial number. It is nevertheless false to turn every person living in a food-insecure household into a starving person or every child in such a household into a child whose body is wasting away. The broad headline borrows its emotional picture from the narrowest and most severe end of the scale.

Cheap food is not synonymous with insufficient food. Hamburger Helper, ramen, beans and rice, casseroles, powdered milk, day-old bread, inexpensive canned vegetables and a pound of meat extended across a family are ordinary forms of working-class household economy. The resulting diet may be monotonous, excessively salty or less nutritious than an affluent family's diet, but a stretched protein is not proof of starvation. A person can consume enough calories while receiving poor nutrition, and a person can suffer malnutrition because of disease, disability, neglect, an eating disorder or isolation even when food is physically abundant. Anxiety, reduced choice, low dietary quality, missed meals, physiological hunger, malnutrition and starvation describe different conditions. Public argument routinely uses one term while inviting the audience to imagine another.

The American abundance problem

America is extraordinarily rich in productive capacity and densely provisioned with public and private assistance, but national wealth does not place an affluent household's purchasing power in every citizen's bank account. A rich country can contain poor people just as a rich parent can have a child who does not independently own the family assets. National abundance establishes that severe deprivation is preventable. It does not establish that prevention requires no labor, administration, geographic compromise, cooperation or choice.

The claim that America grows enough corn to feed everyone is a good example of a true statement being asked to perform work it cannot do. Corn committed to livestock, ethanol and industrial supply chains does not become a balanced human diet because a politician points toward Iowa. It must be processed, combined with other foods, transported, stored, purchased or distributed and prepared. A surplus of calories cannot create an apartment in a neighborhood where housing construction is prohibited or ruinously expensive. It cannot replace identification, complete an application, treat schizophrenia, maintain sobriety, locate an absent parent or persuade a frightened person to trust an outreach worker. Productive capacity begins a delivery system; it does not complete one.

The United States already operates a vast, fragmented welfare state whose scale is easy to miss because it rarely arrives as one obvious cash payment. Depending on a person's age, disability, income, family composition and jurisdiction, assistance can include Social Security, Supplemental Security Income, SNAP, WIC, school breakfast and lunch, Medicaid, Medicare Savings Programs, housing vouchers, utility assistance, Lifeline telephone service, reduced transit fares, paratransit, senior nutrition programs, tax credits, food banks, church aid and municipal services. The Government Accountability Office identified 18 federal food and nutrition assistance programs alone, administered through multiple agencies and a decentralized network of states, localities and nonprofit providers. The complexity permits specialization but also creates overlap, gaps and a maze of separate applications.

Eligibility is not receipt. Federal research has repeatedly found people who qualify for benefits but do not participate. USDA has found that eligible older adults use SNAP at lower rates than other eligible groups. Social Security researchers estimated that more than a third of financially eligible elderly retired-worker beneficiaries did not participate in SSI. Some people do not know they qualify. Some cannot assemble the documentation, manage the application, tolerate repeated recertification or travel to an office. Some calculate that a small benefit is not worth the intrusion. Some distrust government, fear consequences for relatives, misunderstand the rules or refuse assistance because accepting it would violate their conception of independence. Pride is not a rounding error in a country built around self-reliance.

Libraries, community centers, universities, churches and nonprofit organizations respond by setting up folding tables and trying to persuade people to accept help that has already been funded. They offer tax preparation, FAFSA assistance, grant searches, benefits screening, discounted laptops, medical enrollment and professional guidance. The recurring obstacle is often locating an eligible person, convincing that person the offer is genuine and completing the application. An agency can truthfully say that help is available while an eligible person truthfully says that nobody helped. Money can be appropriated but unspent, a voucher can be authorized but unusable, and food can be abundant but unknown to the person who needs dinner. A program recorded on a ledger has not necessarily become a delivered service.

The meal as an entrance

For seven years, I served as a Sous Chef and Dining Room Captain at Miriam's Kitchen in Washington, D.C., an interfaith-rooted but operationally secular organization that served nutritious meals and connected guests with professional assistance. Anyone could eat without proving homelessness, poverty or citizenship. There was enough food for everyone who arrived, although there were no seconds, and the dining room included unhoused people, undocumented construction workers and even touring musicians who had discovered the free meals. The food met an immediate need while giving case managers time to build trust and offer help with identification, benefits, healthcare and housing.

The structure repeats across charitable traditions. Christian kitchens may offer Jesus, fellowship and church community along with dinner. Hare Krishna kitchens may offer food within a devotional practice. A secular meal program offers entry into case management and the administrative safety net. The food is real and the generosity is sincere, but the meal also functions as a humane lure for people who would never walk into an office and announce that they need a social worker. Repeated hospitality lowers suspicion and creates the relationship through which deeper assistance becomes possible.

The absence of endless queues outside an open dining room does not prove that nobody elsewhere is hungry. It raises the question of visibility. A kitchen may keep a low profile to protect guests, avoid overwhelming limited staff or preserve its relationship with the surrounding neighborhood. Existing clients may share information selectively. People outside the service network may assume that meals require proof they cannot provide. An open kitchen can have enough food for everyone who arrives while unknown people elsewhere remain hungry. The missing queue may indicate adequate capacity, inadequate promotion, gatekeeping, geographic inconvenience or several of those conditions at once.

The case managers, outreach workers and administrators move assistance across the final human distance. They locate people, earn trust, replace identification, document eligibility, complete applications, coordinate medical care, negotiate with landlords, preserve appointments and begin again after a client disappears or refuses. This work is slow, skilled, repetitive and emotionally punishing. It cannot be replaced by declaring food or housing a right because every declared right still requires people capable of delivering it to another particular person.

There is no surplus of deployable help

The claim that America has produced more social workers than it has people to serve is not supported by the labor market. The Bureau of Labor Statistics counted about 816,100 social-worker jobs in 2025 and projects approximately 69,100 openings each year through 2035, many caused by transfers and retirements. Those workers are distributed across child and family services, schools, hospitals, mental-health and substance-use treatment, government agencies, outpatient practices and many other settings. A licensed clinical social worker providing reimbursable therapy in a hospital or private practice is not interchangeable with an outreach worker walking an encampment, even when both hold related credentials.

The problem is distribution and retention. Frontline homelessness work commonly offers lower pay, heavier caseloads, less safety, more secondary trauma and fewer predictable hours than hospitals, schools, corporate programs or private clinical practice. A city can contain thousands of trained helpers while failing to fill positions that require daily street outreach and months of case management. Credentials are potential capacity. Only funded, hired, retained and appropriately assigned workers become usable capacity.

The Government Accountability Office reported in 2026 that the joint HUD-VA supportive-housing program had persistent case-manager vacancies. More than a quarter of VA medical centers with at least two case managers had vacancy rates of 20 percent or higher in fiscal 2024, while annual turnover ran between 20 and 26 percent from 2020 through 2024. GAO found 174,045 instances during those years in which eligible veterans were not referred into the program, with the reason undocumented in 87 percent of cases. Earlier GAO work identified case-manager shortages, housing costs, limited housing stock and transportation as barriers even inside programs designed specifically for homeless veterans.

Washington presents the same paradox outside the veterans' system. The District returned more than $10 million in federal grants intended to end homelessness between 2017 and 2021. In 2024, Street Sense reported that 800 people who qualified for housing vouchers could not begin the process because there were not enough workers. Money existed, need existed and eligibility existed, but the bridge between them did not. This evidence refutes both slogans. Visible deprivation does not prove that society refused to fund assistance, and abundant funding does not make every remaining case willful.

Agency and obligation

Modern poverty rhetoric often treats agency as an accusation. If a person mentions refusal, pride, addiction, family breakdown, location, criminal behavior or repeated bad judgment, the observation is received as an attempt to revoke compassion. The result is an oddly dehumanizing defense of vulnerable people in which adults possess no meaningful capacity to choose, resist, conceal, cooperate, manipulate, recover or fail. Parents and children become equally passive victims of a system, while friends, relatives, churches, absent parents and local communities vanish from the account.

Agency does not mean equal capacity. A psychotic person refusing shelter, a mother hiding from an abuser, an addict protecting access to a drug supply, an undocumented worker fearing official attention, an elderly person too proud to apply for SNAP and a healthy adult rejecting a cheaper city are making different choices under radically different constraints. Policy must distinguish incapacity, constrained choice and ordinary preference because each requires a different response. Some people need an uncomplicated benefit. Some need persistent case management. Some need psychiatric treatment or guardianship. Some need enforcement against a responsible parent. Some need permission to fail until they are ready. Some need to hear that public obligation has limits.

Housing intensifies the problem because housing exists somewhere rather than everywhere. A society can guarantee access to safe shelter without guaranteeing a permanent private dwelling in every neighborhood a person prefers. If affordable housing and employment exist in one region while another has prohibited construction and produced extreme prices, relocation belongs in the policy conversation. Otherwise a right to housing silently becomes a right to command an unlimited subsidy at a chosen location. A refusal may be understandable without creating an unlimited duty to redesign every available offer around it.

Democracy may say no

The TikTok argument eventually rested on the word “democratically,” as though democracy were an escalator carrying every decent society toward greater redistribution. Democracy is a procedure for resolving political conflict among citizens who do not share one moral theory. The same electorate can expand nutrition assistance, impose work requirements, build shelters, clear encampments, welcome immigrants, deport noncitizens, subsidize expensive cities or tell people to move somewhere they can afford. Donald Trump's elections are sufficient proof that voters do not reliably follow the progressive arc assigned to them.

Voters respond to compassion, but they also respond to reciprocity, taxes, disorder, borders, fraud, proximity, personal responsibility and the conduct of recipients. They may conclude that existing programs are adequate and poorly administered. They may believe that assistance should concentrate on children, disabled people and those least capable of caring for themselves. They may decide that able adults owe work, that parents owe their children support, that relatives bear some responsibility before strangers do, or that a benefit available in one place need not be reproduced at unlimited cost in another. Those positions can be wise or cruel, but they remain possible democratic outcomes.

Anyone who uses democracy as the source of an unlimited obligation faces a logical problem. If democratic consent creates the duty, democratic refusal can restrict it. If the duty persists regardless of what voters decide, then it comes from a prior moral principle rather than democracy. The speaker may defend that principle, but the word “democratically” does not settle its scope, identify who must pay, determine where a benefit must be delivered or establish whether a recipient owes anything in return.

From avatars to cases

The starving child with the three-job mother is effective because nobody can investigate the family. There is no case file showing income, benefits, family connections, child support, rent, medical conditions, immigration status, EBT balance, school meals or previous offers of help. Every question can therefore be treated as cruelty, and every policy preference can be placed in the child's outstretched hand. The character converts compassion into a blank check while protecting the argument from the complications that accompany every real person.

A serious politics of deprivation begins with classification and particulars. Is the child short of calories, deficient in nutrients or eating repetitive inexpensive food? Is the household eligible for assistance, enrolled but underserved, administratively excluded, unaware, ashamed or refusing? Is the binding constraint income, housing supply, documentation, staffing, transportation, illness, addiction, family violence, location or judgment? Which adult is responsible, which institution has authority, which intervention can reach the family, and what happens if assistance is rejected? These questions are the route from displaying suffering to reducing it.

America has enough food and wealth to prevent starvation and to relieve a great deal of hardship. It also has an enormous safety net, incomplete knowledge of who can use it, a shortage of people willing to perform its most difficult work, citizens who miss or refuse what is offered, and voters who may reject demands for more. Abundance, delivery, agency and democratic consent are separate problems. The anonymous child erases all of them. A real child forces us to solve them.

Sources

U.S. Department of Agriculture, Economic Research Service. Food Security in the U.S.: Measurement

U.S. Department of Agriculture, Economic Research Service. Food Security in the U.S.: Key Statistics & Graphics (2024 data)

National Research Council. Food Insecurity and Hunger in the United States: An Assessment of the Measure (2006)

U.S. Government Accountability Office. Domestic Food Assistance: Complex System Benefits Millions, but Additional Efforts Could Address Potential Inefficiency and Overlap (GAO-10-346)

U.S. Bureau of Labor Statistics. Occupational Outlook Handbook: Social Workers (2026 edition)

U.S. Government Accountability Office. Veteran Homelessness Programs (GAO-26-107517)

U.S. Government Accountability Office. Homeless Veterans (GAO-20-428)

U.S. Interagency Council on Homelessness. Scale Housing and Supports That Meet Demand

Street Sense Media. DC returned millions of federal grants to end homelessness (June 7, 2023)

Street Sense Media. D.C. has everything it needs to fight homelessness—except the people (January 15, 2025)

Social Security Administration. Benefit Adequacy Among Elderly Social Security Retired-Worker Beneficiaries and the SSI Federal Benefit Rate

A single train track can move as many people as a sixteen-lane highway.

Fine.

Nobody cares.

People do not choose transportation by asking which system moves the most strangers across the smallest strip of land. They ask how much it costs, how long it takes, whether it goes where they are going, whether they can bring their things, and whether the experience is pleasant.

Transit advocates keep answering a question passengers did not ask.

They show us diagrams of throughput and land use. They talk about density, carbon, traffic, induced demand and the collective good. Then they act surprised when a middle-class family looks at the train, looks at the car already sitting outside, throws everything into the trunk and drives away.

The train may be efficient for the system.

The car is efficient for the person.

That difference is the whole argument.

I have taken the good trains

I have traveled around the world and ridden plenty of trains. I have taken the TGV in France. I lived in Berlin and regularly used both the U-Bahn and S-Bahn. I lived in England and used the Underground and intercity rail.

I am not an American who saw one sad Amtrak platform and mistook it for the entire history of rail.

The good trains are good.

They still do not compare with the private luxury of a modern car.

Nothing on a train is remotely as personal. Nothing is as convenient. Nothing is as bespoke. First class is still a seat or compartment aboard somebody else's vehicle, traveling on somebody else's route, according to somebody else's timetable, between somebody else's stations.

Once a train reaches first class, the only thing left is to throw more butlers at it.

Better food. Better upholstery. More attentive service. A private compartment. Another person in a uniform asking whether sir would like something else.

The train can improve the service forever. It cannot become your vehicle.

There is no private jet of rail travel. The next level above first class is a private jet, a helicopter or a car. Of those three, the car is the one ordinary people can own.

The exception is one of those magnificent historical sleeper trains crossing Europe or Asia, where everyone dresses for dinner and a Belgian detective eventually finds a body. But that is not competitive transportation. The train itself is the vacation.

My ordinary BMW beat the Acela

You do not need a $200,000 car to beat first-class rail.

I owned a 2001 BMW 530i. I paid about $50,000 for it. It was not some absurdly appointed modern flagship. It was a returned corporate fleet vehicle with basic trim. It had heated seats, a radio, a CD player and German quality. It did not have a giant screen, ventilated seats, massage functions or any of the electronics now found in expensive cars.

It was still ten times better than the Acela as a personal travel experience.

It was private. It left when I wanted. It went directly where I wanted. The temperature was mine. The music was mine. The route was mine. The people inside were people I had invited.

And it had a trunk.

This sounds trivial until you compare it with carrying luggage through a station.

A car is a private room with a motorized storage unit attached. You can bring coats, medicine, chargers, tools, extra clothes, drinks, food and everything else you might conceivably need. You throw it into the trunk, lock the car and forget about it. It can stay there for the entire journey.

You do not drag it across a concourse. You do not lift it into an overhead rack. You do not keep one eye on it while trying to sleep. If you need something, you stop the car, open the trunk and retrieve it.

If I travel with one other person, I also have the entire back seat. If I take an SUV, the comparison becomes ridiculous. I owned a 1999 Jeep Grand Cherokee Laredo and a 1997 Land Rover Discovery. Imagine how much private comfort, food and equipment can be packed into either one.

Rail charges for every passenger.

The car does not care whether the other seats are occupied.

A family of four buys four train tickets. In a car, the spouse, children, cooler and luggage ride for little more than the marginal cost of fuel and wear.

The private dining car

The supposed culinary advantage of first-class travel is equally silly.

There is a magical device called a cooler.

There are also specialty food stores.

Dean & DeLuca used to occupy the old Georgetown Market on M Street. I could walk in, spend $200 on a bespoke picnic, put it in a cooler on the back seat and leave with better food than almost any transportation company could serve me.

Fresh fruit. Vegetables. Drinks. Prepared food. Snacks made by people who devoted their lives to making snacks. Whatever I wanted, selected before the journey rather than from a menu designed for storage, reheating and distribution to hundreds of passengers.

My family and I could serve one another. Everyone could play butler and everyone could be treated like a first-class passenger.

We could ignore every McDonald's and roadside junk-food stop and pull over only when we needed fuel, a bathroom or a beautiful place to eat.

The train's answer to luxury is upgraded service.

The car's answer is to bring the entire luxury experience with you.

When trains actually work

My Pop-Pop commuted from Spring Lake, New Jersey, to New York by train.

He did not do it because a transit advocate showed him a drawing of a sixteen-lane highway. He did not accept inconvenience as an act of environmental penance.

He took the train because it was less of a pain in the ass.

He could park near his origin station, board a predictable commuter train and avoid traffic, driving fatigue, mechanical worries and the nightmare of parking in Manhattan. His job may have helped pay for the train more readily than it would have paid for his car. At the other end, he arrived in a walking city.

From the 1950s through the end of the twentieth century, that bargain often made sense. Park the car. Take the train. Arrive in a dense city where the rest of the trip can be completed on foot.

No sermon was required.

The train won because it was convenient.

The bargain collapses when the destination is not concentrated around the terminal. Union Station is useful if you are going somewhere near Union Station. If you must then take Metro, find a bus, call an Uber or spend another forty-five minutes reaching somewhere in the wider DMV, the train has not replaced the car. It has inserted itself into the middle of a more complicated journey.

People will surrender automotive luxury when a train is genuinely easier.

They will not surrender it because altruism is supposed to feel virtuous rather than inconvenient.

Too expensive for the poor

American rail has an astonishing class problem.

It is too expensive for poor people. Rich people generally do not want to use it. Middle-class people own middle-class vehicles that outclass almost everything about trains, buses, undergrounds, above-ground rail and trolleys.

The truly price-sensitive intercity traveler often takes a bus. The salaried suburban professional takes commuter rail into a downtown office, sometimes using an employer transit benefit. The affluent leisure traveler buys Acela or first class. The family looks at the price of four tickets and takes the car.

Anything running on a timetable must also compete with taxis, Uber and the coming fleet of autonomous taxis.

A taxi charges more than a bus fare, but it picks you up where you are and takes you where you are going. Once waiting, transfers, missed connections, parking, luggage and the value of time are included, the price difference becomes less impressive.

With ADHD, I will take Uber when I must be somewhere at a specific time. The extra money buys certainty. I do not have to calculate when to leave, reach the stop, estimate the wait, protect the transfer and continually wonder whether one delay will ruin the appointment.

Transit planners see a cheap fare.

The passenger sees the cost of being late.

Autonomous taxis will make the comparison more brutal. Remove much of the driver's labor cost and a private vehicle summoned on demand begins competing directly with buses and trains while retaining door-to-door service.

If government wants people to use fixed transit despite those disadvantages, it will have to make transit free, attach a meaningful tax benefit to it or pay people to ride.

Otherwise the tunnels and tracks will die from their own lack of momentum.

If we want to keep them, public money must keep them alive. Fare-paying passengers cannot be expected to sustain a system that charges them for accepting inconvenience.

My free funicular

During COVID, Metrobus temporarily stopped collecting fares.

That was the only time in my life I voluntarily used the ubiquitous Number 16 Pike Ride bus with any regularity.

I treated it as a free Columbia Pike funicular.

If I took Metro into Washington, I could return through Pentagon or Pentagon City, transfer to the bus and let it haul me up Columbia Pike toward Courthouse Road. If I was riding my bicycle and did not feel like climbing the hill, I could stop at a bus stop, put the bike on the rack at the front and ride home.

The bus was not replacing my entire journey. It was solving one specific problem. It was a hill-climbing machine, a bicycle rescue vehicle and a complimentary final leg home.

At zero dollars, its compromises became charming.

At zero dollars, it became fun.

That is what transit advocates refuse to understand. People will use public transportation when it serves them. My Pop-Pop's train succeeded because it served him. It did not require him to serve the train.

D.C.'s Puritan commuter appliance

D.C. Metro is beautiful and joyless.

It was built by Puritans for Puritans.

No eating. No drinking. Little retail. Little busking. Few cafés, kiosks, newsstands, pizza counters or reasons to remain underground once the train arrives.

In Berlin and New York, stations participate in daily life. There are stupid little stores everywhere. You can buy food, coffee, newspapers and whatever else somebody has figured out how to sell to a captive stream of pedestrians. The station becomes a marketplace, a meeting place and shelter from weather. It belongs to the city around it.

D.C. Metro is a sterile corridor.

Stand quietly. Consume nothing. Go to work. Leave.

Its natural populations are tourists, students, hardcore commuters and young nightlife passengers who cannot afford taxis, parking or a DUI. It never fully became a neighborhood or a pleasurable third place.

The architecture is magnificent. The culture is disciplinary.

D.C. did not merely build transit that struggles to compete with cars. It stripped away the incidental pleasures that might have made people enjoy using it.

Fun beats luxury

Rental bikes, e-bikes and scooters reveal the alternative.

They are not luxurious. They expose you to weather. They carry almost nothing. They are dangerous. You pay for the privilege of being vulnerable.

They are also freaking fun.

I spent years riding an XR650L and then a BMW K1100LT. Motorcycles taught me that people will surrender comfort, climate control, storage and physical protection when the journey itself becomes exhilarating.

Rental micromobility makes a piece of that experience available without ownership. Unlock a scooter or bicycle with a phone, ride it one way and leave it at the destination. No garage. No maintenance. No return trip. You retain your own route, steering, departure time and personal space.

It is individualized transportation delivered with the spontaneity of public transit.

It also lets the rider say the trip is good for the environment, which adds virtue to something already enjoyable.

Then municipalities ruin it with geofences, no-parking zones and jurisdictional boundaries. Cross an invisible line between D.C. and Virginia using the wrong rental fleet and the motor may slow or die while the app refuses to end the ride. You must drag or pedal the machine back into the approved zone so you can park it legally.

The government finally found a form of non-car transportation people might choose because it is fun, then converted it into a rolling compliance test.

People will surrender luxury for convenience, economy or fun.

What they will not surrender it for is a lecture.

Making less feel posh

Europe has become exceptionally good at making less feel posh.

Smaller becomes minimalist. Crowded becomes communal. Expensive becomes sustainable. Inconvenient becomes intentional. Walking becomes civilized. Taking the train becomes cosmopolitan. Private comfort becomes vulgar.

This is not an insult to every European or every train. Dense European cities were built around walking and rail. Fuel, vehicle ownership and parking are expensive. In those places, the train often wins honestly.

But Americans are being sold European material constraints as aspirational lifestyle upgrades without first being given European geography or functional transit.

Japan offers an even sharper contrast.

At the system level, Japan looks like the future: immaculate stations, Shinkansen trains, vending machines, convenience stores, punctual service and dazzling cities.

At the individual level, many people accept tiny apartments, little storage, long workdays, long commutes and rush-hour trains packed with human beings.

Japan is magnificent at collective technological abundance and austere at private material abundance.

America is the reverse. We express prosperity through private autonomy: large homes, large refrigerators, cars, garages, spare rooms, yards and enough storage to forget what we own.

Japan offers an extraordinarily futuristic life, provided you are willing to live like a monk, commute like a sardine and really, really love katsu.

The most effective social engineering does not force people to accept material constraint. It gives constraint a flattering identity.

The most successful social engineer does not make poverty compulsory.

He makes it prestigious.

The choice

Affluent institutions want Americans to consume less energy, occupy less space, drive less, own fewer things and rely more heavily on shared systems.

They face a cultural problem.

Americans understand prosperity as the ability to control one's private environment. Your house. Your car. Your schedule. Your food. Your tools. Your luggage. Your route. Your ability to leave.

If the promised future gives ordinary people less while powerful people retain private jets, chauffeured cars, large homes and protected privacy, Americans will not experience it as progress. They will experience it as hierarchy.

Transit can still win.

It can be cheaper. It can be free. It can be reliable. It can connect places people actually inhabit with places they actually need to go. Stations can contain food, shops, music and life. Buses can carry bicycles up hills. Trains can eliminate the misery of driving into dense cities. Rental bikes and scooters can remain fun instead of being electronically killed at municipal borders.

But transit must stop demanding admiration for its altruism.

It must become useful.

Austerity can be adopted as prestige or imposed as punishment. Americans are unlikely to choose it merely because someone displays a diagram showing how many strangers fit on a railway track.

The train used to say, "Take me because I am easier."

Now its champions say, "Take me because I am better for everyone."

That is a much weaker sales pitch.

I live in Arlington, Virginia, where quarter-million-dollar cars sit on the street with the emotional significance of Toyota Tercels. My optician owns a white BMW 440i convertible with a red interior. He parks it outside. People walk past it. Nobody cuts the canvas roof, breaks the glass or drags a penny down the Alpine-white paint. It is simply a car belonging to a man who goes to work, sitting among all the other expensive cars belonging to people who go to work. I lived in England during a period when that kind of display would have felt like an invitation to public correction. Cars were broken into, stolen, keyed and pennied. A white convertible with red leather would not merely have been noticed. It would have been answered. America has plenty of carjackings, catalytic-converter thefts and ordinary property crime, but there is remarkably little generalized outrage at the existence of nice things. The only cars in recent memory that people have attacked as political symbols are Teslas, and that was not an uprising against expensive automobiles. It was an argument with Elon Musk conducted on somebody else's door panel.

This ordinary tolerance for extraordinary property is one of the least appreciated facts about the United States. America does not merely contain billionaires, millionaires and a handful of families behind gates. It contains an enormous affluent class whose wealth has become so commonplace that it often cannot recognize itself. About one in six American households now earns at least $200,000 a year. Add the households earning somewhat less but holding substantial home equity, retirement accounts, pensions, businesses, professional licenses and valuable credentials, and the constituency with something important to lose becomes enormous. The United States has not stabilized capitalism solely by constructing fortresses for plutocrats. It has installed baffles throughout the fuel tank. Tens of millions of households have enough exposure to the machinery that no political wave can travel cleanly from one end of the country to the other.

The household earning $225,000 does not compare itself with a British household, a French household or the overwhelming majority of people who have ever lived. It compares itself with the neighbors earning $400,000, the managing partner earning $800,000, the family paying three private-school tuitions without visible distress and the couple who renovated the kitchen before spending August in Italy. It therefore experiences abundance as inadequacy. This is the Joneses paradox: the richer a mass-affluent society becomes, the more ordinary affluence feels, because people measure themselves horizontally against richer neighbors rather than vertically against history or the world. The United States now sits with Luxembourg and Norway among the OECD countries with the highest median disposable household incomes after adjusting for purchasing power, while Britain, one of the ancestral centers of Western wealth, produces roughly as much per person as Mississippi, America's poorest state, depending on the year and exchange rate. Yet the American professional family still says it is barely middle class because the family across the street has a pool.

The affluent class does not merely own better objects. It consumes time, distance and experience differently. One of my clients lives in Jacksonville Beach, already a place where another family might save all year to spend one week. During a single summer he took his children to several places in Europe, the mountains, the ocean and a lake. Any one of those trips would constitute a major vacation for a normal household. For another person, each might be a separate item on a lifetime bucket list. For his family, they were summer. This is bucket-list compression: the conversion of experiences one class regards as singular lifetime achievements into the annual routine of another class.

Televised America is affluent America. The camera at Wimbledon, the Masters, the World Cup, an NFL game or a Taylor Swift concert is not pointed at a random sample of society. It is pointed at people who possess some combination of ticket money, transportation, accommodation, childcare, schedule flexibility, advance-planning capacity and enough remaining energy to turn leisure into an event. Some saved for years or bought the worst seat in the building, but the visible public is disproportionately affluent. Everyone else is working, commuting, recovering, watching from home or discovering that the parking costs as much as a weekly grocery run. The people most frequently shown living American life are therefore the people least representative of what American life costs. Social media intensifies the error by breaking exceptional lives into daily installments until six vacations, a championship game, a destination wedding and an arena concert look like an ordinary year.

This matters because institutions build for the people who arrive, pay and appear in the photographs. America increasingly designs public life around people whose Tuesdays would be other people's bucket lists. Those people are portable, exquisitely comparison-aware and surrounded by outside options. They do not have to attend the regional university, visit the nearest beach, enter the local stadium, shop on the high street or remain in the city where their employer happens to have an office. They can leave. The organizing economic fact of modern American life is not simply wealth. It is affluent mobility.

When the city could make you come

I started talking about this during the dot-com years, when remote work first became semi-feasible and everybody briefly imagined that geography had died. Geography did not die, but obligation weakened. The old American city was an obligator. If you were a banker, publisher, advertising executive or elite lawyer, New York possessed something you could not acquire elsewhere. If you worked in film and television, Los Angeles held the machinery. San Francisco offered technology and a kind of metropolitan gay life that could not be ordered through the mail. Washington held government. Chicago held commodities and Midwestern corporate power. Boston held universities and medicine. You tolerated the rent, weather, commute, crime, tiny apartment, provincial customs and municipal contempt because the city had the key to your career, community or identity. It did not need to seduce you. You needed it.

Traditional universities possessed the same authority. When I attended George Washington University from 1988 through 1990, it was already among the most expensive universities in America, but nobody confused the price with a resort charge. Freshmen were packed four to a suite. We ate cafeteria food. The dormitories were institutional. Classrooms had chalkboards and overhead projectors. Some buildings were architecturally handsome, but the interiors did not perform hospitality. The university's proposition was simple: we possess something valuable, and you will endure the conditions required to receive it. The experience retained something monastic. The price purchased access, not pampering.

Paris still behaves this way. Paris does not particularly care whether you find Paris convenient. Berlin expects you to learn Berlin. They possess destination sovereignty, the power of a place or institution to make the visitor adapt because the place itself remains the attraction. Even Disney retains more destination sovereignty than most contemporary universities. Families pay thousands of dollars to walk fifteen or twenty thousand steps through Florida heat, stand in queues, follow reservation schedules and organize their bodily needs around a theme park because Disney can still say that the magic exists here and nowhere else. The visitor suffers for Disney.

The old university could say the same thing. The old city could say the same thing. The old seaside town did not need a strategic positioning exercise because the people nearby went there. English families went to English seaside towns. Families from Pennsylvania and New Jersey went to the Jersey Shore. Midwesterners went to the Great Lakes, the mountains or a cabin. Children went to camp while their parents worked. A place possessed a captive audience created by distance, expense, custom and inconvenience.

Cheap aviation, passports, remote work, digital communities and global comparison dissolved much of that protection. The Jersey Shore no longer competes merely with another town on the Jersey Shore. It competes with Cancún, Mallorca, Bali and an algorithmically perfect rental in Croatia. An English seaside town competes with Greece, Spain and Turkey. A stadium competes with the enormous television in the living room. A restaurant competes with delivery. An office competes with the bedroom. A regional university competes with Arizona State, a coding boot camp, four years abroad, online education and simply remaining at home.

Once necessity disappears, seduction becomes policy.

The affluent Dust Bowl

The original Dust Bowl moved people whose land could no longer support their labor. The affluent Dust Bowl moves people whose labor no longer requires any particular land. Its migrants do not arrive with mattresses tied to trucks. They arrive with laptops, home equity, professional salaries and a detailed list of lifestyle requirements. They ask a question that earlier generations of workers could rarely afford to ask: if my income no longer requires me to live here, where would I actually choose to live?

Remote work never became universal, and Manhattan, Hollywood, Silicon Valley and Washington retain real concentrations of power. That is beside the point. A disruption does not need to liberate everybody before it changes the behavior of every competitor. Roughly a quarter of paid working days in America are still performed from home, and the workers with the greatest geographic discretion are disproportionately educated, highly paid and attractive to tax collectors, property developers and universities. The mere possibility of their departure changes the market.

Austin offered technology without San Francisco. Nashville offered music, restaurants and metropolitan excitement without New York. Miami offered finance, internationalism and nightlife without winter. North Carolina offered universities, medicine and technology with trees and larger houses. Bend offered a Colorado-flavored outdoor life without being Colorado. Texas, Florida, Idaho, Montana and Wyoming marketed different mixtures of space, taxes, climate, scenery, privacy, politics and access to airports. People who once moved where the factory, bank or studio required them began sorting themselves by preference.

Poor people remain attached to place by exactly the constraints that make the affluent portable: rent, family care, health, unreliable transportation, insecure employment, moving costs and the absence of a cash cushion. The affluent Dust Bowl is therefore not the liberation of society from geography. It is a geographic privilege exercised by the people most valuable to institutions. They do not merely change addresses. Their purchasing power changes the destination. They raise housing prices, alter restaurants, reshape schools, demand direct flights, import cultural expectations and sometimes overwhelm the political settlement that made the destination attractive in the first place. A laptop migrant can transform a mountain town without opening a factory or employing anyone in it.

Cities responded as startups do at the beginning of a platform war. They spent ahead of revenue to acquire the customer before the market settled. Waterfronts became entertainment districts. Warehouses became creative quarters. Food halls arrived. Stadiums, trails, cultural venues, airport connections and mixed-use developments promised a complete lifestyle. The old city said, "You must come here." The new city asks, "What would persuade you to stay?"

Universities entered the same land grab because they were pursuing the children of the same households. The modern campus offers luxury housing, elaborate recreation, wellness infrastructure, professional studios, global programming, branded social experiences and a curated community representing the whole world. This is regularly described as an amenities arms race, but that phrase makes it sound like several foolish deans became distracted by climbing walls. The buildings are customer-acquisition expenses. Universities are spending desperately to become attractors because they can no longer rely on being obligators.

The risk is obvious. Austin, Nashville and Miami can all enjoy a winning decade, but hundreds of regional colleges cannot all become national destination universities. Nor can every old industrial town become an arts-and-food capital. Institutions borrowed and built as though market dominance awaited, often while enrollment softened and demographic pressure approached. Many will discover that the customers remain mobile after the ribbon cutting. A leaking cabin at a century-old legacy camp can be sold as character. A leaking dormitory at a regional university is insolvency.

Obligation supported austerity. Mobility requires seduction.

When America became the attraction

The 2026 World Cup provided a useful foreign inspection of what Americans had stopped seeing. International visitors arrived expecting political disorder, inadequate trains, expensive healthcare and perhaps some very large people. They encountered all of those arguments, but they also encountered the casual, repetitive material abundance of American life: enormous stadiums, gigantic food choices, suburban houses, pickup trucks, air conditioning, free refills, hospitable strangers and convenience stores that behaved like regional amusement parks. European fans described American stadiums as spaceships. The revelation was not that America possessed a handful of magnificent buildings. Everybody knew America was rich. The revelation was how far the hospitality machinery extended downward.

Some Texas public-school districts possess athletic facilities that look absurd beside professional clubs elsewhere. Allen High School plays in an eighteen-thousand-seat stadium that cost roughly $60 million and includes a weight room, wrestling space and an indoor golf facility. Melissa High School has a 134,200-square-foot indoor championship center with a turf field, an eight-lane track, nine locker rooms, treatment and training rooms, film-review classrooms, wrestling rooms and VIP meeting areas. Norwich City Football Club has professional footballers, a twenty-seven-thousand-seat ground and more than a century of history. Melissa has a championship center larger than certain university facilities because it occasionally becomes hot in Texas.

This does not mean an American teenager receives better football instruction than a Norwich City academy player, or that a Texas high school has surpassed Real Madrid. It means that American mass affluence pours astonishing amounts of capital into the visible environment of ordinary people. Europe often concentrates grandeur in inherited centers, ancient institutions and national monuments. America manufactures mass-market magnificence at replication scale. A European cathedral may overwhelm an American church, while an American high school overwhelms a European professional training ground. The distinction is not simply rich versus poor. It is where wealth becomes visible and whom the institution believes it must impress.

American sports treat the spectator as a customer whose attention must be recaptured between every play. The game requires a giant screen, programmed music, branded contests, premium seating, elaborate concessions, perfect sightlines, spotless bathrooms and a procession of upgrade opportunities. The baseball game needs fireworks, bobbleheads and a concert. The apartment building needs a rooftop pool, coworking lounge and podcast room. The supermarket needs prepared-food stations, theatrical product displays and enough square footage to consume a European town square. The university needs resort housing, Olympic recreation, celebrity speakers, international cuisine, curated adversity and a four-year social world more complete than the one outside its gates.

America's great innovation was not building more Disney Worlds. It was franchising the functions of exceptional destinations across ordinary life. Disney supplied immersion, legibility and family-safe spectacle. Las Vegas supplied hospitality, indulgence and continuous stimulation. Hollywood supplied fantasy, reinvention and production value. Manhattan supplied density, status, cultural variety and the promise that something interesting was always happening. American cities, campuses and lifestyle developments recombined those functions without retaining all the inconveniences of the originals. They built Manhattan without the rubbish bags and subway, Hollywood without rejection, Vegas without visible vice and Disney without compulsory mouse ears.

Eventually they began to out-Disney Disney, out-Vegas Vegas, out-Hollywood Hollywood and out-Manhattan Manhattan, especially for people who prefer the cleaned-up simulacrum to the authentic experience. Why endure Manhattan's tiny rooms and indifference when Nashville, Austin or suburban Atlanta offers rooftop cocktails, ambitious restaurants, live music, luxury apartments and valet parking? Why endure Hollywood's seediness when a university media center supplies professional studios inside a landscaped residential resort? Why tolerate Las Vegas resort fees, casino distances and predatory prices when nearly every city now offers sports betting, celebrity chefs, arena concerts, rooftop pools and bottle service?

Even London has become Londonland, a premium experience constructed inside the physical remains of London. The red bus, black cab, palace guard, market and Victorian brickwork serve as assurances that the visitor purchased the correct London. Working markets become curated food destinations. Industrial buildings become creative quarters. Pubs become immaculate performances of pubs. The city becomes denser in monetizable experience and thinner in ordinary life. London has not disappeared. It has been preserved, intensified, branded and sold back to the world as an immersive London experience.

Every city has effectively become a red-light district standing beneath the global algorithm and displaying whatever portion of itself might persuade a portable affluent person to come inside and spend recreational money. There is a food district, arts district, waterfront district, innovation district, historic district, nightlife district and sanitized dangerous district where visitors can purchase grit without accepting much danger. Even authenticity becomes an outfit. The rusted warehouse receives Edison bulbs, cocktails and a mural explaining that it remains a rusted warehouse. The poor neighborhood becomes an emerging cultural corridor. The fishing village installs high-speed Wi-Fi and accepts reservations through an app.

This is not the destruction of authenticity by imitation. It is the discovery that many customers prefer the imitation once it has cleaner bathrooms, better parking, stronger Wi-Fi and fewer opportunities for the authentic place to say no. Paris can still expect you to accommodate Paris. Berlin can still demand that you learn Berlin. Places retaining destination sovereignty can remain difficult because difficulty has been incorporated into the product. Everyone else must welcome, stimulate, photograph, entertain and reassure.

The human attraction

The competition does not end with places because people now inhabit the same global comparison market. It was once sufficient for a relatively healthy man and woman in the same town to encounter one another and establish a life. Now a man is compared not merely with other men nearby but with Wolverine, professional athletes, actors, fitness influencers and a global feed of chemically and digitally optimized bodies. Superman is no longer muscular enough. If a man is not shaped like Wolverine, he has a dad bod. Women have been returned to the thin aesthetic of 2000 while also being expected to display the proportions and physical optimization of a comic-book heroine. The ordinary human body has become another under-amenitized regional institution losing customers to a destination resort.

The mechanism is identical. Every person appears to possess infinite outside options, while every potential partner knows that he or she is being assessed against the most compelling people an algorithm can locate. Bodies, homes, careers, vacations and personalities must therefore present an experience. Merely being decent, solvent, affectionate and available begins to feel like the cinder-block dormitory of romance. The human being McMansions himself because the plain person fears the same fate as the plain university and the plain seaside town.

There is no reason to pretend I approach this with the serenity of a central banker. I lived through the dot-com bomb and watched companies spend fortunes acquiring customers they would never profitably serve. I lived through the 2008 collapse and watched paper abundance discover gravity. I lived in England when prosperity still encountered routine public hostility, and I now live in a part of Virginia where objects that would once have looked like royal property sit uncovered beside the curb. I recognize the signs of a land grab. At the beginning of a disruption, competitors spend far more than the customers will ever repay because survival appears to depend on achieving dominance before the market closes.

That is what American cities and universities are doing. They are building for the next ten years because they suspect the next ten years will determine who remains in the consideration set. Some will win. Some improvements will be useful, beautiful and humane. Some cities genuinely will become better places to live. None of that changes the financial structure. An attractor must keep attracting, and the customer it wins today retains the right to leave tomorrow.

The affluent class is large enough to sustain this competition far longer than skeptics expect. This is not the dot-com economy's narrow population of venture capitalists, founders and early adopters. It is tens of millions of households whose retirement accounts, houses, professional incomes and expectations continually refill the tank. The revolutionaries may bring a knife to the gunfight, but their pension funds own shares in the gun manufacturer. The same people who denounce capitalism in theory defend their mortgage rates, school districts, university credentials, home equity and children's inheritance in practice. They will tolerate almost any critique of the system provided the proposed demolition begins several houses farther up the street.

This is why the American structure bends, inflates and redistributes failure without producing the satisfying single collapse everyone keeps predicting. Weak universities close individually. Overextended developments fail locally. Affluent demand sorts upward or migrates elsewhere. One city cools while another becomes the new Austin. One resort declines while another appears in an airport's direct-flight map. The baffles interrupt the wave.

America did not become a theme park because Americans lost their taste for reality. It became a field of competing attractions because the people with the most money no longer had to be anywhere. Their mobility converted institutions from authorities into supplicants. The student became the guest. The resident became the recruit. The spectator became the consumer. The city became the experience. Every school, stadium, apartment, restaurant, vacation town and human body began competing to become sufficiently spectacular that the affluent customer would not choose somewhere else.

The old city said, "You must come here." The old university said, "Meet our standards." The old seaside town said, "This is where your family goes." The new institution can say none of those things with confidence. It stands beneath the algorithm with its lights on, its waterfront redeveloped, its dormitories renovated, its body optimized and its authentic warehouse carefully preserved. It asks what would persuade you to stay.

It is a small world after all. That is precisely the problem.

Sources and notes

Household-income estimates come from the U.S. Census Bureau's 2024 household-income tables. International disposable-income comparisons come from the OECD. The frequently disputed Britain-Mississippi comparison is explained in The Atlantic; its precise ranking moves with the year, measure and exchange rate. Work-from-home estimates come from the Survey of Working Arrangements and Attitudes series published by FRED. The Texas facility examples come from Allen Independent School District and the Melissa Championship Center project account. Contemporary World Cup stadium reactions were reported by Stadia Magazine.

There is a point in a bad marriage when the fighting stops, and that is usually the point when the spouse who regarded the fighting as evidence of commitment should become worried. For years, every argument could be interpreted as participation in the relationship. However ugly the evening became, both people were still negotiating because both still believed there was something worth negotiating over. Then one of them quietly rents an apartment. The other is astonished. Why leave now, when we were still working things out? The answer is that the fighting was the working-it-out part. Silence arrived when the decision had already been made.

American institutions have spent decades making a similar mistake. Schools became accustomed to angry parents, universities to angry alumni, newspapers to angry readers, corporations to angry customers, cities to angry taxpayers, television networks to angry viewers, and political parties to angry voters. The anger could be managed precisely because the angry people remained customers of the institution. A parent screaming at a school-board meeting still had children enrolled in the schools. A conservative furious with network television still watched network television. A businessman complaining about state government still operated his business in the state. A homeowner furious about property taxes still owned property inside the jurisdiction collecting them. The complaints sounded like rebellion while functioning as evidence of attachment.

Increasingly, Americans are discovering a different response. They can go their own way.

The phrase already belongs to men who concluded that contemporary relationships offered them an unacceptable bargain. Whatever became of the sprawling MGTOW subculture, its central mechanism was simple and powerful: withdrawal. A man dissatisfied with marriage or dating could spend his life trying to change women, society, divorce law, sexual politics and expectations surrounding masculinity, or he could decide that he no longer wanted the product badly enough to accept its terms. Participation had been mistaken for an obligation when it was actually a choice. Once he understood that, the negotiation changed completely.

That is the useful principle here, stripped of the sexual politics surrounding the original acronym. Going Their Own Way is what happens when people stop asking an institution to change and begin arranging their lives so that they need less from it. Parents homeschool or choose another school. Workers relocate. Audiences abandon networks. Writers publish themselves. Businesses change states. Families cross school-district lines. People move from cities to suburbs, from suburbs to exurbs, from incorporated towns to unincorporated land, or from one state to another. Each decision can be tiny when viewed from the household making it. The importance lies in what happens when thousands of tiny decisions begin pointing in the same direction.

Fuck This Hill. There Are Other Hills.

Much of the world contains peoples whose relationship with land is nearly impossible for a highly mobile American to understand. Land can contain ancestry, religion, graves, language, memory and obligations stretching through generations. The particular ground matters. A replacement parcel fifty miles away cannot replace it because the attachment concerns the place itself rather than its acreage, price or amenities.

The stereotypical white American relationship with geography is almost comically different. When taxes become intolerable, schools deteriorate, crime increases, regulations multiply, neighborhoods change, employment disappears or the general aggravation level exceeds the owner’s tolerance, the moving truck appears. There is another suburb, another county, another state, another subdivision, another five acres. Americans have been doing this since before there was an America. Cross the ocean. Cross the mountains. Leave the farm. Leave the city. Leave the old industrial town. Leave the inner suburb. Leave the expensive coastal state. Home becomes partly a place and partly a kit that can be reassembled wherever the truck stops.

That habit matters because mobility creates a peculiar form of political power. Institutions normally assume that dissatisfied people will remain inside their boundaries. The school district assumes parents will keep sending children. The municipality assumes homeowners will continue owning houses. The state assumes earners and businesses will remain taxable. The newspaper assumes readers still want the newspaper. The political argument therefore concerns what the institution will do for you, permit you to do, charge you for doing, or require you to tolerate.

Going Their Own Way changes the pronouns.

The school district can tell you what your schools will teach until you decide they are no longer your schools. The city council can tell you what your city has decided until you no longer live in the city. The newspaper can explain what our community believes until you stop reading it. The state can describe what we have chosen until you become a resident of another state. None of those institutions disappears when you leave. What disappears is the institution’s ability to define itself as yours merely because it once was.

That distinction becomes particularly important at the local level because local government is where belonging and payment are most tightly connected. Leaving a school district can mean removing both a child and a family from its future enrollment. Leaving a municipality can mean taking a household’s property, consumption and economic activity beyond the municipal tax base. Leaving a state can move income, investment and future taxable activity with the person. Moving into a different jurisdiction can redirect property taxes toward a different school system and local government. The person making the move usually experiences none of this as a grand political act. He has found a house, a school, a tax bill and a community he prefers. From his perspective, the sentence is simply: I live here now.

That is where the cascade begins.

White Flight and the Politics of Exit

White flight is the enormous American demonstration of how consequential millions of individually comprehensible departures can become. Families leaving cities took purchasing power, tax revenue, businesses, children, professional skills, political energy and future investment with them. The racial discrimination, housing policy, redlining, exclusionary development and segregation surrounding that history mattered enormously and cannot be separated from its consequences. But white flight also demonstrates something broader about exit itself: a person does not need to seize control of an institution to profoundly alter it. He can leave it.

This is where a seemingly small event becomes interesting. One family leaves a school district and essentially nothing happens. One affluent household moves outside the city limits and the city continues operating. One business relocates and another may take its place. Even a hundred departures may disappear inside the normal churn of a metropolitan area. Institutions are large enough to absorb individual exits, which makes each departure easy to dismiss.

The secondary effects arrive through accumulation. Fewer children can eventually mean lower enrollment. Lower enrollment can change staffing, programming, school utilization and political constituencies. A changing residential population changes the property-tax base, retail demand, housing market and electorate. Businesses follow customers and workers. Families compare schools partly by the families already attending them. Employers compare jurisdictions partly by the workers already living there. Homebuyers compare neighborhoods partly by the tax burden and services created by everyone who arrived before them. The departure that looked like an isolated household decision becomes one input into the next household’s decision.

Then the direction of causality becomes difficult to untangle. Did families leave because the schools declined, or did the schools struggle because families with money, children, political influence and alternatives left? Did the tax base weaken because businesses departed, or did businesses depart because the tax base was weakening and government was asking more from those who remained? Did a suburb become desirable because affluent families moved there, or did affluent families move there because it was desirable? At sufficient scale the answer can become all of the above. Cascades feed themselves.

That is the part of Going Their Own Way that gets lost when every departure is interpreted as an ideological statement. Most people do not move with a manifesto taped to the windshield. They make decisions at human scale. This house is cheaper. That school seems better. The commute works because I am remote three days a week. My parents live nearby. Property taxes are lower. I can have a yard. I am tired of dealing with this city. None of those sentences sounds revolutionary. Add enough of them together and they can redraw a metropolitan region.

COVID Made the School Optional

Public education provides one of the clearest contemporary examples because COVID accidentally conducted an enormous audit of American classrooms. Parents who previously experienced school through homework, report cards, parent-teacher conferences and stories told over dinner suddenly sat within earshot of the school day. They heard teachers speak, watched lessons unfold, encountered the vocabulary of the classroom, observed disciplinary norms and discovered which subjects received confidence, attention and moral emphasis. Some came away impressed. Others became convinced that academic priorities, cultural priorities or both had drifted far from what they expected from public education.

The lasting change was larger than any argument over curriculum. Millions of parents learned that the school itself was optional in ways they had never seriously considered. Homeschooling had infrastructure. Online curricula existed. Microschools could be organized. Religious and private schools were alternatives. Charter schools, tutors, co-ops and learning pods expanded the menu. Even families remaining entirely committed to public education discovered another form of exit: move six miles and acquire a different public school system.

That last form may be more consequential than the spectacular political battles because it connects education directly to geography and taxation. A family that removes a child for homeschooling can still own the same house and pay the same property taxes. A family that leaves the school district changes considerably more. The children leave. The parents leave. Their house changes hands. Their spending patterns change. Their votes disappear from one jurisdiction and appear in another. Depending on the move, their property taxes may begin supporting another school system, another county, another municipality. What began as dissatisfaction with something happening in a classroom can end with a moving truck crossing a jurisdictional boundary.

The institution sees the enrollment number. The municipality sees the tax roll. The realtor sees demand moving between neighborhoods. The receiving school sees another student. The receiving county sees another household. The family sees Tuesday.

That mismatch of scale is why cascades are so easy to miss while they are happening.

Going Their Own Way Is Bigger Than Geography

Media went through the same transition without moving trucks. Conservatives spent decades complaining about newspapers, network television, Hollywood, publishing and eventually social-media companies. Complaining implicitly acknowledged the authority of those institutions. Someone furious that a newspaper would never publish his argument still wanted publication in that newspaper. Someone angry about television coverage still regarded television coverage as important. Technology gradually reduced the value of permission. Talk radio, cable television, blogs, YouTube, podcasts, streaming and independent publishing allowed audiences and creators to construct alternatives.

The important event was not that everyone suddenly agreed. They stopped needing to occupy the same room. The old institution could retain its values, standards and audience while the people who disliked the bargain assembled somewhere else. The political battle over who controlled the gate weakened because there were suddenly many gates.

Money behaves similarly because money possesses even less sentimental attachment to institutions than people do. Companies relocate. Investors reallocate. Retirees change states. Entrepreneurs incorporate elsewhere. Remote work allows some earners to separate employment from geography altogether. A government still possesses enormous authority within its jurisdiction, but jurisdiction itself becomes part of the bargain whenever the governed possess a practical exit.

The interesting question is therefore rarely whether one particular tax, school policy, zoning decision, cultural argument, crime statistic or regulation will cause some dramatic exodus. Usually it will not. People tolerate enormous amounts of friction because moving is expensive and disruptive. The cascade begins when several ordinary incentives align. A household dislikes the schools a little, taxes a little, housing costs a little, regulations a little, crime a little and the general direction of the place a little. Meanwhile another jurisdiction offers a somewhat better school, somewhat cheaper house, somewhat lower taxes and people the family already knows. Eventually the arithmetic crosses an invisible line. Nobody needs to become furious. The alternative merely needs to become easier than staying.

The Receiving Side of the Cascade

Departure changes the destination too. A family carries consumption, labor, children, taxes and social networks. A business carries payroll, suppliers, investment and future expansion. Enough arrivals create demand for houses, schools, restaurants, churches, stores, professional services and political representation. The place people escaped to begins acquiring institutions capable of supporting more people like them. That makes it more attractive to the next wave.

This is how sorting becomes self-reinforcing. People select different packages of order, autonomy, surveillance, community, privacy, services, taxes and inconvenience. Some willingly pay high taxes for excellent schools and extensive public services. Some choose a tightly regulated municipality because they value predictability. Others accept fewer services in exchange for lower taxes or greater autonomy. Some want an HOA. Others would rather look at their neighbor’s rusting washing machine than receive a letter about the height of their own grass. America’s abundance of overlapping jurisdictions allows remarkably different preferences to exist within a short drive of one another.

The important word is their. Going Their Own Way does not require everyone to agree about which package is best. It depends upon people being able to select among packages. My preferred bargain can be intolerable to you. Your preferred bargain can seem absurd to me. Political temperature falls considerably when neither of us requires the other to live under precisely the same arrangement.

This also explains why the cascade can become dangerous to institutions without being threatening in any ordinary sense of the word. Nobody has to menace a school board, city council, university, newspaper or employer. Nobody needs to announce retaliation. People can simply stop showing up. The school can keep its curriculum. The newspaper can keep its editorial position. The city can keep its policies. The state can keep its tax structure. The employer can keep its rules. The individual retains the equally mundane decision to determine whether the resulting bargain is worth buying.

Exit Produces Sorting, and Sorting Produces Institutions

Going Their Own Way becomes more consequential as repeated exits produce durable alternatives. A few homeschooling families constitute a departure. A large homeschooling ecosystem becomes an educational institution. A podcaster with a microphone represents independent media. Thousands of creators with millions of listeners constitute a media system. Families leaving one state represent migration. Enough families carrying businesses, wealth and political preferences reshape the state receiving them.

Exit therefore has stages. Dissatisfied people complain while they remain invested in an institution. Some eventually leave. Departures sort people into alternatives that better match their preferences. Successful alternatives develop infrastructure. Infrastructure attracts investment and talent. That infrastructure makes departure easier for the next person. What required courage or eccentricity for the first generation becomes a menu option for the second. Children grow up inside the alternatives and regard them as ordinary.

That final step may be the most important. The first generation remembers the old newspaper, university, city, neighborhood or school system. It may retain nostalgia and periodically wonder whether returning is possible. Its children may have no such attachment. The homeschooled child does not necessarily experience homeschooling as an escape from public education. It was simply school. The family raised in the new county does not necessarily think of itself as having fled the old city. That was where Mom and Dad used to live.

The cascade has then crossed from behavior into inheritance.

This is why institutions should pay attention to quiet departures long before they become dramatic demographic trends. Anger represents engagement. Complaint represents attachment. A furious parent still believes the school can be recovered. A furious reader still cares about the newspaper. A furious resident still believes the city is his city. The argument itself contains evidence that the relationship remains valuable enough to fight over.

Silence arrives later.

By then the family has another school. The reader has another publication. The business has another state. The homeowner has another county. The taxpayer has another jurisdiction. The person who once argued about what we should do has quietly changed the meaning of we.

That is Going Their Own Way. It is not necessarily a movement, and it does not require coordination. It can emerge from millions of people making decisions that seem almost boring when considered separately. One enrollment withdrawn. One house sold. One subscription cancelled. One business incorporated elsewhere. One family crossing a county line. One taxpayer deciding that the combination of services, schools, regulations, costs and culture available somewhere else suits him better.

The cascade appears only afterward, when somebody looks at the enrollment, the tax base, the audience, the population or the map and asks where everybody went.

The answer may be that they stopped fighting.

The fighting meant they hadn’t left yet.

There is a piece of pedestrian wisdom so obvious that nobody thinks it requires a political philosophy: having the right of way does not make it safe to step in front of a moving truck. You may be standing at a marked crosswalk with the WALK signal illuminated. The driver may have a red light. The traffic code may be completely unambiguous about which of you is entitled to proceed. If the driver nevertheless fails to stop, none of those facts will alter the mass of the truck, shorten its stopping distance, or make your femur more resistant to fracture. You can be legally right, morally right, factually right and catastrophically injured at exactly the same time. Looking both ways before crossing does not surrender your right of way, excuse the driver, endorse reckless driving or transfer responsibility for the collision to you. It means only that you understand the difference between a rule governing another person's conduct and a physical mechanism capable of preventing that person from violating the rule.

We understand this distinction effortlessly when trucks are involved, yet we have developed a strange tendency to forget it when the subject becomes constitutional rights, policing, protest, institutional responsibility and personal safety. Somewhere along the way, the sentence “they are not allowed to do that” begins quietly transforming into “they cannot do that.” The first statement is about law. The second sounds as though it is about physics. Those are radically different propositions, and confusing them can become dangerous long before a judge gets an opportunity to straighten things out.

The law is not a force field.

I mean that almost literally. The Constitution does not surround an American citizen with a transparent membrane through which government power cannot pass. The First Amendment does not physically prevent a police officer from arresting somebody for protected speech. The Fourth Amendment does not make an unreasonable search mechanically impossible. A statute prohibiting discrimination does not prevent an employer from discriminating. A contract does not prevent somebody from breaching it. A restraining order does not erect an invisible wall around the person it protects. What law does, among many other things, is establish rules, duties, prohibitions, procedures and remedies so that conduct can be restrained prospectively when possible and judged afterward when prevention fails. The existence of a legal remedy is itself an acknowledgment that people sometimes do things they are forbidden to do.

Federal civil-rights law makes this particularly plain. Section 1983 of Title 42 provides a civil cause of action against people who, acting under color of state law, deprive someone of rights secured by the Constitution and federal law. The statute is tremendously important, but notice the chronology embedded in the concept: a person has been deprived of a right and seeks a remedy for that deprivation. The courthouse is enormously powerful, but the courthouse is downstream from the event.

That temporal distinction is the whole subject of this essay. It is also why I want to dispose immediately of an especially dangerous variation of the force-field fantasy: the idea that legal remedies themselves can make physical risk worth taking. I am not claiming that most protesters think this way, nor am I attributing the calculation to any particular person. But anyone who does begin calculating that an unlawful arrest, an unjustified shove, an excessive-force incident or some other well-documented civil-rights violation could produce a handsome settlement is playing an extraordinarily peculiar lottery. There is no control on the encounter that allows you to select “injured enough to collect $2 million, but healthy enough to enjoy it.” Human bodies and chaotic confrontations do not have settlement-value settings.

A shove can become a fall. A fall can become a traumatic brain injury. A supposedly less-lethal projectile can destroy an eye. A vehicle can move six feet when someone expected it to move two. A hand can disappear from view. An officer can misunderstand a movement. A frightened person can make an impulsive decision. A person who expected to spend a night in jail and become the sympathetic plaintiff in an excellent civil-rights case can instead spend months learning to walk again. In the worst version of this grotesque lottery, the plaintiff never receives the award at all. The estate does. The surviving family members may eventually receive money from the government while wishing more than anything that there had never been a cause of action in the first place.

A large civil judgment should therefore never be confused with evidence that the law protected somebody from injury. Quite often it means exactly the opposite. Damages are an attempt to compensate for something the legal prohibition failed to prevent. Money can pay medical bills, replace lost earnings, punish certain forms of misconduct and acknowledge an injustice. It cannot restore the afternoon before the injury occurred. A legal system can assign a monetary value to blindness. It cannot give somebody the eye back.

This is not an argument for meekness, obedience or allowing government officials to do whatever they please. It is certainly not an argument that excessive force becomes justified when the person receiving it has behaved foolishly or offensively. It is an argument for maintaining separate ledgers that political arguments constantly try to collapse into one. There is the question of legal authority: what was each person legally entitled to do? There is the question of responsibility: who caused the situation, escalated it, behaved recklessly or violated another person's rights? And there is the question of risk: regardless of who is right, what is physically capable of happening during the next ten seconds? A person can answer the first two questions perfectly and still make a disastrous mistake about the third.

Why Are You Using Real Bullets?

The death of Renée Good in Minneapolis on January 7, 2026, is worth discussing here precisely because I do not need to settle the enormous political and legal argument surrounding her death in order to make this point. Good was fatally shot during an encounter with federal immigration agents. Federal authorities and critics of the shooting offered sharply different interpretations of what happened and whether the use of deadly force was justified. The Minnesota Bureau of Criminal Apprehension initially expected to participate in a joint investigation with the FBI but said the arrangement changed later that day, leaving the FBI in charge and the state agency without the access to evidence and interviews it said it needed for a complete investigation. The BCA subsequently said it remained willing to participate if federal authorities resumed cooperation.

Those disputes matter enormously. They matter to Good's family, to the agent, to the government and to anybody who cares about the lawful use of state power. They are also unnecessary to my argument. If the shooting is ultimately judged legally justified, the law was not a force field. If it is ultimately judged unjustified, the law was still not a force field. The legal conclusion changes responsibility. It cannot change what happened on January 7.

In video recorded immediately after the shooting, Good's wife, Becca, can be heard asking, “Why did you have real bullets?” I want to be exceedingly careful with that sentence. A woman who has just watched her wife be shot is not delivering a seminar on constitutional theory. Shock is not testimony about what somebody rationally believed before a traumatic event, and it would be indecent to diagnose Becca Good's politics, intelligence or state of mind from words spoken in that condition. I am interested in the sentence for another reason. Taken entirely on its own terms, it captures with awful economy the difference between the institutional world we imagine and the physical world we inhabit.

The bullets were real because the guns were real. The agents were not actors portraying federal agents in a civic pageant. Their weapons were not theatrical representations of state authority. They were people carrying equipment capable of injuring and killing other human beings, operating in an environment in which decisions could become irreversible almost immediately. Whatever restrictions the Constitution, federal law, departmental policy and the law of deadly force placed upon their use of those weapons, none of those restrictions altered what the weapons physically were.

That sounds almost insultingly obvious when written down, which is exactly why the sentence bothers me. There are circumstances in modern American life in which people seem to approach armed state power with an extraordinary confidence that the rules governing that power also guarantee its restraint. A person can scream at an armed officer, insult him, call him a racist, a fascist, a coward or anything else the First Amendment may protect in the circumstances, and there may be an extremely strong legal argument that none of those words permits the officer to strike or shoot him. Fine. That still leaves the entirely separate question of whether deliberately creating an emotionally charged, face-to-face confrontation with an armed stranger is a sensible way to test the proposition.

The First Amendment can be on your side while the person standing six inches away is angry, frightened, confused, poorly trained, exhausted, mistaken about your intentions or simply bad at his job. The right remains a right. Your possession of it does not improve his judgment.

This is where accusations of “victim blaming” tend to enter the discussion, and I think the accusation often prevents adults from discussing risk intelligently. Responsibility and prudence are not mutually exclusive concepts. If I tell my friend to look both ways before crossing the street, I am not announcing a new legal doctrine under which motorists may mow down inattentive pedestrians. If I tell someone not to leave an expensive laptop visible on the passenger seat of an unlocked car, I have not legalized theft. If I tell someone that getting inches from an armed officer and deliberately trying to enrage him increases the possibility of an ugly encounter, I have not granted the officer any additional authority whatsoever. The officer's legal obligations remain exactly where they were.

This distinction used to be called prudence.

Rights and Remedies Are Different Things

American constitutional law itself contains some bracing reminders that government is not a universal guarantor of personal safety. In DeShaney v. Winnebago County Department of Social Services, the Supreme Court confronted an appalling case involving a child who suffered catastrophic abuse by his father despite extensive involvement by social services. The Court held that the Due Process Clause generally does not impose an affirmative constitutional obligation on the state to protect an individual from private violence. The constitutional protection against deprivation of liberty by the state was not, in the Court's understanding, a general guarantee that the state would successfully protect a person from other people. The distinction is morally uncomfortable precisely because the underlying facts were horrific, but it demonstrates how far the actual constitutional structure can be from the popular notion that possessing rights means government has undertaken to keep each of us safe.

Town of Castle Rock v. Gonzales is perhaps even more brutal as an illustration. Jessica Gonzales had obtained a restraining order against her estranged husband. He abducted their three daughters, and after repeated attempts by Gonzales to get police to enforce the order, he eventually arrived at a police station and opened fire. Police killed him and discovered the bodies of the three girls in his truck. The Supreme Court ultimately rejected Gonzales's procedural due-process claim that she possessed a constitutionally protected property interest in police enforcement of the restraining order. The precise legal issue was technical and the decision remains controversial, but the human lesson is much simpler: an order bearing the authority of a court is still an order. It is not a wall.

That does not make restraining orders useless. It explains why they are useful. They establish legal boundaries and enable enforcement. They create consequences. They provide evidence and authority for intervention. In many cases those mechanisms undoubtedly deter conduct that otherwise would occur. But deterrence is probabilistic. Paper cannot tackle a person approaching your front door. The order becomes physically effective only through the behavior of human beings who know about it, respect it or enforce it.

We understand the same distinction perfectly well in ordinary private law. A contract does not prevent breach. If contracts prevented breaches, contract litigation would barely exist. A lease does not prevent a landlord from doing something prohibited by the lease. Employment law does not physically prevent unlawful discrimination. Criminal statutes do not make murder impossible. Indeed, every criminal prosecution demonstrates the obvious point that declaring conduct illegal and making conduct physically impossible are not remotely the same enterprise.

Yet the language of rights encourages a peculiar psychological slippage because rights are properly described in categorical terms. You have the right. The government may not violate it. The officer cannot lawfully do this. All of that is correct, but “cannot lawfully” can lose its final word in ordinary conversation until “cannot” begins carrying more weight than it deserves. An officer cannot lawfully use objectively unreasonable force under the Fourth Amendment. An officer obviously remains physically capable of doing so, which is why excessive-force litigation exists in the first place. Federal prosecutors themselves describe use-of-force analysis as requiring evaluation of whether force was objectively unreasonable while recognizing that officers can be forced to make judgments in circumstances that are tense, uncertain and rapidly evolving.

The legal system then performs one of civilization's most important tricks: it slows everything down. Video can be viewed frame by frame. Witnesses can be interviewed separately. Forensic evidence can be collected. Lawyers can research precedent. Experts can reconstruct trajectories. Judges can hear motions. Jurors can deliberate. Appellate courts can review what the trial court did. Questions that had to be confronted amid noise, movement and adrenaline can eventually receive hundreds of pages of briefs and opinions.

This is wonderful. It is also exactly the opposite of a force field.

The person standing in the street does not have appellate time.

The Sanitization of Civil Disobedience

Some of our confusion may come from the way successful protest movements are remembered. History compresses. Decades become chapters; chapters become paragraphs; paragraphs become photographs. Eventually we remember Rosa Parks sitting peacefully on a bus, Martin Luther King Jr. speaking from the steps of the Lincoln Memorial, marchers crossing the Edmund Pettus Bridge, and then Congress passing landmark civil-rights legislation. The moral arc becomes so smooth in retrospect that it is easy to forget what the people inside the events actually experienced.

The people who practiced nonviolent civil disobedience did not possess a theory that nonviolence made them physically safe. Quite the contrary. The power of the method depended partly on their willingness to accept danger without answering violence with violence. Civil-rights demonstrators were arrested, beaten, attacked by dogs, knocked down by high-pressure fire hoses and murdered. Freedom Riders did not board buses because federal law created an invisible protective bubble around interstate travelers. They boarded knowing that the law and the physical reality on the ground could be separated by an enormous and dangerous distance.

That distinction matters because modern protest can sometimes inherit the iconography of civil disobedience without inheriting its understanding of danger. Nonviolence was not a guarantee that the other side would also be nonviolent. Moral legitimacy was not expected to render clubs soft. A constitutional right to assemble did not make an angry crowd, hostile police department or violent opponent physically incapable of hurting anybody.

This is one reason the question “Why are you using real bullets?” is so haunting. There was never a historical period in which the coercive apparatus of the state consisted only of symbols. Government ultimately possesses the capacity to compel behavior through physical force. Democratic government surrounds that power with law, training, accountability, elections, judicial review, civil liability, internal discipline and constitutional restrictions because coercive power is dangerous. The restrictions are not evidence that the danger has disappeared. They exist because it has not.

Institutions Care Differently Than People Do

There is a related mistake in the way we talk about institutional care. Universities now routinely speak the language of safety, wellness, belonging, mental health and community. Employers do it. Government agencies do it. Schools do it. Large organizations increasingly sound less like contractual institutions performing defined functions and more like unusually well-funded parents.

Individual people inside those organizations may care about you enormously. A professor can genuinely worry about a student. A supervisor can care deeply about an employee. A police officer can risk his life for a stranger. An administrator can lose sleep over someone in crisis. Nothing in this argument requires the bleak assumption that everybody inside an institution is a sociopath reading from a liability manual.

The institution itself, however, does not love you. It has policies, incentives, legal duties, insurance, public-relations concerns, budgets, regulatory obligations, lawyers, risk managers and institutional interests. Sometimes those interests align beautifully with yours. Sometimes they do not. The fact that an institution maintains an office devoted to your safety does not mean it has accepted responsibility for every danger you encounter, much less acquired the ability to eliminate it.

This is where I part company with a certain progressive conception of safety that seems increasingly willing to outsource personal risk assessment to institutions. If an institution says an environment is safe, if the activity is constitutionally protected, if trained authorities have rules governing their conduct, if there are cameras everywhere and lawyers available afterward, there can emerge a confidence that somebody else has already performed the dangerous part of the calculation.

But nobody can outsource physics.

This is not uniquely a progressive vice. Conservatives perform their own version whenever they assume that police officers, soldiers, churches, corporations or political leaders are trustworthy because those institutions belong to their cultural team. The conservative who says, “I haven't done anything wrong, so why would I need a lawyer?” is wearing his own constitutional force field. So is the businessman who says a counterparty “can't” do something because the contract prohibits it. So is the gun owner who imagines that possession of a weapon has eliminated rather than altered his risk. Human beings are extraordinarily talented at turning systems they trust into imaginary guarantees.

The progressive version interests me because contemporary protest culture can combine very expansive ideas about institutional responsibility for personal safety with extremely confrontational behavior toward the armed representatives of those same institutions. There is something difficult to reconcile about believing that a university bears substantial responsibility for protecting one's psychological safety while simultaneously assuming that a federal agent carrying a sidearm and a rifle will maintain perfect emotional regulation while being screamed at from inches away. The agent should maintain that regulation. His training, authority and weaponry give him a greater obligation to do so, not a smaller one. But “should” is precisely the word around which this entire essay revolves.

The truck should stop.

Look anyway.

Prudence Is Not Cowardice

There is a temptation to treat personal risk management as capitulation, particularly in political movements built around resistance. If I alter my behavior because another person might behave unlawfully, haven't I allowed intimidation to work? If I refrain from saying something because a police officer might respond improperly, haven't I surrendered my First Amendment rights? If everybody behaved that way, wouldn't government misconduct become self-enforcing?

Sometimes, yes. There are circumstances in which exercising a right despite danger is morally necessary. Nearly every political freedom we possess was defended at some point by people willing to accept risks that a purely actuarial analysis would have told them to avoid. I am profoundly grateful that such people exist. A society consisting entirely of people who retreat whenever power threatens them would not remain free for very long.

But accepting a risk knowingly is categorically different from denying that the risk exists. Courage requires the first. Delusion requires the second.

A Freedom Rider who knew a bus might be attacked and boarded anyway made a moral choice about the relative value of personal safety and political action. Someone who boards because “they can't attack us, that's illegal” has misunderstood both the adversary and the nature of law. Two people can perform the same physical act while operating from completely different understandings of reality.

That is the distinction I want restored. I am not arguing that nobody should confront power, protest, engage in civil disobedience or deliberately risk injury for a cause. I am arguing that anyone doing those things should understand that constitutional protection determines legal rights, not ballistic outcomes. If you decide that a principle is worth risking your body for, at least know that this is the transaction you are making. Do not convert courage into magical thinking by assuming that the existence of a right has eliminated the price you have decided you are willing to pay.

The same applies at a much smaller scale. You can assert every right available to you during an encounter with police while still keeping your hands visible, maintaining distance, avoiding sudden movements and deciding that the sidewalk is not the ideal venue in which to win a constitutional argument. Compliance in the moment does not necessarily mean agreement. A courtroom exists precisely because there are disputes that are safer to conduct with lawyers than with adrenaline.

There are limits to that advice, obviously. Authorities can abuse demands for compliance. Governments throughout history have described dissent itself as disorder. Police commands can be unlawful. People who obediently comply can still be injured or killed. Nothing about prudent behavior guarantees safety either. That would simply replace one imaginary force field with another.

Risk management is not omnipotence. It is probability.

The Awful Meaning of “Eventually”

One of the most seductive phrases in American civic life is “the courts will sort it out.” Often they do, and thank God for that. But consider how much human experience is concealed inside the future tense.

Eventually the conviction may be reversed. Eventually the search may be declared unconstitutional. Eventually the officer may be disciplined. Eventually the government may settle. Eventually the employee may win the discrimination case. Eventually the landlord may be ordered to pay damages. Eventually the appellate court may publish an opinion establishing that what happened was unlawful all along.

Before eventually comes everything else.

That is not a criticism of courts. Careful adjudication takes time because careful adjudication should take time. Evidence needs to be collected. Both sides deserve to be heard. Facts are disputed. Law is complicated. Governments and defendants have rights too. The alternative to retrospective deliberation is often immediate judgment without adequate information, which creates dangers of its own.

But the unavoidable consequence is that law operates on a different clock from physical reality. An event may become irreversible in a fraction of a second while the legal system requires years to decide exactly what that fraction of a second meant. Qualified-immunity doctrine adds another layer in civil litigation involving government officials, asking not merely whether a right was violated in applicable cases but whether the right was sufficiently clearly established under existing law. The doctrine is explicitly intended to balance accountability for irresponsible exercises of power against protecting officials performing their duties reasonably. Whatever one thinks of where courts have struck that balance, its very existence demonstrates how much legal analysis can stand between an injury and a remedy.

There is therefore nothing inconsistent about believing fiercely in constitutional rights while behaving cautiously around people capable of violating them. In fact, I think the two positions belong together. I want government power constrained precisely because I understand that government power is real. I want courts available precisely because officers, prosecutors, bureaucrats, presidents and legislatures can be wrong. I want damages available precisely because injuries cannot always be prevented. I want restraining orders precisely because dangerous people sometimes need to be restrained. The institutions of law are necessary because human beings are fallible, emotional, self-interested, frightened, tribal, occasionally malicious and frequently mistaken.

Treating those institutions as magical protection misunderstands the reason we built them.

Look Both Ways

So I return to the crosswalk because the crosswalk contains almost the entire argument without requiring a law degree.

You approach the curb. The signal says WALK. The approaching truck has a red light. The driver is legally obligated to stop. You are legally entitled to proceed. If he runs the light and strikes you, the fact that you looked both ways would not make him less responsible. The fact that you failed to look would not magically give him a green light. The legal questions remain legal questions.

But before stepping off the curb, you look at the truck.

You do not do this because you distrust the concept of traffic law. You do it because you understand why traffic law exists. Trucks are heavy. Drivers make mistakes. Brakes fail. People look at phones. Human beings misunderstand signals. Occasionally somebody deliberately runs a red light. The law can deter those things, punish those things and compensate some of their consequences. It cannot guarantee that they will not happen to you this afternoon.

That habit of mind should travel with us beyond the intersection. Your constitutional rights are real. Your civil rights are real. Your contractual rights are real. Your right to protest is real. The duties imposed upon police officers are real. The restrictions placed upon government are real. Court orders are real. Civil remedies are real. None of them should be surrendered merely because someone capable of violating them happens to be standing nearby.

But the nearby person is real too.

His anger is real. His fear is real. His mistakes are real. The pavement is real. The truck is real. The gun is real.

And the bullets are real.

The law can tell us afterward who was wrong. It can sometimes compensate the person who was wronged. It can punish misconduct, establish precedent, change policy and make the next violation less likely. At its best, it can take an injury suffered by one human being and turn it into a rule that protects millions of others.

That is an astonishing power.

It just isn't a force field.

The Khmer Rouge governed Cambodia for three years, eight months, and twenty days. That was enough time for nearly two million people to die, at least a quarter of the country's population according to the United States Holocaust Memorial Museum, while the regime attempted to transform Cambodia into a self-sufficient agrarian utopia. Phnom Penh fell on April 17, 1975, and within hours soldiers were ordering roughly two million residents out of the city and into the countryside. Hospitals were emptied. Families were separated. People were pushed into forced labor. The transformation did not arrive after years of disappointing reform or a gradual radicalization of government. The transformation was the government. The old society itself had become the obstacle.

That is where I want to begin because I prefer opportunists to true believers every time. It sounds morally backward because sincerity is supposed to be a virtue and opportunism an insult. We admire the person who means what he says and distrust the politician who changes position after discovering that yesterday's conviction is polling at 34 percent. That is sensible when choosing friends. I am much less convinced that it is sensible when choosing the person who commands armies, directs laboratories, reorganizes economies, controls intelligence services, writes constitutions, or has access to nuclear weapons. In those circumstances I become intensely interested in whether the person wants something ordinary enough to lose.

Give me Bill Clinton. I mean Clinton almost archetypally. Give me the politician who wants to be loved, reelected, invited back, remembered fondly and photographed shaking the right hands. Give me Nixon looking at Communist China after a career built partly on anti-Communism and deciding that yesterday's ideological enemy might be strategically useful tomorrow. Give me the revolutionary who takes the palace and discovers that palaces have surprisingly comfortable bedrooms. Greed tells me what you want. Vanity tells me what you want. Ambition tells me what you want. Cowardice tells me what you fear. Nepotism tells me whom you love. Narcissism tells me which image of yourself must survive. None of those are virtues, but they are handles. The opportunist needs tomorrow because tomorrow contains the things he wants.

The true believer may have discovered something more important than tomorrow. It may be God, equality, revolution, race, nation, science, democracy, liberation, history, the planet, the proletariat, the market, the Constitution, technological transcendence or a purified humanity. I am agnostic about the content. I am not frightened merely because somebody believes something absolutely. I believe some things absolutely myself. I believe in Jesus. I believe the Second Amendment is an essential canary in the coal mine of American liberty. I believe rivers should not catch fire, children should not drink lead, getting lead out of gasoline was a triumph, and animals and wild places possess value independent of their usefulness to us. Conviction is not my problem. My problem begins when conviction produces a particular moral permission structure, one that can be expressed in four words that frighten me more than almost any others in politics: by any means necessary.

Put almost anything admirable before those words and listen to what happens. Save the world by any means necessary. Protect democracy by any means necessary. Defend liberty by any means necessary. End poverty by any means necessary. Achieve equality by any means necessary. Advance science by any means necessary. Preserve the nation by any means necessary. End human suffering by any means necessary. The object can be magnificent. The four words poison the grammar because any abolishes the limiting principle. Once the end is sacred enough, the means cease to possess independent moral weight. If a million people must suffer so that ten million can eventually inherit justice, the arithmetic begins to perform itself. If one generation must be sacrificed so that every generation afterward can live correctly, objecting to the sacrifice can start to look like moral cowardice.

Social scientists have actually studied a related phenomenon under the name sacred values. Work by Scott Atran, Jeremy Ginges and colleagues found that people do not always behave according to ordinary bargaining logic once an issue has moved into a morally non-negotiable category. In politically charged disputes, adding material incentives to a compromise could make some participants less willing to compromise rather than more willing, because the offer itself treated a sacred commitment as though it had a market price. Sacred values are not inherently pathological. Some things ought not to be purchasable. The danger appears when the priceless thing begins making everything else expendable.

Pol Pot is one of the purest examples because he removes the temptation to make this essay about religion. There is no need for a supernatural apocalypse when human beings are perfectly capable of manufacturing secular eschatologies. The Khmer Rouge did not merely seek a different tax system or a more equitable distribution of land. It sought a reconstructed human order. Cities, markets, religion, professions, institutions, inherited loyalties and enormous portions of existing Cambodian life became evidence of contamination by the world that had to disappear. The familiar story that people were killed merely for wearing glasses is too neat to describe what actually happened. Educated people, professionals, former officials, religious figures, people with foreign connections and others associated with the previous society were vulnerable because the revolution had made the previous society itself incriminating. The Khmer Rouge warning preserved by the Holocaust Museum was chillingly concise: “To keep you is no gain; to lose you is no loss.”

What particularly terrifies me about Cambodia is the speed. Revolutions usually discover that they need the people they have spent years denouncing. Somebody eventually has to maintain a turbine, deliver a baby, repair a pump, account for grain, run a railway, understand an electrical grid, manage a hospital and know which pipe leads where. Vanguard movements generally assume that there will still be a vanguard around to run things after victory. The Khmer Rouge came frighteningly close to treating the inherited human infrastructure itself as disposable. Phnom Penh fell in the morning and by afternoon soldiers were driving residents into the countryside. It was not a hundred-year transition. It was an attempt to turn history around immediately.

That is eschatology in the lowercase sense that interests me. A corrupt present is identified. The corruption is declared structural rather than incidental. Reform therefore becomes inadequate. Rupture becomes necessary. Suffering during the rupture becomes regrettable but historically meaningful. Purification follows. A redeemed society waits on the other side. Religion did not invent this architecture and modern secularism certainly did not abolish it. It simply supplied new names for the eschaton.

China's Cultural Revolution provides another enormous example. Stanford sociologist Andrew Walder reconstructed the first five years using published histories from 2,213 county and city jurisdictions, covering 99 percent of the relevant local jurisdictions. His estimates run from 1.1 million to 1.6 million deaths and 22 million to 30 million people subjected to political persecution. The detail that interests me most is that the vast majority of casualties in his reconstruction came from organized repression by authorities rather than the insurgent violence normally associated with images of youthful Red Guards. The restoration of order became extraordinarily violent too.

That is another feature of true-believer politics worth remembering. The revolution does not necessarily stop consuming people when the revolutionary side wins. Purity has no obvious natural stopping point. Yesterday's revolutionary can become tomorrow's reactionary because once one layer of impurity has been removed, the remaining impurity becomes easier to see. The person who helped expose the traitors may himself be insufficiently vigilant. The person who helped destroy the old hierarchy may have acquired suspicious authority in the new one. The eschaton retreats another hundred yards.

Max Weber identified the moral problem more than a century ago when he distinguished an ethic of conviction from an ethic of responsibility in Politics as a Vocation. The distinction was not between people with principles and people without them. The serious politician needs convictions. The problem is whether fidelity to the principle becomes sufficient moral accounting, or whether the actor also accepts responsibility for the foreseeable consequences produced by acting upon it. Good intentions do not sterilize bad means. Pure motives do not cause unintended consequences to vanish. Politics is the place where the person saying “but I meant well” can have a body count.

I am therefore obsessed with second-order consequences, third-order consequences and consequences so far downstream that nobody putting the policy into effect ever has to meet them. I am interested in blowback. The architect of a grand intervention naturally concentrates on the intended consequence because the intended consequence is the reason he built the machine. The unintended consequences live outside the beam of the projector. You want people to appreciate pristine environments, so you create ecotourism, and people who once sat obediently beside resort swimming pools put on Tevas and march into every pristine environment they can locate. You destroy an institution because it is oppressive and later discover that it was also containing something worse. You reorganize agriculture because the existing system is backward and discover that relationships you dismissed as inefficient were transmitting information nobody knew how to reproduce centrally. The more ambitious the intervention, the more interested I become in the least glamorous question in political philosophy: and then what?

This is where climate eschatology enters my thinking. I don't believe in climate change. I do believe in environmentalism, and I do not experience those positions as contradictory. I don't want mercury in fish, lead in children, toxic sludge in groundwater, spontaneously combustible rivers or species casually extinguished. I don't need the end of the world to want a clean river. What interests me politically is what happens to the permission structure when somebody else sincerely believes that the world itself is approaching catastrophe. If the alternative is planetary ruin, what intervention becomes too audacious to contemplate?

Stratospheric aerosol injection is almost comically perfect as an image because it is sunglasses for Earth: deliberately placing reflective material high in the atmosphere to reduce incoming solar energy. The important thing for my argument is not whether somebody has already deployed it globally. They have not. The important thing is that serious institutions study solar geoengineering precisely because people sufficiently worried about warming can imagine planetary intervention as a rational response. The National Academies has treated solar geoengineering as a subject requiring substantial research and governance precisely because its environmental, social and geopolitical consequences could be profound. Existential urgency changes the risk calculation. Ordinarily the fact that something acts on a planetary scale is an argument for restraint. Under an eschatological frame, the planetary scale of the alleged emergency becomes an argument for planetary-scale action.

That is the true-believer problem at maximum magnification. There is no spare Earth in the laboratory next door. Nobody can obtain meaningful consent from every human being who would live under the intervention. The people making the decision can be brilliant, educated, compassionate and convinced that they are acting for humanity. None of those qualities alters the topology of the risk. The size of the good intention does not shrink the possible size of the unintended consequence.

Then there is Oppenheimer. After Trinity, the first successful detonation of a nuclear weapon, J. Robert Oppenheimer later recalled a line from the Bhagavad Gita: “Now I am become Death, the destroyer of worlds.” The scientific accomplishment had reached a scale for which the language that came back to him was not merely technical. It was cosmic and divine.

I don't need to claim that Oppenheimer literally believed himself to be God to find the moment extraordinary. Human beings had acquired a power previous civilizations would have described through the vocabulary of gods, nature and apocalypse. That is what technological history keeps doing. It takes something once belonging to myth and converts it into engineering. Flight. Instantaneous communication across continents. Seeing inside bodies. Editing genes. Splitting atoms. Redirecting rivers. Potentially altering the atmosphere. The danger is not that scientists are uniquely evil. The danger is that intelligence does not solve the moral problem created by intelligence.

This is also why I do not accept the flattering image of science and scholarship as floating outside human systems of orthodoxy. Universities have hierarchies, initiations, credentials, specialist vocabularies, canonical literatures, accepted methods, senior authorities, mechanisms for determining who may teach, ceremonial robes and institutions devoted to reproducing themselves into the next generation. Graduate school can function socially like formation, the doctorate confers standing, tenure confers authority and the people who have spent decades ascending such systems are no more magically immune to hubris than people ascending churches, armies, parties or corporations. The dangerous move occurs when epistemic authority becomes moral authority. “I know more than you about atmospheric chemistry” may simply be true. “Therefore I know which risks civilization ought to accept” does not follow. Physics can determine whether a bomb detonates. Physics cannot determine whether a city ought to be underneath it. There is no laboratory instrument that measures ought.

The most powerful orthodoxy may therefore be the one whose priests are convinced that only other people have religions. The religious believer says revelation gives him the truth. The revolutionary says history gives him the truth. The nationalist says the nation gives him the truth. The activist says justice gives him the truth. The scientistic believer says the method gives him the truth. The language changes. My question remains the same: what does your certainty give you permission to do to everybody else?

The same question explains why disruptive climate activism interests me more than climate argument. If someone sincerely believes billions of lives and the continued viability of civilization are at stake, gluing oneself to a roadway, blocking traffic, attacking the protective barrier around a famous painting with food, interrupting public events or accepting arrest may make perfect moral sense inside the believer's denominator. Ten thousand irritated commuters are microscopic if your comparison class is planetary catastrophe. The behavior is useful to this essay because it shows how changing the perceived stakes changes the apparent cost of the means. What looks grotesquely disproportionate outside the eschatology can feel embarrassingly moderate inside it.

I do not care whether this psychology presents itself from the left or the right. I do not care whether the eschaton is religious or secular. I care about the believer's relationship with the future. Can he accept an 80 percent improvement, or does 80 percent constitute collaboration with the remaining evil? Can he lose an election and try again? Can evidence make him retreat? Can consequences embarrass him? Can he discover that an opponent was right about one thing? Can reality veto the theory? Those are much more interesting questions to me than which bumper sticker is on his car.

There is another person who frightens me even more than the person who says he wants to save the world. It is the person who says, “I want to change the world.” Change it into what?

The phrase is treated as automatically noble. Universities put it in recruitment material. Startups put it on walls. Politicians put it in speeches. Philanthropies turn it into mission statements. But change has no moral direction. Penicillin changed the world. So did the atomic bomb. Gutenberg changed the world. So did Mao. Norman Borlaug changed the world. So did Pol Pot. “I want to cure cancer” at least contains a stopping condition. Cancer is cured. “I want clean water” can be measured. “I want to bring art to people” tells me what the project is. “I want to change the world” tells me only that the speaker wants history to notice his fingerprints.

That is where Milan Kundera's Immortality enters the article for me. Human beings do not merely fear biological death. Some of us want to survive as images in the minds of people who never knew us. The politician starts imagining the biography. The billionaire imagines the foundation. The revolutionary imagines the statue. The scientist imagines the unit of measurement carrying his name. The artist imagines the canon. The eschaton becomes personal. I may die, but the world will have been altered so deeply that it cannot stop remembering me. That form of immortality can become more dangerous than ordinary opportunism because ordinary opportunism often anchors a person to the existing world, while the immortality seeker may need to wound the world deeply enough to leave a permanent scar.

Competent administrators rarely become immortal. Nobody writes an opera about the deputy minister who kept sewage treatment functioning for thirty-five peaceful years. History disproportionately remembers founders, conquerors, revolutionaries, destroyers, inventors, prophets and people associated with enormous ruptures. That creates a perverse prestige economy around scale. If your ambition is merely to leave the neighborhood slightly better than you found it, history may ignore you. If your ambition is to change civilization permanently, history might remember your name even when the change is monstrous.

This brings me to the event that supplies the other half of the essay, because history also gives us moments when almost everything depends upon somebody declining to become history.

On October 27, 1962, at the most dangerous point of the Cuban Missile Crisis, the Soviet submarine B-59 was submerged while American anti-submarine forces above it used signaling depth charges in an attempt to make it surface. Conditions aboard the Soviet submarine were miserable. Communications with Moscow had been lost. Captain Valentin Savitsky believed the submarine might be under attack and readied its nuclear torpedo. The Americans pursuing B-59 did not know that the submarine was carrying a nuclear weapon. Vasily Arkhipov, chief of staff of the Soviet submarine brigade and fortunately aboard B-59, recognized that the American forces were signaling rather than opening a war and helped calm Savitsky down. The submarine surfaced. The nuclear torpedo stayed aboard.

That one decision is impossible to price because the counterfactual is impossible to observe. Nobody can tell us exactly what President Kennedy, the Pentagon or NATO would have done if a Soviet nuclear torpedo had detonated against the American naval force during the Cuban Missile Crisis. The nearest plausible target included the carrier Randolph, with thousands of sailors and Marines aboard, and the United States had nuclear antisubmarine weapons available in the theater. What we can say is that a Soviet nuclear attack on American forces on October 27, 1962 would have occurred in the middle of a crisis already balancing at the edge of nuclear escalation.

Civilization may therefore owe an unknowable debt to a man who did not insist upon being right quickly enough.

Think about the psychological materials available to produce the opposite decision. Fear. Exhaustion. Military honor. Humiliation. Anger. An American force overhead. A captain who believed the war might already have started. A nuclear weapon ready to use. Every ingredient for an existential outburst was present. Arkhipov's contribution was not a brilliant theory of history. It was restraint under uncertainty. He acted as though the premise might be wrong.

That may be the most important characteristic I want in anybody who possesses enormous power: not lack of conviction, but the capacity to interrupt conviction with doubt before the consequences become irreversible.

The contrast with Pol Pot could hardly be cleaner. Pol Pot had a vision of the society that should exist and power sufficient to begin forcing the existing society toward it. Arkhipov faced a situation in which acting decisively could have made him a participant in one of the most consequential moments in human history, and instead helped create the possibility that nothing happened. One man pursued transformation so completely that approximately a quarter of his country died. Another man may have preserved an enormous portion of the existing world by refusing to assume that his side's worst interpretation of events was correct.

The first becomes history because of what he did.

The second matters because of what he declined to do.

That is why I keep coming back to opportunists. Opportunism is not the highest virtue. It is not even a virtue. It is merely one crude way of attaching powerful people to consequences. The politician wants another term. The billionaire wants another decade. The dictator wants his children to inherit the palace. The narcissist wants the statue. Those appetites can be disgusting, but they create stopping conditions. They give reality something to threaten.

The better brake is what Arkhipov appears to have possessed: epistemic humility under pressure. The willingness to admit, at the exact moment when pride, honor, ideology and fear are screaming for action, that the information may be incomplete and the interpretation may be wrong.

That is not weakness. On October 27, 1962, it may have been a survival system.

This is ultimately what I want from true believers, including the ones whose beliefs I share. I do not want Christians without faith, environmentalists without love of nature, scientists without curiosity, patriots without loyalty, activists without conviction, soldiers who believe nothing is worth defending or politicians whose only philosophy is tomorrow's poll. I want people whose beliefs contain stopping conditions. I want the environmentalist who can conclude that the intervention creates more damage than it prevents. I want the revolutionary who can admit that the revolution is hurting the people it was supposed to liberate. I want the scientist who can discover that something is technically possible and still say it should not be done. I want the general who can win but still decide the price of victory has become unacceptable. I want the religious believer who trusts God enough not to appoint himself God's project manager. I want the constitutional absolutist who understands that constitutional liberty includes other people exercising liberty in ways he despises.

Most of all, I want the person who can say I might be wrong while he still has time to do something useful with the possibility.

Not because truth does not exist. Not because conviction is embarrassing. Not because morality is relative. Not because all perspectives deserve equal respect. I want that opening because complex systems contain relationships nobody completely understands, because interventions produce consequences their architects did not intend, because human beings surprise us, because beautiful theories eventually encounter plumbing, because intelligence does not eliminate hubris, because institutions reward conformity as well as truth, because history does not provide control groups, and because eight billion people are not raw material for somebody else's perfected experiment.

My own Christianity contains a wonderfully simple version of the same restraint. I am a Jesus freak, but I do not believe I have been appointed assistant regional manager of the Apocalypse. Jesus tells his followers that they do not know the day or hour. The Messiah may be part of my faith, but managing the Messiah's travel arrangements is not my job. I believe in the Messiah. I distrust anybody who thinks the Messiah requires his help getting here.

I think of the Godhead as white light hitting a prism, with human religions and traditions appearing as different colors after that light passes through culture, history, language and temperament. I happen to live most comfortably in the Christian ray. Jesus is where the light makes the most sense to me. That does not give me ownership of the light source, and it certainly does not give me permission to break the prism because everybody else is standing in the wrong color.

That is the same standard I want applied everywhere. The activist, the professor, the scientist, the priest, the revolutionary, the patriot, the billionaire, the general, the environmentalist, the president and me all remain human beings standing in front of systems larger than ourselves. None of us gets an exemption because our intentions are pure.

Believe whatever you believe. Believe it fiercely. Argue for it. Evangelize it. Vote for it. Organize around it. Spend your life building it. Tell me I'm an idiot for rejecting it. I can live with all of that.

Just tell me there is something you will not do to me in order to get there.

Because the moment your answer becomes by any means necessary, I no longer care how beautiful your destination is. I am looking for the exits.

And if the person beside the launch control can still say, at the worst possible moment, we may be wrong, I would very much like him to remain in the room.

In January 2024, the United States counted 771,480 people experiencing homelessness on a single night, the highest number recorded in HUD's national point-in-time series. In 2023, 54.2 million Americans were classified as needing substance-use treatment, but only 12.8 million received it, meaning more than three quarters of the people judged to need treatment did not get treatment as federally defined. Housing First, one of the dominant responses to chronic homelessness, performs much better at the specific thing its name promises: randomized-trial evidence finds participants about 2.5 times as likely to be housed 18 to 24 months later, while showing no clear corresponding reduction in substance use. None of these numbers proves conspiracy, corruption, indifference, or failure. They suggest something more interesting. America has become capable of spending enormous amounts of money keeping vulnerable people housed, fed, medically stabilized, enrolled, monitored, protected from immediate catastrophe and repeatedly returned from the edge without necessarily solving the conditions that put them there. A lost cause, in ordinary English, is something into which further investment is no longer expected to produce a worthwhile result. My question is whether we have constructed a humane public version of that judgment about people: not abandonment, not extermination, not cruelty, but maintenance after expectation has been withdrawn. We will keep you alive. We will keep trying to keep you indoors. We will reverse the overdose. We will replace the phone. We will renew the benefits. We will respect your autonomy. What we may no longer be doing, in some cases, is seriously betting on your return.

That distinction matters because maintenance is not nothing. Housing First really does house people. A systematic review of randomized trials found that participants were 2.46 times as likely as control groups to be housed 18 to 24 months later, with fewer emergency-department visits and hospitalizations. The same review found no clear difference in substance use, and another randomized Housing First study likewise found no reduction in daily substance use after 12 or 24 months. That does not make Housing First fraudulent. It means housing is an outcome and recovery is another outcome. Public policy becomes confused when it treats success at the first as proof that the second question has been answered.

The addiction numbers make the distinction harder to ignore. SAMHSA estimated that 54.2 million Americans aged twelve or older needed substance-use treatment in 2023. Only 12.8 million received it, 23.6 percent of those classified as needing treatment. The remaining three quarters were not all homeless, were not all incapacitated, and were not all asking for treatment, but the scale tells us something important about the world in which arguments about addiction policy occur: the United States is nowhere near a condition in which everybody who needs recovery treatment is receiving it and policymakers are merely choosing whether to be strict or permissive about the final few holdouts. Most of the federally defined treatment need sits outside treatment altogether.

Homelessness produces the same uncomfortable split between keeping someone from disaster and restoring someone to ordinary participation. HUD's national count found 771,480 people homeless on a single night in January 2024. HUD does not merely count beds or meals and call the problem solved; its national data systems track housing placement, returns to homelessness and other outcomes. The bureaucracy itself therefore recognizes the difference between servicing homelessness and ending an episode of homelessness. But the sheer number of people remaining outside or in shelter means that the distinction between management and durable exit is not philosophical decoration. It is the problem.

This is where I want to use the phrase black sheep paternalism. Every sufficiently large family eventually learns the difference between loving somebody and believing in his judgment. There is the brother whose rent gets paid directly because nobody wants to discover what happens if the rent money arrives as cash. There is the cousin who remains unquestionably part of the family but is not entrusted with the principal of the trust because everyone knows that generosity can become accelerant. The family does not revoke his membership. It may spend astonishing amounts of money preserving it. What quietly disappears is investment in the future. Nobody stops caring whether he eats. They stop asking what he might become.

That is the possibility I want to examine in American social policy. Not whether social workers secretly want homelessness, whether nonprofits conspire to preserve their grants, or whether harm reduction is an elaborate plan to kill addicts. Those theories require motives that are unnecessary to explain institutional behavior. A system can be staffed almost entirely by decent people and still settle into an equilibrium in which keeping someone alive, housed or administratively attached becomes easier to accomplish than restoring capability. Institutions respond to incentives, measurements, legal constraints and the limits of human agency. No villain is required for a system to become excellent at maintenance.

I know the difference because I spent roughly fifteen years volunteering at Miriam's Kitchen in Washington, working as a sous chef and dining-room captain. Miriam's began with the uncomplicated act of feeding hungry people and later developed social services, outreach and housing work. Its own strategic history says essentially that: the organization started in 1983 because people needed meals, then realized that seeing the same people return year after year meant meals alone were insufficient. Its filings describe the dining room as an entry point through which guests could be connected to income, healthcare and housing. One filing reported roughly 70,000 meals annually, basic services for about 4,000 people and more intensive social-service work with roughly 1,200 of them.

That is not how I remember an industrial complex. I remember a flower.

The food was the flower. People came because breakfast and dinner were worth coming for, because the room was clean, because there were bathrooms, coffee, conversation and a place to sit without being treated as contamination. Once people came through the door, the social workers could find them. Those were the bees. Someone who might otherwise be almost impossible to reach was suddenly sitting in the same place often enough for somebody to ask whether he had identification, Medicaid, Social Security, veterans benefits, healthcare, a telephone, housing paperwork, clean clothes, treatment, education or a job prospect. Miriam's own historical performance reporting shows exactly that mixture: meals on one side, housing, income and healthcare connections on the other.

The guest did not have to accept any of it. He could eat and leave. There was still a price, because every civilized room has a price. You had to behave. You could not threaten the other guests, terrorize the staff, start fights or make everybody else pay for your inability to regulate yourself. There were usually some very large men around capable of making that boundary clear when conversation failed. I do not consider that a violation of autonomy. A dumpster makes fewer demands on your conduct, but the dumpster also does not cook breakfast. The price of sharing a room with other human beings is acknowledging that they exist too.

That distinction between unconditional sustenance and conditional participation is where Lost Cause begins to get difficult. There is a kind of compassion that says, "You are hungry, so eat." There is another kind that says, "You are in trouble, so here is a ladder." There is another that says, "You keep refusing the ladder, but we are not going to let you freeze." Eventually somebody has to decide whether the ladder remains the objective or whether keeping the person alive at the bottom of the hole has become the objective.

That decision is rarely announced. Nobody writes LOST CAUSE across a case file. Institutions use gentler language: stabilization, harm reduction, client choice, supportive services, permanent housing, continuity of care. Much of that language describes useful and humane things. The danger is not the vocabulary. The danger appears when stabilization stops being a stage and becomes a destination.

There was a man who lived for a long time on a metal bench near Columbia Pike and South Barton Street in South Arlington. He did not merely sit there occasionally. The bench became where he lived. Over time he became filthy and increasingly unwell. A woman from social services used to visit him regularly, alone, without security, bringing water and little necessities and talking with him. Nobody dragged him away because his life offended the neighborhood's aesthetic expectations. For a long time the response was astonishingly restrained: check on him, offer things, talk, leave him alone.

Eventually two Arlington police officers appeared. I watched from Starbucks for hours. They wore blue rubber gloves because he was dirty, but there was no theatrical show of force. They stood several feet away, often near the curb, spoke to him and waited. Eventually an ambulance arrived. Eventually he went onto a gurney and was taken away. I do not know what happened to him afterward.

I know what happened to the bench. Armrests were welded across it so nobody could comfortably lie there again, and similar divisions subsequently appeared on other benches nearby.

The scene has stayed with me because nobody in it needs to be a bastard. The social-services worker could be doing exactly what she should. The police could be exercising exemplary restraint. The emergency personnel could be acting at exactly the threshold at which intervention became medically necessary. The people who altered the benches could reasonably believe that public furniture should remain usable by the larger public. Every individual decision can make sense and the total picture can still be strange.

For a long time the neighborhood tolerated the man. When his condition became intolerable to his body, he was taken somewhere else. Then the environment was altered so another person could not use it in quite the same way.

That is not evidence of a conspiracy. It is evidence of an equilibrium.

Human beings create equilibria everywhere. A chronically homeless man sleeps at Starbucks until he repeatedly screams at customers, at which point somebody complains. Perhaps a frightened customer gets a restraining order. Perhaps the store stops providing the accommodations it previously offered. Perhaps the police eventually enforce boundaries that nobody wanted to enforce earlier. The result can look cruel if the only person in the frame is the homeless man. Put everyone else back into the photograph and the problem changes. The woman being screamed at is also a citizen. The barista is also working. The person who wants to sit on the bench is also part of the public. Compassion does not abolish competing claims.

This is why autonomy cannot settle the question by itself. "Let him make his own choices" works cleanly only when the costs of the choices remain inside his own skin. Addiction, homelessness, untreated psychosis, violence, public intoxication and severe self-neglect routinely spill into other people's lives. At the same time, the opposite principle, "we know what's best, therefore we may compel you," has no natural stopping point. If preserving life overrides everything else, civilization rapidly becomes a padded room.

Life is immensely valuable. It is not infinitely preservable.

Human beings sometimes speak about death as though mortality were a systems failure. We are not Tolkien elves who would live indefinitely if only we optimized nutrition, medicine, housing, public policy and personal virtue. We are much closer to mayflies. Eighty years feels enormous because it contains almost everything we will ever experience, but eighty years is still eighty years. Everyone dies. Some people die after doing everything right. Some survive astonishing amounts of stupidity. Medicine, sanitation, housing, sobriety and social order can buy years, sometimes decades, and those years are worth buying. They do not repeal the terms of the species.

That matters because a society cannot define compassion as preventing every preventable death without eventually consuming liberty in the attempt. The alcoholic can be confined because alcohol might kill him. The addict can be confined because fentanyl might kill her. Motorcycles can disappear. Dangerous food can disappear. Solitude can disappear. Risk can disappear. Eventually life itself becomes the management of remaining alive.

The other extreme is equally barren. "His choice" can become a phrase we use when we have ceased trying.

This is where harm reduction becomes intellectually interesting rather than politically symbolic. Keeping somebody from contracting HIV, reversing an overdose or giving someone a stable apartment is not surrender. These interventions can preserve exactly the thing recovery requires: a living person. Housing First substantially improves housing stability and reduces some emergency healthcare use. It does not appear to make substance use worse, and it should not be accused of doing so. But neither does housing automatically cure addiction. The distinction is important because harm reduction makes the most moral sense as a floor beneath a person, not as a ceiling above him.

A social system therefore has at least three possible relationships with the person at the bottom. It can abandon him. It can control him. Or it can maintain him while continuing to offer ways back into fuller participation. The third is usually the least terrible option and therefore the one liberal societies gravitate toward. But it contains a hidden temptation. Maintenance is easier to standardize than transformation. A meal can be counted. A housing placement can be counted. A naloxone reversal can be counted. A Medicaid enrollment can be counted. A benefit application can be counted. Becoming capable, responsible, employable, trustworthy, sober, connected and independent is considerably harder to put into a quarterly dashboard.

The numbers themselves can therefore tell two stories at once. Seventy thousand meals is seventy thousand instances in which somebody ate. That matters. It is also seventy thousand meals that may or may not have changed the trajectory of the person eating them. A housing placement is a person no longer sleeping outside. That matters enormously. It does not tell you whether the person stopped drinking, found work, repaired a family, learned anything, made friends, accumulated savings or stopped requiring supportive services. None of those omissions invalidates the housing placement. They simply remind us that human flourishing cannot be collapsed into a single administratively convenient variable.

The mistake is assuming that because we cannot guarantee transformation, we should stop expecting it.

That is black sheep paternalism at institutional scale. You still get the jersey. You are still on the team. You get Gatorade and orange slices. Your name remains on the roster. Nobody throws you out of the stadium. But you sit on the bench permanently because everybody has privately decided there is no plausible game in which putting you on the field improves the outcome. At first they keep coaching you anyway. Eventually they stop.

The dangerous moment is not when the team cuts you. The dangerous moment is when it loves you enough to keep you forever and believes in you too little to ask anything difficult of you.

Melville's Bartleby understood the other side of this problem long before there were federal grants for supportive housing. Bartleby simply "would prefer not to." His employer keeps attempting accommodations because he cannot bring himself to discard another human being, but accommodation cannot produce desire. Eventually the problem becomes almost insoluble: what does one person owe another person who increasingly refuses participation in his own rescue?

Recovery traditions know this problem intimately. Twelve-step language begins with powerlessness, but it does not end with permanent passivity. Admitting that one cannot control alcohol is supposed to begin a process of radical personal change. Families learn a parallel lesson through detachment: you cannot recover another person on his behalf. That is not the same thing as concluding that he cannot recover.

The distinction is everything.

A lost cause is not necessarily somebody who cannot be saved. Sometimes it is somebody from whom the rescuers have stopped expecting a return on investment. Sometimes that judgment is rational. There are sunk costs in human affairs just as there are in war, business and medicine. Resources are finite. Attention is finite. Families become exhausted. Social workers become exhausted. Communities have other people to protect. Nobody possesses an infinite obligation to destroy himself attempting to rescue someone who continually refuses rescue.

But people are not failed factories or abandoned military campaigns. The arithmetic cannot be identical because a person can surprise you. An alcoholic can become sober after thirty years. A homeless veteran can finally trust the wrong social worker on exactly the right morning. Someone who burned every bridge at twenty-five can become the person rebuilding bridges at forty-five. The fact that transformation cannot be compelled does not make it imaginary.

So the humane society has an almost impossible task. It must be willing to maintain people whom it cannot presently repair without allowing maintenance to become a declaration that repair is impossible. It must respect autonomy without laundering abandonment through the language of choice. It must impose enough order that everyone else can live in the same city. It must sometimes accept "no." It must sometimes ask again tomorrow. It must recognize that some interventions work, some do not, and no program converts mortality into immortality.

Most of all, it has to resist the temptation to confuse having stopped expecting something from a person with having become compassionate toward him.

Feed him. House him if you can. Give him water. Replace the identification. Reverse the overdose. Treat the infection. Pay the rent directly if handing him cash would kill him. Put boundaries around him when his behavior harms other people. Leave the door open after the fiftieth refusal.

Just don't write the DNR on his future.

The most successful international development program in the world may be the one nobody invented.

It has no headquarters in Washington, Geneva, Paris or New York. It has no president, no board of governors, no annual donor conference, no five-year plan and no mission statement. Nobody flies business class to a resort hotel to discuss stakeholder engagement. There are no consultants producing a 140-page report explaining what the recipients need before the recipients are allowed to receive it.

It is a roofer in Virginia sending $500 to Guatemala.

It is a nurse in London sending pounds to Nigeria. It is an Indian engineer in California sending dollars to his parents. It is a Pakistani cabdriver in Chicago, a Filipino caregiver in Toronto, a Bangladeshi restaurant worker in Queens, a Moroccan in France, a Salvadoran in Maryland. It is somebody getting paid in a rich economy and sending part of the money to somebody he knows personally in a poorer one.

We call these transfers remittances, which may be one of the dullest words ever invented for one of the largest voluntary transfers of wealth in human history.

The World Bank estimated that officially recorded remittances to low- and middle-income countries reached $685 billion in 2024. The real number is larger because money traveling through informal channels does not necessarily appear in the official totals. India alone was expected to receive $129 billion, Mexico $68 billion, China $48 billion, the Philippines $40 billion and Pakistan $33 billion. In Tajikistan, remittances were estimated at 45 percent of GDP. In Tonga they were 38 percent. In Nicaragua and Lebanon, 27 percent.

Then comes the number that should rearrange the furniture in every conversation about international development: according to the World Bank, remittances to low- and middle-income countries have grown larger than foreign direct investment and official development assistance combined. During the decade leading into 2024, remittances increased by 57 percent while foreign direct investment declined by 41 percent.

The world's poor did not wait for us to solve development.

Their relatives got jobs.

Mamá needs $400

There is an important difference between institutional development money and a remittance that has nothing to do with whether the people working at the World Bank, IMF, USAID or an international NGO are good people. I assume most of them are trying to accomplish exactly what their institutions say they are trying to accomplish.

The difference is informational.

If I want to improve the life of a family in rural Guatemala from an office in Washington, I first need to determine what the family needs. I need data. I need criteria. I need a program. I need a mechanism for distributing the money. I need accountability. I need to determine whether the program worked. Somewhere in this process there will eventually be a person whose roof leaks.

The woman's son already knows the roof leaks.

Mamá needs $400.

Send $400.

That is the entire needs assessment.

Remittances have an extraordinary informational advantage because the sender and recipient generally know one another. The migrant does not need to commission a poverty survey to discover that Dad needs medicine, his sister needs school fees, the roof needs repairing or somebody needs $2,000 to buy equipment for a tiny business. The family has information that no development institution could acquire at comparable cost because they live inside the problem.

The IMF itself describes remittances as unusually targeted toward recipient needs and notes that research has repeatedly associated them with poverty reduction. World Bank research has found that remittances help households smooth consumption during shocks, ease working-capital constraints for farms and small businesses, and increase spending on education, health, entrepreneurship and property.

This is development without the development industry standing between the money and the person.

No debt, no austerity, no application

A World Bank loan and $500 sent home by your daughter are not substitutes. Governments need infrastructure. Countries sometimes need financing on a scale that families obviously cannot provide. Nobody is building a national electrical grid because Uncle José sent money from Houston.

But at the level of the household, remittances possess features institutional finance cannot reproduce.

There is no repayment schedule.

There is no interest rate.

There is no structural adjustment program.

There is no requirement that the recipient privatize the electric company, reform the pension system, cut a subsidy, change the tax code or satisfy a lender before the next tranche arrives.

There is not even necessarily a requirement that the money be spent wisely.

That last part matters more than it appears.

Development institutions understandably want to make sure development money is used for development. A brother sending money to his sister trusts his sister to decide what development means this month. Maybe it is tuition. Maybe it is antibiotics. Maybe it is concrete. Maybe it is a refrigerator. Maybe it is repaying an ugly local debt. Maybe it is a motorcycle that makes a job possible. Maybe it is chickens.

The recipient possesses the money and therefore possesses the decision.

That is not merely economic assistance. It is autonomy.

Cash changes who you have to ask

Money is power in the most boring and immediate sense of the word. It expands the number of things you can do without asking somebody else.

If you live close to subsistence, your choices can be controlled by whoever controls access to work, food, credit, land, transportation or government assistance. That person may be a perfectly decent employer or official. He may also be a landlord, political boss, local oligarch, corrupt official, predatory moneylender, abusive relative, cartel, landowner or whoever happens to control the narrow economic passage through which you must travel.

Then your daughter starts sending dollars every month.

Nothing magical happens. You are not suddenly rich. The government does not disappear. The local strongman does not evaporate in a puff of smoke.

But perhaps you no longer need his loan.

Perhaps your child can attend a different school. Perhaps you can pay transportation to a city where the jobs are better. Perhaps you can survive long enough to tell an employer to go to hell. Perhaps you can buy fertilizer instead of owing part of the harvest to the person who financed it. Perhaps you can repair your house without asking a politician for a favor. Perhaps you can accumulate enough cash to become the person who owns the tiny store instead of the person who buys on credit from it.

That is why I think remittances are politically more interesting than we usually admit.

The World Bank describes strong evidence that they reduce poverty, and the IMF has found that remittances can help households maintain consumption particularly during periods of fiscal consolidation. That sounds bloodless until you translate it into ordinary life.

When the government tightens its belt, your cousin in New Jersey does not necessarily tighten yours.

The migrant monetizes the village

There is another form of remittance that we often fail to recognize because the money never crosses an international border.

A young person leaves a subsistence village and gets a factory job in a manufacturing center. Maybe she assembles electronics. Maybe she makes clothing. Maybe she works in food processing or automotive parts. Her parents have spent much of their lives producing things they consume themselves or trading inside a local economy with very little cash.

Then money starts coming home.

This changes more than the family's consumption. It monetizes choices that previously did not exist.

Cash buys fertilizer, tools, animals and building materials. It makes it possible to pay somebody else for labor. It creates customers for the little shop. It gives somebody a reason to start a transportation service. It pays school fees. It produces savings. It may eventually become land.

You can criticize global supply chains until everyone involved is blue in the face, and plenty of criticism is deserved. Factory labor can be brutal. Workers can be exploited. Communities can lose their youngest and strongest people. A village that once depended on a landlord may become dependent on a factory 400 miles away.

But dependence has changed shape, and sometimes the new dependence gives the worker more bargaining power than the old one.

The ability to leave is itself an economic asset.

Hermano bought the farm

My favorite example lives in my building.

I call him Hermano. He is Guatemalan, works construction, drives an old Toyota Tundra, smokes too much, drinks too much and remains convinced that I need to accompany him to Guatemala so that somebody can find me una novia. I remain unconvinced about the final proposal.

If you look only at his American balance sheet, you could construct a very persuasive story about deprivation. He rents here. He works with his body. He does not possess the visible portfolio of assets Americans associate with arrival.

Then you look south.

Hermano bought a farm in Guatemala with the money he earned here. He built the family compound. His extended family lives on it. The farm did not precede the migration as some ancestral estate he occasionally helps maintain. The migration produced the farm.

In America he can look like a working-class immigrant in an old truck.

In Guatemala he is the lord of the manor.

The mistake is assuming that the apartment in America represents his wealth. It may represent his overhead.

The wealth went somewhere else.

That is remittance economics in one human being. Labor performed in a high-wage economy becomes capital in a lower-cost one. The difference between what his work can command here and what those dollars can purchase there becomes land, buildings and security for people he loves.

No NGO bought Hermano's farm.

No development bank bought Hermano's farm.

Hermano bought the fucking farm.

This is soft power nobody has to administer

There is another consequence of all this money moving around the world that receives remarkably little popular attention.

Remittances connect households in poorer countries directly to the economic health of richer ones.

If your son works in America and sends dollars, you now have an interest in his continued ability to work in America. You have an interest in the dollar retaining purchasing power. Your local bank and money-transfer businesses have an interest in systems capable of moving dollars. Merchants become accustomed to receiving money ultimately sourced from the American economy.

The dollar travels home even when the worker does not.

Nobody needs to have designed this as an American soft-power strategy for it to function as one. Hundreds of millions of family relationships can accomplish what a propaganda campaign cannot: they make the prosperity of the United States materially relevant to households that may never contain a person who has visited it.

The migrant does not simply send dollars.

He exports demand for dollars.

The same thing happens with euros, pounds and currencies earned in other wealthy labor markets, but the scale and international role of the dollar make the American case particularly interesting.

Unlike a conventional aid program, this influence is radically decentralized. There is no single faucet to shut. Formal transfers can be regulated, taxed or made difficult, but people have been moving money across borders for considerably longer than Western Union has existed. Cash, informal transfer networks, digital payment systems and increasingly cryptocurrencies and stablecoins give people additional ways to move value when formal rails become expensive or restrictive.

The international financial system keeps becoming harder to monopolize because the technology keeps becoming easier to carry in your pocket.

The recipients are allowed to waste it

One of the reasons remittances offend the managerial imagination is precisely what makes them powerful.

The recipient is allowed to screw up.

She can make a terrible investment. He can buy the wrong truck. They can lend money to the idiot cousin. Somebody can spend too much on a wedding. Somebody can open a shop that fails after eleven months.

This is normally regarded as a weakness.

I think it is adulthood.

We do not require affluent Americans to demonstrate optimal capital allocation before letting them possess money. We allow them to buy boats they never use, kitchen renovations nobody needed, cryptocurrency at the top, ugly handbags and Pelotons that become clothing racks.

Then we become extremely concerned about whether a poor person in Nicaragua will allocate $500 according to the optimal development model.

Give him the money.

It is his brother's money anyway.

Remittances are not magic

There are costs. Sending money remains surprisingly expensive. World Bank data put the global weighted average cost of sending $200 at about 4.9 percent in early 2025, still well above the United Nations Sustainable Development Goal target of 3 percent.

Heavy dependence on remittances can also produce distortions. A community can become reliant on income generated somewhere else. Young workers leave. Families receiving foreign income may pull ahead of neighbors without relatives abroad. Large inflows can affect exchange rates and labor markets. Remittances are excellent at helping households and considerably less capable of building sewer systems, reforming courts or running a central bank.

None of that makes the phenomenon smaller.

It makes it real.

The World Bank itself describes the aggregate growth effects of remittances as more complicated than their household effects. The poverty-reduction evidence is strong; the question of how that translates into national economic growth contains countervailing forces.

Fine.

I am interested in the household first.

Because nations do not eat dinner.

People do.

Development from below

International development has spent decades asking how rich countries can transfer enough knowledge, capital and institutional capacity into poor countries to accelerate their development.

Migrants discovered another method.

Move the worker.

Let the worker plug into the richer economy. Let him earn a wage that would have been impossible or improbable at home. Then let some portion of that wage travel backward through the family.

The migrant becomes his own tiny multinational corporation.

Production happens in Virginia. Capital accumulation happens in Guatemala. Management is distributed between family members who already know one another. The investment committee may consist of somebody's mother yelling through WhatsApp that the west wall still leaks.

It is messy, inefficient, frequently sentimental, occasionally stupid and almost impossible to centrally control.

That is precisely why I love it.

We have spent generations designing institutions to lift the global South from above while millions of ordinary people quietly built an enormous development system from below.

They migrated.

They worked.

They sent money home.

They bought medicine and tuition and fertilizer and concrete and motorcycles and chickens and land. They built houses. They supported children. They kept parents alive. They opened shops. They accumulated assets. They transformed American dollars and European euros into decisions that somebody else no longer got to make for them.

The World Bank can measure it. The IMF can study it. Governments can regulate it. Economists can argue about its macroeconomic effects.

But nobody owns it.

Somebody's daughter gets paid on Friday.

Mamá needs $400.

Send $400.

I have a friend in my building whom I call Hermano. He calls me Chris, which makes the arrangement embarrassingly unequal because I cannot reliably remember his actual name. For a while I also kept deciding that he was from El Salvador until he repeatedly reminded me that he is from Guatemala. This is perhaps useful information to disclose before I begin making sweeping observations about immigration. I know the man considerably better than I know his biographical metadata.

Hermano is about five-foot-four, handsome, solid, with a little potbelly, and almost invariably dressed like a cowboy. He drives an old Toyota Tundra and works construction, which sometimes pays well and sometimes sucks because construction sometimes sucks. He smokes too many cigarettes, drinks too much, has a wife and children I adore, and remains convinced that my principal failure as a man is that I do not have una novia. This is apparently a problem he can solve for me in Guatemala. I have declined his assistance. We mostly talk about work, coffee, tea, women, trucks, ordinary things. I don’t interview him about the immigrant experience because he is my friend, not my fieldwork.

By the vocabulary that has settled over American discussions of immigration, however, Hermano belongs to a population we are forever calling the most vulnerable. I have come to hate that phrase. I don’t hate the people it is intended to describe. I hate what the phrase quietly says about them, and what it allows the person saying it to believe about himself.

The most vulnerable among us. The words sound compassionate. They also establish the hierarchy before the conversation has begun. I am the person standing securely up here, educated and competent enough to understand the complicated world. You are the person down there upon whom the world acts. I have choices and preferences. You have needs and circumstances. I pursue opportunities while you are displaced by forces beyond your control. I take risks while you are exposed to them. I can tell you what I want. You require an advocate who has been trained to explain what you need.

There is something oddly anthropological about it. We have turned adults into a population to be observed, interpreted and protected, and somehow never notice how superior we have made ourselves in the process. The comfortable American with a graduate degree peers down at the immigrant who crossed several countries, navigated borders and bureaucracies, found somewhere to sleep, found work, figured out transportation, learned enough English to function, established a network, acquired a vehicle, supported relatives thousands of miles away and built a life in a place where he may not even have the legal right to remain, and concludes that the immigrant is the helpless one.

I sometimes think we should stop teaching immigrants English before they become fluent enough to read what we write about them. They might be insulted.

The wrong balance sheet

Hermano does not own much in America. He rents. He has an old Tundra. He works with his body. I don’t know his immigration status, and it has never mattered to our friendship. If you took a snapshot of his American household and fed it into the usual machinery for measuring economic success, you could probably produce a fairly convincing account of what he lacks.

The trouble is that you would be looking at the wrong country.

Hermano bought a farm in Guatemala. Not inherited family land that he helps maintain, and not somebody else’s farm to which he occasionally sends a little money. His farm. His American labor and his remittances bought the land and built the family compound on it. His extended family lives there and benefits from what he created. In America he is a construction worker who rents an apartment and drives an old Tundra. In Guatemala he is, for lack of a less ridiculous phrase, the lord of the manor. He created the manor.

This is one of the things Americans misunderstand when we measure an immigrant entirely by what he consumes here. We look at his apartment and think we are looking at his wealth. We may only be looking at his overhead. The crowded apartment, the old truck and the physically punishing job can be interpreted as evidence that the American Dream has failed him when, from his point of view, America may be doing precisely what he came here to make it do. The wealth didn’t disappear because he didn’t spend it here. It became land somewhere else.

The World Bank estimated that officially recorded remittances to low- and middle-income countries reached about $685 billion in 2024, and the actual amount is probably higher because not every transfer travels through an official channel. India alone was expected to receive about $129 billion, Mexico $68 billion, the Philippines $40 billion and Pakistan $33 billion. In some smaller economies the numbers are extraordinary relative to the entire national economy: the World Bank estimated remittances at 45 percent of GDP in Tajikistan, 38 percent in Tonga and 27 percent in Nicaragua. Most remarkably, remittances to low- and middle-income countries were expected to exceed foreign direct investment and official development assistance combined. World Bank remittance estimates

That is not pocket money. It is one of the largest development systems on Earth, except that it has no headquarters, no president, no annual summit and no fellow in Washington explaining its theory of change. It consists largely of people working somewhere richer and sending part of what they earn to people they love somewhere poorer.

I find something magnificent about that. I have spent enough of my life hearing about the IMF, the World Bank, USAID, development programs, structural adjustment, aid packages and the entire elaborate machinery by which wealthy countries attempt to improve, influence, cajole or reorganize poorer ones. Meanwhile somebody’s daughter in Northern Virginia is cleaning houses and wiring $400 to her mother. Somebody’s son is driving Uber all night and sending dollars home. Somebody is roofing houses in August heat while his family uses the money to educate a child, repair a roof, buy land, start a business or simply eat better.

There is no austerity program attached to the transfer. Mamá needs $400. Send $400.

I sometimes wonder how much geopolitical power is buried inside this enormous river of private money, and whether governments understand it considerably better than we discuss it publicly. Nobody needs to have designed it for the effect to exist. Millions of households outside the United States have a direct financial interest in an American economy capable of employing their relatives and producing dollars worth sending home. Remittances do not merely export money. They help export the usefulness of the dollar itself.

More importantly for this essay, they tell us something about the people doing the sending. A person may willingly accept a comparatively austere life here because his economic unit is not the individual American consumer. It may be a family spread across two countries. Five or six adults sharing housing and working different shifts can look like deprivation if your ideal is one household, one detached home, two cars and a kitchen island the size of Guam. It can look rather different if those adults are pooling costs while accumulating assets somewhere else.

Hermano is not supporting a family farm that existed before him. The farm is the receipt.

The American Dream depends enormously on which American Dream you were trying to buy.

The dream is a wager, not a warranty

We have made the American Dream strange by talking about it as though America once issued successful people at the border and has recently stopped doing so. There was never a guarantee. There was an opportunity to take a shot, surrounded by millions of other people taking theirs.

Americans understand this perfectly well in almost every other context. The Bureau of Labor Statistics followed private-sector business establishments born in 2013 and found that only 34.7 percent were still operating ten years later. Nearly four out of five survived their first year, but by year five only about half remained. Bureau of Labor Statistics business-survival data We nevertheless encourage people to start businesses. We don’t regard every closed restaurant as proof that entrepreneurship was fraudulent.

College works the same way. Among first-time, full-time students who began bachelor’s programs at four-year institutions in 2014, 64 percent had graduated from the institution where they started within six years. More than a third had not. National Center for Education Statistics graduation data We still tell people that education creates opportunity because opportunity and guaranteed completion have never meant the same thing.

Immigration is a wager of the same species with considerably higher stakes. The immigrant arrives in a country where other people want the same apartment, the same job, the same customer, the same loan and the same patch of economic oxygen. Some of those people will be generous. Some will be indifferent. Some will be racist, sexist, nationalist, regionalist, nepotistic, corrupt, envious or simply assholes. Somebody will prefer his cousin. Somebody will hate your accent. Somebody will underbid you. Somebody making a decision that wrecks your month may not have thought about you for ten consecutive seconds.

That is the world. It is not a load-balancing system.

The American Dream was never the proposition that every person who works hard receives the same outcome. It is closer to the proposition that there are enough openings in the system to make taking the shot worthwhile. Preparation helps. Persistence helps. Luck helps enormously. Failure remains available in industrial quantities.

This is why I don’t need every immigrant story to end with the protagonist owning a chain of restaurants and sending three children to medical school. That is another sentimental way of refusing immigrants ordinary humanity. Some immigrants will become fabulously successful. Some will build stable middle-class lives. Some will spend forty years doing physical labor and renting. Some will make money here and build wealth somewhere else. Some will hate America and leave. Some will get deported. Some will make terrible decisions. Some will drink too much. Some will fail completely.

Adults do that.

Respecting the immigrant means respecting the possibility that he understood failure was on the menu when he ordered.

Maybe he came for the free stuff

I don’t even require noble motives.

If somebody halfway around the world comes to believe that the United States or Western Europe is temporarily allowing unusually easy entry, generous benefits or some other exploitable opportunity, I am not morally scandalized that he might try to take advantage of it. I might think the policy creating the incentive is idiotic. The government may eventually change the rule, close the door, deny the claim or send him home. Those are questions for immigration policy.

But the migrant looking at the opportunity and thinking, I should shoot my shot before that window closes, is behaving recognizably like a human being.

I would have a Plan B. If I thought I might get a year somewhere richer, or five years, or ten, and could work, meet people, have an adventure and send money home before the political weather changed, I would calculate the risk against the reward. So would millions of other people. The idea that migrants are incapable of making precisely these calculations is another strange consequence of insisting upon their vulnerability.

This is also 2026, not 1926. Migration no longer necessarily means disappearing from your family. A person can leave Guatemala, Pakistan, Ethiopia or Bangladesh and remain in daily visual contact with people at home. Grandma can see the apartment. Dad can complain about the haircut. Children can talk to a parent every day. Families can live with a video connection open for hours. None of this substitutes for physical presence, but it radically changes the human cost of distance compared with the world in which an emigrant boarded a ship and became a monthly letter.

I moved to Hawaii as a child in 1976. Calling grandparents on the mainland was an event because long-distance telephone calls cost real money. We wrote letters. We waited. The modern migrant has a magic portal in his pocket.

We sometimes talk about people from poorer countries as though they have arrived not merely from another nation but from another technological epoch. They have smartphones. They know what airplanes are. They have Google Maps. They have family group chats. They know America contains both opportunities and bastards.

They came anyway.

Three crayons for eight billion people

Another thing that bothers me about the language surrounding immigration is how much human complexity disappears when Americans begin talking about race.

I grew up understanding myself as Irish, English, German, Czech or Slovak depending on which piece of family history we were discussing, Hungarian and Austrian. Then I grew up in Hawaii, where an enormous amount of that European genealogical spreadsheet could be collapsed into one extremely efficient category: haole.

Hawaii understood racial compression before the mainland caught up.

Eventually Irish and German and Polish and Italian and Hungarian and Czech and a dozen other European identities became, for many American purposes, white. We now perform the same trick on much larger portions of humanity. Mexicans, Salvadorans, Guatemalans, Colombians, Venezuelans, Indians, Pakistanis, Bangladeshis, Afghans, Arabs and goodness knows who else can disappear into brown. Ethiopia alone contains extraordinary linguistic and ethnic diversity, but American racial language can take an entire continent of astonishingly different peoples and histories and reduce it to Black.

We have somehow developed a politics capable of finding seventeen identities inside one American undergraduate while finding three identities among eight billion human beings. Microscope inward, paint roller outward.

Then we become surprised when people assigned the same crayon don’t necessarily like one another.

A Mexican does not necessarily experience a Venezuelan as a member of some transnational political fraternity called Brown People. A Cuban in Miami and a Cuban in Havana may share language, ancestry, food and family history while possessing radically different political identities precisely because exile, revolution and property are part of that history. An expatriate Persian whose family left after the Iranian Revolution may have a relationship to the Islamic Republic completely unlike that of someone who remained. Ask either diaspora long enough about the homeland and you are not receiving the view from nowhere. You are receiving history from someone standing inside it.

Western Europeans have demonstrated for generations that racial difference is hardly necessary for anti-immigrant hostility. The English can become exercised about Poles and Romanians without consulting a color chart. Humans are perfectly capable of getting down to nationality, religion, language, class, region, village and which side of the mountain your grandmother came from.

Race can matter enormously. Racism exists. It also does not explain every conflict involving immigrants. Sometimes people resent newcomers because they are newcomers. Sometimes they fear competition. Sometimes housing is scarce. Sometimes cultures clash. Sometimes politicians exploit the issue. Sometimes people simply prefer people familiar to them. Humans were inventing out-groups long before anybody developed the American census categories.

Mexico’s recent experience makes the point especially interesting. As U.S. enforcement and asylum policies have made northward migration harder, Mexico has increasingly had to confront migrants not merely as people passing through but as people who may remain. That creates Mexican versions of questions Americans imagine are uniquely American: who may enter, who may stay, what newcomers are owed, how much migration communities can absorb, and what happens when yesterday’s transit population becomes tomorrow’s neighbor.

Once brown people are arguing with brown people about immigration, perhaps the three crayons were never enough.

The badassery of leaving

The people who fascinate me most are not wealthy expatriates who fly first class into another country and accidentally discover ten years later that their visa expired. I mean the people who go overland. The people who travel enormous distances because they have decided that whatever lies at the other end is worth the attempt.

There is a peculiar selection effect involved in migration that our language of vulnerability obscures. The person willing to leave home, cross borders, tolerate uncertainty, enter an unfamiliar labor market and risk failure is not a random sample of the population he left behind. Something selected him. Ambition, desperation, courage, recklessness, duty, curiosity, greed, love, stubbornness, economic calculation or some cocktail of all eight.

Sometimes the migration is a sacrifice. One strong son goes north because his labor is worth more there and the family can multiply the value of his earnings at home. Sometimes it is a daughter. Sometimes several relatives go. They live together, pool money, work opposite shifts, buy the first van, then the second, then perhaps a house or land somewhere. Jewish immigrants did versions of this. Irish immigrants did it. Italians did it. Salvadorans and Guatemalans do it. Indians, Pakistanis, Bangladeshis, Ethiopians and countless others do it. The nationalities change. The family mathematics is ancient.

And sometimes the result is precisely what happened with Hermano. The man who looks comparatively poor from the American side of the transaction takes the difference between what his labor can earn here and what those dollars can accomplish there and turns it into property. He buys the farm. He builds the compound. His family inhabits it. The construction worker renting in America becomes the landowner in Guatemala because the two lives are not competing stories. They are the same economic strategy viewed from opposite ends of the wire transfer.

I don’t see pathetic people when I look at that. I see adventurers.

Some of them have endured violence. Some have endured poverty. Some have inherited enough family trauma to keep three generations of American therapists employed. Some simply wanted more money. Some probably heard there was an opening and went for it. None of those motives requires me to shrink the person into an adult-sized baby so that I can demonstrate my compassion.

There is something almost insulting about insisting that courage only counts when the person succeeds. The failed migrant may have been every bit as brave as the successful one. The failed entrepreneur still opened the restaurant. The student who dropped out still enrolled. The actor who never became famous still got on the bus to Los Angeles.

Luck is real. Competition is real. Failure is real. So is taking the shot.

Ask him

Perhaps this is the simplest thing I want from the whole argument.

Before deciding that an immigrant has failed at the American Dream, ask him what he was trying to accomplish.

Maybe he wants the suburban house and citizenship and children at Princeton. Maybe he wants ten years of American wages and a farm in Guatemala. Maybe he wants to send enough money home for his sister to finish school. Maybe he wants an apartment, streaming services, coffee money and enough left over for a mildly embarrassing bag habit. That last one happens to be fairly close to my own definition of material sufficiency, which may explain why I am unimpressed when somebody tells me I should want a larger house.

Envy can turn any standard of living into deprivation because there is always somebody above you. The person earning $60,000 can compare himself with a billionaire if he wants to, but that is rather like a four-year-old announcing that he has failed athletically because an Olympic pentathlete is faster. The comparison is available. Its usefulness is another matter.

A recent immigrant may have a comparison set entirely different from the fifth-generation American who thinks a detached house in the neighborhood where his grandparents lived is the minimum acceptable evidence that the American Dream remains alive. Those expectations may change in the next generation. The immigrant’s children may want more. His grandchildren may eventually earn $160,000 and make TikToks explaining that life is impossible because the only house they consider acceptable costs $1.4 million. Congratulations. Assimilation is complete.

But don’t decide in advance what success is supposed to mean for Hermano. Ask Hermano.

He might surprise you.

Give them back their adulthood

Immigration policy is a legitimate argument. A country can have borders. A country can decide whom it admits, under what conditions, in what numbers, and what happens when someone violates those rules. People can favor dramatically more immigration or dramatically less. They can argue about asylum, labor markets, benefits, enforcement, deportation, assimilation and citizenship until everyone involved has died of old age.

None of that requires us to infantilize the people crossing the border.

In fact, I think the argument becomes more serious when we stop doing it. The migrant can be brave and opportunistic. He can be generous and selfish. He can support a family and drink too much. He can be exploited and exploit an opportunity himself. He can encounter racism and harbor prejudices of his own. He can make brilliant decisions and stupid ones. He can deserve compassion without becoming a creature defined by his need for it.

He can be an adult.

That is what I love about Hermano. He does not need to become my symbol of immigration, and I have no interest in polishing him into the Noble Immigrant. He is a funny little Guatemalan cowboy with an old Tundra, a family, a construction job, cigarettes, plans, a farm and family compound in Guatemala that exist because he bought and built them with the money he earned here, and an inexplicable determination to solve my romantic life.

Somewhere in the immense machinery of American social analysis, there is probably a category into which he can be placed and a professional vocabulary for describing all the forces acting upon him.

I prefer Hermano.

The World Bank can count the dollars he sends home. The Census Bureau can count whatever it counts. Immigration authorities can decide which box he belongs in. Economists can calculate his labor-market effects. Political activists can fight over what America owes him.

I am interested in something simpler.

Look at what the man has already done.

Then have enough respect for him to stop calling him one of the most vulnerable among us.

He crossed the world to take his shot.

At least give him the dignity of admitting that he took it.

My building has terrible elevators. It's one of the five hundred things that keeps the price down, and I've made peace with it, except peace isn't really the word, because peace implies I had to get somewhere first. What I actually have is a game. A private, free, endlessly repeatable game of three card monte, except the trick isn't finding the card, the trick is not caring which card it is.

Here's how it works. I walk up to the elevator. I press the button. I have no desire to go up. I have no desire to go down. I have no requirement to be anywhere, because I have a phone and the entire internet and all telecoms in my pocket, and there is no location on earth where I am not already exactly where I need to be. So I stand there. And the elevator comes. Immediately. Every time. If I'm alone it goes straight to eight, straight to one, no stops, like the building itself is relieved someone finally asked for nothing.

Then someone else shows up.

And they want it. Now. They're late, or their feet hurt, or they've decided with total conviction that the up button will somehow summon the elevator faster if they also hit down, just in case, just to cover their bases, just to have optionality against a machine that does not negotiate. And the elevator, sensing this, takes forever. Stops at every floor. Picks up ghosts. Takes so long that the person eventually gives up and heads for the stairs, defeated by eight flights they chose over four more minutes of wanting something in front of a box that wasn't listening.

I used to think I'd found a benevolent trickster. Not Descartes' evil genius, cooking up a universe to deceive me, but the opposite—some kindly force rewarding desirelessness with speed. Let go and the elevator comes. Let go, let God, and the doors open.

But that's not quite it, and I only recently caught the actual mechanism, which is better than the trickster story because it doesn't require a trickster at all.

I don't time it. I never have. No stopwatch, no glance at my watch, no mental tally of how long I've been standing there. Time is irrelevant to me everywhere in my life, and the elevator doesn't get to be the one exception. Which means there is no "slow" for me to have caught the elevator being. Slow requires a clock. Late requires an expectation of on-time. I have neither, so the elevator can't fail me, structurally, the same way you can't lose a game you never agreed to keep score on. It's not that desirelessness beats the clock. It's that there's no clock, so there was never anything to beat.

Call it Schrödinger's elevator. Until somebody wants it badly enough to force the question, the car isn't fast or slow, isn't on floor two or floor seven, it's just unresolved, humming along in every possible position at once behind a set of doors that haven't opened yet. The box doesn't have a speed until somebody's desire collapses it into one. Alone, I never do the collapsing, so the elevator lives permanently in superposition, which from the outside looks exactly like it always arriving instantly, because there was never a slower version of events for it to have been instead.

None of this is resignation in the feeble sense, the kind where you've simply run out of fight and calling it peace after the fact. It's intentional resignation, which is a completely different animal. I have full capacity to want the elevator right now, to be irritated, to hit the button four more times like that's ever done anything. I'm choosing not to, on purpose, every single time, which takes exactly as much doing as choosing to want something does. It's not the loss of agency. It's agency, spent deliberately on not-wanting instead of wanting, which is a stranger and harder thing to pull off than it sounds.

Alone, this whole practice is a null experiment. Nothing to prove, nothing to measure, no opponent. It only becomes visible—dramatic, even—when I'm no longer alone in the hallway. When someone else's wanting shows up next to my not-wanting, and suddenly there's a contrast to see. Which means the whole show was never really about me and the elevator. It was about what happens when my empty room meets somebody else's crowded one.

And that's the real lesson the elevator's been giving me for free every day: willfulness doesn't work, because you are never the only will in the hallway. Everyone standing there has their own agenda, their own tolerance, their own private weather system of frustration—late for work, sick of this building, why do I even pay for elevators, might as well take the stairs, an entire novel running in somebody's head while I've got nothing up there but fluff. You can be as willful as you want. You will still be negotiating, whether you admit it or not, with every other will in the box. The only thing that has ever actually moved anything, in an elevator or anywhere else, is collaborative will, not willful will. Getting eight people to a shared floor requires something closer to consent than command.

I live in this building. I pay rent. I have zero elevator entitlement, and I mean that literally, not as a spiritual posture. A guy in my building once got furious waiting, the kind of furious where he starts narrating his case out loud to nobody, and he said there should be six elevators in a building this size, this is unacceptable, and I told him I'd lived in a lot of places in Europe where everything is a walk-up, or there's one single elevator retrofitted into a hundred-year-old stairwell that only sometimes works, and that two-thirds if not three-quarters of Europe has no elevator at all. Nine out of ten of my friends in Germany lived in walk-ups with coal-fired ovens, and to heat the place you went down to the basement, shoveled coal into a tote, and hauled it back up every flight yourself. And the top floor, the one everyone here would call the penthouse, was the cheapest apartment in the building, because it was the most stairs. Nobody there is standing in a lobby demanding six elevators. The entitlement is the thing making people furious, not the wait.

So part of what I actually tell myself, standing there, is a small private scale of gratitude that has nothing to do with what I'm owed. Three elevators running is Thanksgiving. That's cornucopia. Two running is a perfectly good weekend day, nothing to complain about. One running and I'm grateful there are any elevators at all in an eight story building. If all three are down, I'm grateful I have legs and a stairwell and can still get to my own apartment, on the eighth floor, out of eight floors, under my own power. And if you want to go all the way to the actual bottom of the ladder, the real worst case, it's not a slow elevator. It's the building on fire with everybody in it. Measured against that, a slow elevator isn't a failure of any kind. It's a wild, undeserved success. There is no rung on that ladder where I end up resentful, or disappointed, or frustrated. I built the scale so that every possible outcome lands somewhere between good and better, on purpose, so there's nowhere left for disappointment to live.

That's the real trick behind the serenity prayer, and why it's stronger than people give it credit for. It doesn't grant you control over the elevator. It doesn't magically speed anything up. What it does is stop you from hitching your wagon to a tractor you don't have the keys for. An engineless tractor is honest about it. It's dead, it's not going anywhere, nobody's confused. But a tractor with keys you don't hold is worse, because it could move, it's just never going to move for you, and you'll stand there straining against an ignition that was never yours to turn, mistaking the possibility of motion for an invitation. The elevator was never yours to control. The prayer doesn't fix that. It just stops you from lying to yourself about whose keys those are, and once the lying stops, you're just a person standing in a hallway, waiting for a box, with nowhere else you'd rather be and no clock telling you otherwise.

Schrödinger never cared whether the cat was in a hurry. That was always the joke. The box doesn't know what you want. It just knows whether you're still asking.

Every elite university in this country already runs a shadow prep program for its own incoming class, and it isn't secret, it's just expensive enough that nobody has to call it what it is.

A kid from a family with money doesn't show up to freshman year cold. He's had a decade of small, compounding advantages nobody labels as remediation because nobody needs to: the $60,000-a-year prep school that teaches him how to write the essay admissions officers want, the private tutor three nights a week, the summer program that puts him on a foreign campus before he's even applied anywhere, the deferred year some universities quietly offer their own waitlisted, well-connected applicants, sending them off to spend a year at a partner university abroad while they wait for their real seat to open up. He comes back from that year having lived inside an unfamiliar academic culture, fluent in its rhythms, its vocabulary, its unspoken rules, before day one of his actual undergraduate career even starts. Nobody at that university calls this a deficiency program. It doesn't carry that stigma, because the kids receiving it were never going to be seen as deficient to begin with. It's just what getting ready looks like when you can afford to buy it.

Nothing about that instinct is wrong. Practice, immersion, a real runway before the real test starts, all of it works, and it's why the families who can afford it keep buying it generation after generation. The problem was never the method. The problem is that it's rationed by income, and everyone the method isn't rationed to gets told, in a hundred small ways, that showing up unprepared is a personal failing rather than the predictable result of never having had the thing money buys quietly for somebody else.

Britain already runs a public version of exactly this runway, and it isn't controversial there, because it isn't optional and it isn't means-tested. After compulsory schooling, before university, British students spend two years in what they call sixth form or college, doing intensive, specialized, exam-driven work in the subjects that actually matter for their intended field, arriving at university already functioning at a level of depth an American freshman typically won't reach until junior year. Nobody in that system experiences those two years as remedial. They experience them as the normal, expected, universal step between childhood and university, because everyone takes them, rich and poor alike, and a step nobody's exempt from can't function as a mark of deficiency.

That's the shape worth building here, minus the stigma this country would otherwise attach to it if it landed only on the kids who need it most: two additional years, call them the AP Years, sitting between high school and university, open to anyone, not sorted by race, not sorted by zip code, but by whoever didn't get what money already buys quietly for somebody else. That includes the kid from a rural county whose high school never offered a single AP course. It includes the kid who spent two years of a pandemic on a laptop and emerged behind in math and reading and is getting mocked for it online now, by people who've never once asked what actually happened to that generation of students. It includes first-generation kids of every background walking toward a campus that already assumes a cultural fluency nobody ever handed them. It is not a segregated program for teaching anyone how to be someone else. It is the same runway wealthy families buy for their own kids, funded publicly, offered out loud, to whoever actually needs the two years instead of whoever can afford to skip needing them.

The academic side is the easy part, and it's almost beside the point. The AP Years get front-loaded with the general-education requirements every degree already demands, math, writing, history, the sciences, cleared two years early, at Advanced Placement depth, so the actual four-year clock starts with nearly the whole degree still ahead as the student's chosen field. That's worth having. It is not why students wash out.

Students wash out because of what happens in the dining hall, not the lecture hall. Academic capability is rarely the thing that ends a college career at an elite institution. What ends it is a slower, quieter erosion, the daily experience of not knowing whether you belong in a room, sitting across from people whose entire lives have prepared them to move through that room unconsciously, without a second thought given to which fork or which reference or which unspoken rule just got broken. Psychologists call it belonging uncertainty, and the research on it is blunt: it costs capable students their degrees at a much higher rate than any exam does. Nobody drops out of Yale because calculus was too hard. People drop out because four years of feeling like a guest in someone else's house, evaluated and slightly foreign every single day, is exhausting in a way no amount of raw intelligence protects against.

And it runs both directions, which is the part nobody wants to say plainly. A kid who arrives having spent his whole life despising the rich, the snobbish, the effortlessly credentialed, for good and often earned reasons, is about to sit in a seminar next to the actual children of the people he's spent his adolescence resenting, sometimes literally the sons and daughters of oligarchs and heads of state. That contempt doesn't evaporate because he got in. It walks in the door with him, and if nobody ever helps him metabolize it, name it, decide what to do with it, it becomes exactly as capable of ending his university career as any professor's disdain could. Wealthy families don't just buy their children academic preparation. They buy years of casual, repeated proximity to people like that, so the shock of the encounter never has the chance to build into contempt or paralysis in the first place. Nobody's ever bought that same exposure, or that same working-through, for anyone else.

Here's the part that makes this urgent instead of merely nice to have. Some kids arrive having grown up somewhere that required a different set of reflexes to survive, a readiness to answer disrespect immediately and physically, because nobody in that environment was coming to intervene on their behalf if they didn't establish it themselves. That reflex isn't a character flaw. It's a rational tool, built for a place that actually demanded it. It is also the single fastest way to end a university career that exists. There is no retake for throwing a punch in a seminar room. The exam has a makeup date. The reflex that kept you safe at fifteen does not get a second chance at twenty, and an institution that hands someone a scholarship and a syllabus without ever once addressing that gap has handed him a loaded advantage and an unmarked minefield in the same envelope.

The same word can carry two entirely different weights depending on who's holding it, and almost nobody explains this before it matters. A rich kid dropping hard language in a seminar, or on a podcast, is usually just handling a toy, borrowing an edge that costs him nothing because nothing behind him has ever made that word a live commitment. For someone who came up somewhere that word could actually start something, where language was one of the tripwires that decided whether a night ended peacefully or didn't, the identical word isn't vocabulary, it's holstered. Neither person is wrong to have learned what they learned. But only one of them is walking into the room carrying something that can actually go off, and nobody's ever sat him down beforehand to tell him which pocket it's in.

Some things about the room itself have to be spelled out too, because they're invisible until you're already inside them. I studied a year in England in 1990, at a university nowhere near as posh as Oxford or Cambridge, and even there the resentment of wealth ran so hot that expensive cars kept getting keyed in the car park, "pennied," the locals called it, over and over, all year. The richest kids on campus had noticed, and they adapted: they drove up in beat-up Citroëns and secondhand Golfs instead of the cars their families could actually afford, wore grotty jeans and secondhand jumpers, and did their best impression of a rough London accent instead of the cut-glass one they'd actually grown up with. It worked, mostly, until enough drink got into them at a party, at which point the real accent came sliding back out every single time, unmistakably posh, without them seeming to notice it had ever left. Nothing about who they were had changed. They'd just learned, correctly, which version of themselves the room outside their door was willing to tolerate, and which version was safest kept for when the room emptied out. That's the actual skill on offer here. Not becoming someone else. Knowing, cold, before the first drink loosens anyone's guard, which version of yourself a given room can hold.

That is what the guaranteed one-on-one relationship is actually for, and it's the part of the program that matters more than any AP credit. Not a drop-in office hour. Not a triage counselor carrying four hundred names. One person, the same relationship the Oxford and Cambridge tutorial system has run for centuries, whose entire job for two years and then four more is making sure this specific student has somewhere to bring the belonging question, the contempt question, and the old reflex honestly, before any of the three quietly ends the whole thing from the inside. Not by telling him the reflex was wrong where he learned it. By making sure he knows, cold, before the first provocation ever arrives, that the room he's walking into runs on entirely different rules, offers an entirely different kind of protection, and will not read the old tool as strength.

There's a piece of that work that has nothing to do with belonging or reflexes and everything to do with plain isolation. Some universities grade on a curve, which means every point a classmate earns is a point taken directly from the pool everyone else is competing for, and the person sitting next to you in the seminar isn't a colleague, structurally, he's a rival, whether either of you wants it that way or not. A student who lands entirely inside that competitive pool for his social life too, whose only people are the same people he's fighting for grades against, is isolated in a way that has nothing to do with money or culture and everything to do with architecture. The fix isn't complicated, and it's exactly the kind of thing a dedicated advisor can make sure happens early, before the isolation sets in: get him rooted in a group that isn't also his competition, a floor, a dorm, an activity, anything anchored outside the graded pool, so there's at least one place in his week that doesn't run on zero-sum math. That's not resilience training. It's making sure nobody has to build resilience alone, against a structure explicitly designed to pit him against the only people around him.

Right now universities take full credit the day the acceptance letter goes out and treat everything after that, the academic gap and the cultural one alike, as the student's own problem to solve alone. A university that admits someone owes that person more than the letter. It owes the follow-through, all the way to a degree actually in hand, not an acceptance letter waved around as though the hard part were already finished.

That's the whole idea. Not a favor. Not charity. Not a program to make anyone more palatable to anyone else. Just the ordinary, unremarkable head start that has quietly belonged to whoever could pay for it, built instead as the standard, guaranteed inheritance of anyone this country asks to walk into a room it never actually prepared them for.

There are not enough people to go around, and every institution that wants a visible symbol of progress right now is drawing from the same thin, slow-growing pool.

Cambridge has seventy-five Black academics out of sixty-one hundred. The entire United Kingdom has roughly three hundred and fifty thousand Black residents out of sixty-eight million. America's Black population runs well over thirteen percent, but the specific credential most elite institutions are actually hiring for, the terminal degree, the endowed chair, the board-ready decade of executive experience, draws from a far smaller slice of that thirteen percent, for reasons that trace straight back to the historical resourcing gaps the entire push was supposed to close in the first place. Every university that wants a headline hire, every corporate board filling an ESG requirement, every media outlet that wants an expert to put on air, every awards committee assembling a program, is reaching into the same room. A PhD takes years to produce. A tenured chair takes longer. The demand for visible representation scaled up in a decade. The supply moves on a generational clock. That gap is not a scandal. It is arithmetic, and arithmetic does not care how good anyone's intentions are.

What arithmetic produces, reliably, is overexposure. The same handful of names get asked onto every board, into every keynote, onto every shortlist, because they are the names everyone already knows, and being known becomes its own qualification, detached from whatever earned the recognition in the first place. Corporate governance researchers have a term for this when it happens to directors: overboarding, the practice of asking the same small set of visible minority executives to sit on far more boards simultaneously than their white peers ever are. Nobody has to coordinate for this to happen. A hundred honest search committees, each reaching for the one list everybody already has memorized, will produce the identical outcome with zero conspiracy anywhere in the chain.

None of which makes the person selected a mark. A school mascot tries out. He knows exactly what costume he's putting on and what the job requires before he ever agrees to wear it. That's a real distinction, and it matters, but it isn't the only honest way to be chosen for a symbolic role and still come out ahead. Leonard Bernstein invited the Black Panthers into his Park Avenue apartment in 1970, and Tom Wolfe wrote the essay that gave the whole phenomenon its name: radical chic, the fashionable set displaying proximity to danger and cause the way they'd display a new sculpture. Nobody sitting in that room was naive about what kind of evening it was. They were there anyway, because a room full of money and press attention is still a room full of money and press attention, and a savvy operator takes the opportunity the fashionability provides and banks whatever it's worth before the fashion moves on, same as anyone would.

King Kalākaua ran the same calculation on a grander stage. His 1881 world tour made him the first reigning monarch in history to circle the globe, and heads of state treated him as exactly what he was: Japan, Austria-Hungary, Britain, the United States, all received him with full military honors and formal audiences, and he became the first foreign leader ever hosted at a White House state dinner. That was one track. The other track, running underneath it in the same newspapers covering the same visit, kept marveling that a Pacific monarch could behave with such sophistication, as though competence in a Hawaiian king were the anomaly requiring explanation rather than the baseline. Both tracks were true at once. The state banquet was real. So was the condescension in the morning edition. Kalākaua wasn't tricked into the tour, he arranged it himself, for his own explicit diplomatic purpose, and he almost certainly understood exactly which parts of his reception were respect and which parts were spectacle, and used both anyway to advance what he came for.

Claudine Gay's appointment as Harvard's first Black president in 2023 ran the identical two-track pattern in modern dress. The institution announced it as history, the press covered it as history, and the arrangement functioned, structurally, as a wager the moment the ink dried: an institution putting a human being's tenure on the line as proof of its own progress before that tenure had run long enough to prove anything at all. When plagiarism allegations surfaced within months and she resigned within weeks of the scrutiny going mainstream, the same press that had run the coronation ran the collapse, at the same volume, on the same front page. By now the entire episode has largely receded from conversation, crowded out by newer stories competing for the same finite attention the original hire was built to capture. That fading is not incidental. It's the mechanism completing itself: the announcement was never really about the person, it was about the institution's need for a headline, and once the headline stopped generating attention in either direction, the institution had already collected everything it needed from the arrangement.

Here is the part that outlasts the wager, win or lose: the credential itself does not care how loudly anyone announced it, and it does not un-happen when the announcement sours. A degree earned is a degree held, permanently, regardless of the press release that came with it or the backlash that followed it. The court that closed the door on race-conscious admissions in 2023 could not reach backward and revoke a single diploma already granted. The scrutiny that follows any high-profile collapse can chase a reputation, but it cannot touch the actual work sitting underneath a real credential, because the diploma never asks how the political weather felt the year it was issued.

What the asterisk actually costs isn't the person it lands on. Whoever the asterisk targets either has the credential or doesn't, and the asterisk changes neither fact. What it costs is everyone standing nearby who shares a surface trait with whoever cracked, the professor down the hall who built a career the slow, unglamorous way with no press release ever written about her, the colleague who never asked to be collateral in somebody else's institutional wager and now answers for it anyway, in the raised eyebrow, in the assumption that has to be argued down before the actual work even gets a hearing. That's the tax nobody voted to pay, charged to people who were never in the room when the bet got placed.

And there's a second cost, quieter, running for a generation instead of a news cycle. Every family that watches one of its own walk through a door like this starts reorganizing what winning looks like around it. The Boulé Railroad worked precisely because a Harvard degree, once inside a bloodline, changes what the next generation aims for without anyone having to say so out loud. That's a real gain. It's also a real narrowing, because a household that starts treating the acceptance letter as the family's new trophy has quietly demoted whatever used to sit in that chair before, the ball field, the stage, the workshop, to something lesser. Nobody announces the demotion. It just happens, one dinner-table story of pride at a time, until the kid who wanted to be a running back instead of a lawyer is fighting his own family's new hierarchy of what counts as making it, not just the world's.

The mascot fades. The wager gets forgotten within a news cycle or two, same as it always does. The credential and the family's new definition of a life worth having are the only two things left standing when the applause moves on to the next headline, and only one of them was ever actually about the person the applause was pointed at.

Batman has fought a man obsessed with riddles, a man obsessed with fear, a man who wants to watch the city burn just to prove it can. He has never fought the villain who does the most damage of all: the Meddler.

The Meddler doesn't want to watch the world burn. He wants to fix it. That's what makes him worse than the ones who don't. The Joker knows he's the villain. Two-Face is at war with himself and knows it. Bane wants to break the world on purpose and says so. The Meddler is the only one incapable of ever finding out what he is, because every intervention reads to him as rescue, every casualty reads as an unfortunate cost of doing good, and every explosion reads as proof the job needed doing even more urgently than he thought. You cannot shame him out of it. You cannot negotiate with him. There is no crack to work with, because his self-image and his body count are never filed in the same place.

Nitroglycerin was invented to make mining safer than gunpowder. That's not incidental to the story, it's the whole story: the product was a genuine humanitarian improvement, adopted in good faith, and it turned out to be so unstable that it killed the people carrying it precisely because of how sincerely it was meant. Nobody straps a crate of it to a wagon expecting the explosion. They do it because they believe, correctly, that they have made things safer than the alternative. Alfred Nobel spent years trying to stabilize the stuff after watching it kill his own brother, and what he built instead was dynamite—a controlled version of the same force, with a fuse and a limit.

The Meddler never builds the fuse. He doesn't think he needs one. He hands out the nitroglycerin labeled medicine, because in his mind it is medicine, and when it goes off in someone else's hands three countries away, the file gets marked unforeseeable rather than should have been foreseen. Every time.

This is not a story about any one party. Every faction with enough power to act on the world eventually grows its own Meddler, and every Meddler tells himself the identical story: I am the one who stayed and did something while everyone else looked away. The DEI consultant who cannot leave an institution's hiring practices alone. The nation-builder who cannot leave a country's government alone. The NGO that cannot leave a village's water supply alone without also rewriting its politics. The interventionist who cannot watch a civil war on the news without concluding that the moral course is to join it. Different departments, same employee.

What makes the Meddler durable, and dangerous, is that his ethic is not selfish. That's the part people miss when they try to satirize him as a grifter or a hypocrite. He is not doing it for the money and he is not doing it for the applause, or not only. He has convinced himself, often correctly in the particulars, that he is discharging an obligation larger than himself: to repair the world, to leave no injustice unaddressed, to treat every suffering stranger as fully his responsibility whether or not that stranger asked for the help. Taken as a private discipline, applied to one's own house, one's own street, one's own community, that instinct is close to the best thing in a person. Scaled up to the size of a government, a media apparatus, or an empire, the same instinct becomes a machine that cannot stop intervening, because stopping would mean admitting that some suffering in the world is simply not his to fix. The Meddler cannot admit that. It's not in the architecture.

And so the wreckage accumulates in a place he never has to look. The coup that was supposed to install democracy and installed a stronger dictator instead. The aid program that displaced the local economy it was sent to save. The moral crusade at home that radicalized the very people it set out to convert, because being told your grievances don't count, often enough, in enough op-eds, does not make a person neutral. It makes him recruitable by whoever shows up next promising to take his side. Every one of these gets filed as an unintended consequence, a tragic complication, a lesson for next time. Never as evidence. The Meddler updates his tactics constantly. He never once updates his self-conception, because the one conclusion his entire operating system cannot process is: maybe the job was never mine to do.

This is why the Meddler is harder to write, and harder to beat, than any cape villain who wants to be caught. You cannot out-argue a man who agrees with all your facts and simply concludes that the facts prove he needs to intervene more, faster, and with fewer people asking permission. He is not lying to you. He is not lying to himself either, not in any way he could locate and correct. He looks in the mirror every night and sees exactly what he set out to see: the one who stayed, the one who cared, the one who acted while the rest of the world made excuses. He never sees the terrarium he flooded trying to water it. He only ever sees the terrarium he meant to save.

A guy on TikTok says you can't feed a family of four on $300 a month unless you're willing to eat nothing but grains and beans. I want to do the math before I do the arguing, because the math changes what the argument actually is.

A burrito costs $2. And not the ascetic version, either—beans, rice, avocado, tomatoes, onions, lettuce, the whole thing, no meat, no fish, no chicken thrown in, protein coming entirely from the beans and rice pairing. That's a real, satisfying meal, not a grim ration, and it's already carrying the extra naughty fee of an avocado tax, the guac upcharge every Chipotle regular knows by heart. It also happens to be the better choice for the planet, since skipping the meat is the single biggest lever anyone has for cutting their own food footprint. Add real protein back in, ground beef or chicken, and the price climbs to roughly $3 to $3.25 a burrito once you account for raw cost and cooked yield. I'll run both numbers.

At $2 a burrito: four people, one each, one meal: $8. Three meals a day: $24. Thirty-one days: $744.

At $3.25 a burrito, with meat added: four people, one each, one meal: $13. Three meals a day: $39. Thirty-one days: $1,209.

Flip the math around and cook for one instead of four, and the same logic runs in the opposite direction. Budget $20 a day for one person, shopping and cooking at home instead of eating out, and that's $600 a month. At that budget, a single person can eat an actual steak-and-shrimp dinner, real cuts, not a garnish spoonful, every single night and still have room for breakfast and lunch. Surf and turf, home-cooked, daily, for a month, on less than what the USDA says a family of four needs just to survive on beans and rice.

Even the working professional who thinks they're eating responsibly is inside the same trap, just at a higher price point. I worked on K Street in the mid-2000s, and the daily cap for a sandwich-chips-drink lunch at the office was around $6 to $8. Run that through twenty years of inflation and it lands at roughly $10 to $13 in today's money, which is almost exactly what a bowl costs now. Chipotle, Sweetgreen, Cava, Chopt: the whole genre has a real name now, "slop bowl," social media slang that made it into the New York Times as shorthand for the assembly-line pile of rice, protein, and sauce that gets called a salad or a burrito bowl once it's fifteen dollars and comes in a compostable container. A Harvard food sociologist put the actual math on it plainly: once a bowl runs $60 to $70 for a family of four, you're past the point where the "healthy, responsible" choice and the "just cook rice and broccoli at home" choice have anything to do with each other. That's the K Street junior-executive baseline, the same self-image of eating well that a family on food stamps gets accused of not affording, except the executive's version costs three or four times what the family's actually would, and nobody calls it a crisis. They call it lunch.

Now put a family of four on the other end of the spectrum: never cooking, never buying groceries, living entirely on restaurants, delivery, and the prepared-food aisle. McDonald's, four Big Macs: about $24 a meal. Golden Corral dinner buffet, the same family: $50 to $70 a night. Applebee's $9.99 combo deal, four of them: $40. A large delivery pizza or two, easily $30 to $40 with tip. And that's before DoorDash: menu prices alone run 10 to 20 percent above in-store, and once the delivery fee, service fee, and tip stack on top, the all-in cost of a DoorDash order runs 50 to 75 percent higher than the identical meal picked up yourself. A $40 in-restaurant dinner becomes a $65 to $70 delivery order without a single ingredient changing. Even the low end of that lifestyle, one fast-food run and one delivery order a day, blows past a thousand dollars a month before the family has eaten a single vegetable that wasn't a garnish.

That's the actual Delta: $744 to make burritos for a month, or somewhere north of $1,500 to $2,000 to never once touch a stove. Nobody in America is comparing $300 to the burrito math. They're comparing $300 to the DoorDash math and calling the difference a moral failing of the country instead of the math it actually is. Dignity, in the version everyone's actually arguing for, has to be pumpkin spiced. It has to arrive in a bag with a receipt stapled to it. A home-cooked burrito, no matter how good, doesn't count, because somewhere along the way, convenience got rebranded as the floor instead of the extra.

Run it the other way and the government agrees with me. The USDA's own Thrifty Food Plan, the lowest of its four official tiers, the one that assumes every meal is cooked from scratch with zero takeout and zero convenience food, puts a real family of four at close to $1,000 a month. Three completely unrelated ways of pricing a month of food, one built out of $2 vegetarian burritos, one built out of $3.25 burritos with real meat in them, and one built out of federal nutrition data, all land in the same neighborhood: $700 to $1,200. None of them lands anywhere near $300.

So the guy in the video is right. $300 a month, three meals a day, four people, doesn't work. The arithmetic doesn't close, and no amount of budgeting skill gets $744 down to $300.

Here's where it gets interesting: once you settle the actual number, watch what the argument does next. Nobody doubles down on $300. Nobody says the real number is $500 or $600, splitting some honest difference. The objection moves. Suddenly it's not about survival anymore, it's about what survival should include. What about snacks. What about treats. What about the birthday cake. One TikTok comment I saw argued every family, no matter what, deserves a fresh Cobb salad. Not canned tuna, not a can of beans dressed up with an onion. A cold-prepped, restaurant-tier Cobb salad, every day, as the baseline.

Nobody in this argument thinks the family goes hungry on rice, beans, and eggs. They think it isn't fun enough. That's not a poverty argument. That's a deliciousness argument, wearing a poverty argument's coat because the poverty argument gets more sympathy.

Go back eighty years and the actual floor for what an American family ate looks nothing like $744 worth of ground beef with nothing added. WWII rationing put meat behind a ceiling so low that meatloaf, oatmeal and breadcrumbs stretching a small ration of ground beef into a family-sized meal, became the defining dish of the era, taught in government pamphlets and home-economics classes as a patriotic act. Sliced bread itself got rationed in January 1943. Some Americans ate horse meat because beef ran out. That's the actual historical floor. Nobody today, at any income level, needs to cut their burger with oatmeal to make it to Friday. The $744-a-month floor isn't the bottom. It's already a staggering upgrade from the bottom, and it's the bottom we're currently calling a crisis.

The number was never the disagreement. Everybody, eventually, agrees the number is somewhere around $700 to $1,000, not $300. The real disagreement, the one nobody wants to say out loud, is whether that $700 has to come with a Cobb salad attached to count as a solved problem. And that fight, the fight over garnish instead of survival, is the kind of fight only a country that has already solved hunger gets to have.

This is a Saturday morning ramble. No forced center, no single throughline dragging every paragraph back to one tidy thesis by main force. Some of what's below is about tools decaying, some of it is about false permanence, some of it is about hyperfocus, and I'm not going to force-feed any of it into a shape it doesn't want. Force-feeding is illegal when it's done to a hunger striker. It shouldn't be legal to do it to an essay either.

In Lee Child's eleventh Reacher novel, Bad Luck and Trouble, a defense contractor called New Age builds a surface-to-air missile system codenamed Little Wing, engineered to defeat every anti-missile system in existence. Someone asks how long that supremacy lasts. Two years, comes the answer. Not because the engineering is bad. Because two years is roughly how long it takes anyone else to look at what you built and figure out how to turn it back on you.

That's not a rule about missiles. That's a rule about everything.

I was listening to Bill Maher's Overtime, Maher going back and forth with Dana Bash and Jeremy Peters about the Electoral College. Everyone in that conversation is holding some advantage they've mistaken for permanent. It isn't. None of them ever are.

The baffles hold the fifty states apart, and that's the whole point

Fifty states voting means fifty baffles. Drop the baffles and the sloshing begins—one big tank instead of fifty separated compartments, and whichever mass of liquid is heaviest at the moment throws the whole vessel around. That's the Electoral College. It's not a bonus for small states. It's the thing keeping the country from slamming from one side to the other every time turnout shifts by a few points.

The case for the Electoral College is usually framed as protection for the small states: Wyoming and Montana get relevance they wouldn't have in a straight national count, and without that, a handful of dense population centers could steamroll everyone else on every question, every time. That's real, and it's not a new argument. It's Madison's argument.

But the same shield runs in the other direction. The Electoral College is also what stops a "silent majority," a rural or populist coalition large enough to swamp any national popular vote, from doing exactly the same thing to blue states and cities. Take the mechanism away entirely, run the country on raw popular vote, and you haven't guaranteed the coastal, urban vision a permanent majority. You've just unlocked the same weapon for whoever manages to out-mobilize the other side next, with no federalist floor left underneath anyone.

This is the part of the argument that keeps getting skipped: 2024 already answered the question. Donald Trump won the national popular vote, not just the Electoral College. The people arguing that killing the Electoral College guarantees a permanent progressive future are making a bet that already lost once, within the last two years, using the exact voting behavior they're trying to permanently disable a check on.

Treat America honestly and it looks less like one country than a European Union with fifty member states instead of Germany, France, and Italy—Wyoming standing in for Malta, California standing in for Germany, a union that's been slowly accreting new members since the thirteen colonies first signed on. Nobody in the EU assumes Germany's economic weight settles every argument permanently. Nobody should assume California's population does either.

Sovereignty is a tool people pick up when it's useful and put down when it isn't

Watch how selectively the sovereignty argument gets deployed. Sanctuary cities and sanctuary states insist on their right to set their own immigration enforcement posture, independent of federal policy, and that's a states'-rights argument if you've ever heard one. The same coalition making that argument, in the same year, generally wants the opposite everywhere else: one national standard on abortion, one national standard on gun policy, one national floor on labor law. Federalism when it protects you, nationalism when it doesn't.

Republicans have the mirror problem. States' rights has been the load-bearing conservative argument for the better part of a century, and it's the same party currently comfortable with an expanded federal executive, ICE operating with wider latitude, National Guard deployments into cities that didn't ask for them. Neither side actually holds a fixed constitutional philosophy here. Both sides hold a convenience philosophy, and they reach for whichever version of federalism is closest to hand in the moment, the same way Jackie Chan grabs a ladder, a bar stool, a noodle cart, whatever's within arm's reach in the scene. Jackie Chan doesn't win fights because he owns a weapon. He wins because he never gets attached to one.

The coalition itself is rented, not owned

Nobody actually owns the populist bloc that elected Trump twice. That energy is grievance-driven, not doctrinal: it swung from "no more Obama-Clinton" to "no more Biden-Harris-woke," and there's no guarantee it stops there. The same bloc that voted isolationist in the campaign and found itself watching an American aircraft carrier group involved near Iran might just as easily swing toward a populist candidate on the left the next time the establishment feels equally distant, whoever that candidate turns out to be. A coalition built entirely on being against something doesn't have brand loyalty. It has a target, and the target rotates.

Rhetorical weapons decay exactly like physical ones

"Everything is fascism" worked as an alarm the first several dozen times it got said with real conviction. It stopped working roughly the same way any siren stops working once it's on constantly: not because the underlying claim became less true in every instance, but because repetition trained the audience to stop hearing it as an alarm at all.

Black fatigue is the sharpest version of this. Mary-Frances Winters coined the term in 2020 to describe the accumulated toll of navigating a society built around constant, exhausting code-switching and the low hum of systemic racism. That's the original coinage. It's now more commonly encountered online meaning almost the opposite: white fatigue with hearing about race at all. The term got released into public discourse and immediately captured by the group it wasn't written to describe, run in reverse. Nobody controls a phrase once it's out. Little Wing doesn't stay yours either.

Look at the horizon, not the ditch

Every motorcycle instructor tells you the same thing: you go where you look. Stare at the pothole and you steer straight into it. Rowing coaches say the same thing about keeping your head in the boat, because the moment you turn to look around, you tilt the shell and throw off everyone else's stroke. Sailors getting seasick are told to find the horizon, not go below deck and stare at the floor. Climbers are told never to look down. The eyes lead the body, every time, in every discipline that depends on balance.

Democrats have spent the current cycle with their eyes fixed on a set of cultural fights treated as existential and non-negotiable, and polling on the actual electorate doesn't back up the stakes being assigned to them. Pull the aggregate public back far enough and immigration enforcement, crime, and cost of living dominate what people say they care about, while a long list of specific cultural positions, academic language around gender, abolitionist framing on policing, maximalist rhetoric on trans-specific policy fights, poll as niche concerns carried by a small, intensely committed slice of the coalition. Hyperfocusing on the pothole steers the whole party into it.

This isn't new, and it isn't one-sided. Guns, God, and abortion ran the Republican Party's non-negotiable list for fifty years, from 1970 forward. Two of the three effectively resolved in the party's favor: gun rights are stronger and more culturally entrenched now than at any point in that fifty-year stretch, and Dobbs handed the pro-life movement the outcome it spent five decades organizing around. The one item still sitting as unresolved, God, isn't even load-bearing for the party's current populist wing the way it was for the Moral Majority generation. The GOP spent half a century staring at three potholes and eventually drove straight over two of them. Whatever the party stares at next determines where it goes next, exactly the same as it always has.

Every cable panel and every strategist on the current media circuit keeps handing Democrats messaging advice, as if the problem is the wrapping paper. It isn't. A party carrying an entire portfolio of non-negotiable positions, each one held by a true-believer faction unwilling to let it go, doesn't have a communications problem. It has a hoarding problem. What that party needs isn't another consultant with a better slogan. It needs a Marie Kondo, someone willing to hold up each item in the pile and ask whether it still earns its place, and to throw out the ones that don't, regardless of how attached the true believer standing next to it happens to be.

There's an irony sitting underneath the hoarding, too. Solomon's whole test was offering to split the baby to find out which woman actually loved it. A hoarder never gets to that test, because every single item in the pile gets treated as somebody's baby, non-negotiable by definition, nothing ever offered up to be weighed against anything else. That's a strange place for a nominally pro-choice coalition to end up: unable to sacrifice a single position, on anything, ever, as though every plank in the platform were a child nobody's willing to see divided.

There's a name for what's driving the hoarding, and it isn't flattering: ideological rigidity, sometimes called purism, where a coalition prioritizes moral or policy purity over the broad appeal it would need to actually win. Political scientists point to a specific mechanic behind it: politicians fear losing their own primary to someone further to their flank more than they fear losing the general election to the other party. That reorders the whole incentive structure. Staying pure enough to survive the primary becomes the actual job, and winning over the persuadable middle becomes a secondary concern, sometimes not a concern at all.

Hollywood ran the same experiment first and got the same result. Marvel, DC, the recent Star Trek and Star Wars entries, all took audience pushback as a marketing problem: message it better, cut a different trailer, put out an op-ed about why the criticism is bad-faith. Nobody at any of those studios seriously entertained the idea that the actual product wasn't something the audience wanted to buy. There's an old sales axiom for why this keeps failing: it's a hundred times easier to sell someone something they already want than to convince them to want something they don't, which is why the salesman who moves bespoke, desirable products drives a nicer car than the one grinding out commodity widgets. Selling what nobody's asking for isn't a messaging job. It's a full counterinsurgency, treating every voter like a mountain villager who needs years of hearts-and-minds work before he'll trust you, when the entire operation could have been skipped by building something people wanted to begin with.

Blue-topia is 1984's slogan wearing a blue rosette

Some of what I heard on Maher's show sounded less like a policy argument than an article of faith: kill the Electoral College, and what follows is a kind of permanent progressive peace, since history only ever bends one direction. That's the same structure as Orwell's "War Is Peace." The thing being proposed as the guarantee of permanent consensus is precisely the thing that currently forces candidates to build broad, cross-regional coalitions instead of running up the score with an already-committed base. Remove the guardrail and you don't get less division. You get a system where fifty point one percent governs the other forty nine point nine with no structural incentive left to ever persuade them again. That isn't a blue utopia. That's a permanent-war machine.

Nobody gets to keep the weapon

Little Wing's two-year clock isn't really about missiles. It's about the fact that any advantage—electoral, rhetorical, coalitional, tactical—exists on loan from the moment your opponent starts studying it. The false sense of security isn't believing you have an edge. It's forgetting that the other side is watching the same clock, and that the two years, or the two elections, or the two viral cycles, run out exactly as fast for you as they ever did for them.

I get technical itches.

They are not business requirements. They do not emerge from a needs assessment, a productivity study, or a careful comparison of available products. Nobody asks me to solve them. In many cases, no normal person would recognize the underlying condition as a problem.

The itch arrives first. The use case has to catch up later.

A while ago, I bought a cheap purple Meshtastic device from somebody on Etsy. I wasn’t a prepper, I wasn’t a ham radio operator, and I didn’t have anyone nearby I particularly needed to contact over LoRa. I just wanted a tiny radio node. I put it in my eighth-floor apartment window, gave it a clear view across Arlington, named it Abracadabra, and left it running. Eventually it stopped being a gadget and became infrastructure: a permanent router participating quietly in a local mesh whether I was looking at it or not.

That became Leaving a Purple Meshtastic Node in the Window, an entire essay about how infrastructure begins when you stop checking whether a thing is entertaining and simply allow it to remain present.

Then there are my Tachikoma ThinkPads. I own a Lenovo W520 and an unnecessary number of X220s, all from the brief historical window when ThinkPad keyboards were still correct. They all run Linux Mint. Different machines live in different bags, but I want whichever one I open to contain the same current working mind.

So I built a little Syncthing network around another DigitalOcean Droplet. The Droplet holds the canonical copy of my Documents directory. Each ThinkPad wakes up, synchronizes with the cloud brain, becomes the current body for a while, then returns its new memories before shutting down. The next ThinkPad wakes up knowing everything the last one knew.

That one became My Tachikoma ThinkPads: A Self-Hosted Sync Network for One Human and Many Terminals.

Dropbox would have been easier. One modern laptop would have been easier. None of that is the point.

The point is the itch.

This one started with PINE

My first computer was an IBM PC 5150 with two 5.25-inch floppy drives, no hard drive, and a green monochrome monitor. I got it when I was thirteen, in 1983. My father had an Osborne 1. I used WordStar. Later, WordPerfect felt like the name of a product and a statement of fact. A dot-matrix printer chattered on fanfold paper nearby.

By the early 1990s, email meant terminals, university systems, command lines, and PINE. Email was somewhere you went to read messages. It was not a full-screen behavioral environment with animated sidebars, suggested replies, chat bubbles, calendars, documents, artificial intelligence, twelve account switchers, and a red badge announcing that civilization had failed because 14,382 messages remained unread.

I still spend most of my working life in text. I write. I edit. I answer people. I move between terminals. Most of my email replies do not require a browser rendering engine capable of playing 4K video while simultaneously consuming the electrical output of a small hydroelectric dam.

I began fantasizing about a dedicated computer that existed only to hold my email sessions. Not a physical computer, exactly. More like a room on the internet.

From any of my ThinkPads, I would open a Linux Mint terminal and SSH into that room. Inside, Alpine, the modern descendant of PINE, would already be running. The session would persist even if the laptop closed or the Wi-Fi disappeared. I could enter through another ThinkPad later and find everything exactly where I left it.

Eventually I would have three Alpine sessions:

• chris@gerriscorp.com

• my Stainless Communications account

• my Meritus Media account

Each mailbox would live in its own tmux window. I could move among them without logging in and out of web interfaces, hunting for tabs, or wondering which Google profile currently owned the browser. Three identities, three text cockpits, one persistent server.

It would use almost no resources. It would be quick. It would be quiet. It would be mine.

This is an absurd amount of infrastructure to avoid opening Gmail.

I wanted it desperately.

Building pineBox

The project had to be called pineBox, always with camel caps when referring to the idea. The actual Linux hostname is lowercase pinebox, because hostnames are not where one expresses typographic individuality.

A pine box is also a coffin, which made the name perfect. PINE is dead. Alpine is fine. Everything old would live again inside pineBox.

I created the smallest useful DigitalOcean Droplet I could buy:

• Debian 13

• one shared virtual CPU

• 512 MB of RAM

• 10 GB of SSD storage

• $4 per month

• a public IPv4 address in New York

This is not a powerful computer by modern standards. It is also ludicrously more powerful than any computer on which I used email in 1993.

I added a one-gigabyte swap file because 512 MB is charming until Debian briefly remembers that it was born in the twenty-first century. I created my own chris account, gave it carefully limited administrative privileges, installed my SSH key, disabled root login, and disabled password authentication. I installed UFW and configured the firewall to deny everything coming in except SSH. Automated security updates are enabled.

There is no website on pineBox. There is no database. There are no public users, forms, containers, dashboards, mailing lists, or applications. It is one small locked room containing one human’s email client.

On my W520, named Menta, I added an SSH configuration entry so I do not have to remember the server’s numerical address. Now I type:

ssh pinebox

That is the door.

Inside, I installed Alpine and tmux. Tmux is what allows a terminal session to survive the terminal itself. I can disconnect, close the laptop, turn it off, open another ThinkPad three days later, reconnect, and reattach to the same running session. Alpine never knows I left.

That persistence is the real itch. The laptops become interchangeable terminals. The state lives somewhere else.

The architecture is almost offensively simple:

ThinkPad X220 ─┐
ThinkPad X220 ─┤
ThinkPad W520 ─┼── SSH ──> pineBox ──> tmux ──> Alpine
ThinkPad X220 ─┤
ThinkPad X220 ─┘

Only one body needs to be awake. The cockpit remains.

Then Google required the future

Reading Gmail from a 1990s-style terminal client in 2026 involves an amusing amount of modern identity machinery.

Google no longer permits an email client to stroll in with a username and ordinary password. Correctly, to be fair. My Google Workspace account requires OAuth 2.0. Alpine 2.26 supports XOAUTH2, but that meant creating a Google Cloud project, configuring the Google Auth Platform, defining the audience, creating a desktop OAuth client, supplying Alpine with the client ID and secret, and completing Google’s browser-based authorization flow.

The final result is wonderfully anachronistic. A text-mode email client running on a tiny Debian server launches an OAuth authorization process through Google’s global identity infrastructure, after which it returns to looking almost exactly like PINE did when I was in college.

Once authorized, the inbox appeared.

There it was: live Gmail inside Alpine. Plain text. Sender, subject, date, size. No promotional cards. No hovering interface. No algorithm trying to be helpful. Just email behaving like email.

The first itch was gone immediately.

I could SSH into pineBox from Menta, enter the persistent tmux session, and read my real Google Workspace inbox in Alpine. If I disconnected and returned later, the green tmux status bar and my mailbox were still there waiting.

This was objectively unnecessary and spiritually exact.

Reading worked. Sending waited at zero percent.

The first outbound test message sat at:

Sending mail | 0%

Nothing happened.

This looked at first like an Alpine or OAuth problem, but the diagnosis was simpler. DigitalOcean blocks outbound SMTP traffic on ports 25, 465, and 587 by default. This is an understandable anti-abuse policy. Tiny, instantly deployable cloud servers with fresh public IP addresses are enormously useful to legitimate customers and equally attractive to people who would like to send several million exciting investment opportunities before breakfast.

DigitalOcean already hosts my excellent Syncbox, the canonical brain behind my Tachikoma ThinkPads, and it has been excellent for that job. I am not angry at DigitalOcean for treating outbound SMTP carefully. The company has to protect its network and address reputation from people whose technical itch is “crime.”

My own use case is almost comically harmless. I am not running a mail server. I am not sending newsletters or transactional messages. There is no automation. I am one person writing ordinary replies through an authenticated Google Workspace account, probably a handful per day.

So I opened a support ticket and explained the entire strange little project honestly. DigitalOcean asked for more context before forwarding the request to its Security team. I replied that pineBox is a private, single-user email workstation running Alpine inside tmux, reachable only by SSH, with no public application or mail daemon. Gmail requires the standard authenticated submission ports and does not offer an alternate port such as 2525.

I even confessed the actual reason:

In short, this is one person using a terminal email client for normal correspondence, partly because it is fast and partly because I miss how pleasantly simple email was in 1993.

That request is now with DigitalOcean.

If the ports are opened, I will retest the connection, send the poor message still waiting at zero percent, verify replies and Gmail’s Sent folder, and finish wiring the Gerris cockpit for sound. Then I will add Stainless Communications and Meritus Media as separate Alpine configurations and tmux windows.

If the request is declined, there are other options. I could move pineBox to a provider that permits authenticated SMTP submission. I could route outbound messages through an alternate service. I could even build a local sendmail-compatible helper that accepts Alpine’s finished MIME message and sends it through Gmail’s HTTPS API, bypassing SMTP entirely.

That last option is unnecessarily elaborate, faintly deranged, and very Unix.

I am trying not to hope DigitalOcean says no.

Why not just use Gmail?

Because I already do.

This is not a boycott of webmail. I am not claiming Alpine is objectively superior for every person or every message. Attachments, rich formatting, calendars, searches across years of mail, and administrative work will still sometimes send me back to a browser.

The mistake is assuming that because one tool can do everything, every interaction must happen inside it.

Most of my daily email work is text. Someone writes something. I read it. I write something back. For that small, ancient transaction, Alpine is extraordinarily good. It uses the whole terminal without filling it with interface. It responds immediately. It does not care which ThinkPad I am holding. Tmux preserves the exact state of the session between visits.

The browser treats each laptop as a separate place where I must reconstruct my working context. PineBox treats the laptops as windows into one place that already exists.

That is the same itch behind the Tachikoma network. My Documents do not belong to whichever computer happens to be open. They live in a continuous shared mind, and each ThinkPad temporarily embodies them. Now my active email environments can work the same way.

The Meshtastic node follows the same pattern from another direction. It became meaningful only when it stopped depending on my attention. It remains in the window, powered and present, doing its quiet job.

I keep building small systems that continue to exist when I am not looking at them.

The pleasure of exact overkill

Most consumer technology is designed around an imaginary average person. That is reasonable. Products need markets. Interfaces need conventions. Companies cannot build a bespoke metaphysical operating environment for every middle-aged man with seven ThinkPads and unresolved feelings about green monochrome displays.

Fortunately, Linux will.

My technical itches are rarely about acquiring capabilities I could not otherwise obtain. Dropbox can synchronize files. Gmail can display email. A phone can send messages over several globally maintained networks without a purple LoRa box in my window.

The itch is usually about arrangement.

I want to decide where state lives. I want one continuous environment available through many bodies. I want small machines performing narrow jobs well. I want systems that become boring because they are stable, not because I lost interest. I want to understand the pieces well enough that the result feels inhabited rather than rented.

PineBox costs four dollars per month. It runs on half a gigabyte of RAM. It solves a problem almost nobody has, using software whose interface would be familiar to a university student from thirty years ago, authenticated through security infrastructure that would have sounded like cyberpunk at the time.

It has already succeeded.

I can sit at a W520 built in 2011, open a Linux Mint terminal, type ssh pinebox, and enter an email cockpit that persists somewhere in New York whether that ThinkPad is awake or not.

Soon, Inshallah, it will send.

Until then, I can read six messages very efficiently.

That is too much fun.

It is also extremely stupid.

These are not opposing qualities.