This Isn't the Flex You Think It Is

There's a really good reason no one has ever done this before.

Not caution. Not red tape. Not some oversight that the bold finally found the nerve to fix. A reason, and a hard one: the kind that has stopped every faction that ever got close to the button. We're going to walk through it together, plainly, and by the end you'll be able to say the reason out loud yourself. That's the whole point. I don't want you to take my word for it. I want you to derive it.

So let's take our beliefs off the table for a minute, so we can see this thing clearly, for what it actually does and not for what the people pushing it right now think it might do.

First we have to agree on one thing, and it shouldn't be hard. We'd like our government to work, to get things done, and to keep the country ahead of an aggressive world. That last part runs on science. Think about every story we tell where something comes from the sky to conquer us. It never arrives with bronze-age tools. By definition it shows up with technology so far past ours that we can't answer it, and that is the engine of the whole horror. We are, all of us, quietly terrified of being out-teched by someone who means us harm. Staying ahead is the thing that has kept that story fiction. That is what the rule I'm about to describe puts at risk. Let me explain.

The steering wheel and the kill switch

The federal government has a proposed rule, Office of Management and Budget docket OMB-2026-0034, that changes who decides which research gets funded. For seventy years the answer was simple: scientists. Experts read the proposals, score them on whether the work is any good, and the government funds the best. Political officials set the priorities (fund more ocean research, more cancer research, more clean energy), but they don't overrule the experts on whether a given project is competent. The new rule changes that. It makes expert review merely "advisory" and hands political appointees the final say, including the power to cancel grants that are already underway if the work doesn't match the President's agenda.

Here is the distinction that matters, and it is the whole argument. Setting priorities is a steering wheel. Cancelling active, expert-approved work for being politically off-message is a kill switch. One of those, used wisely, drives the car. The other one just kills the car, and you can't un-kill it.

The people enacting this think they're installing a steering wheel. They are not. They are wiring a kill switch and pointing it at the engine.

You're not handing this to your team. You're handing it to the chair.

The fatal mistake is imagining this power belongs to you forever. It doesn't. You aren't handing the switch to the current administration. You're bolting it onto the office, and offices change hands. You had better win every election from now until the end of time, because the day you lose one, the other side inherits the exact same switch, and they remember.

But here is the part that should stop even the people most confident they'll never lose. It doesn't matter. Suppose they're right. Suppose one side holds power straight through to 2045 and never hands over the keys. Run that tape, their best case, and the country still loses:

The green line is what we'd have if we left the experts alone. The blue line is the rest of the world, and it does not wait for us to settle our politics. It climbs right past us early and never looks back. The two roads down are the point. The bold dark red line is the world where one party never loses, a smooth and uninterrupted slide. The faint one is the world where power keeps flipping, a staircase of cliffs. They end in exactly the same place.

That's the tell. The damage was never really about the handovers. The handovers are just one way to fall. The disease is pulling merit out as the test and dropping loyalty in its place, and that kills capacity whether power changes every four years or never changes at all. If anything, "never changes" is worse, because then there's no one left even trying to fix the mistakes, and they pile up, uncorrected, for good.

Why "breaks" isn't drama. It's arithmetic.

Destroying is instant. Rebuilding takes a generation. That asymmetry is the entire game.

Cancel a forty-year measurement record for four years and it isn't paused. It's gone. You can't go back and measure the years you skipped. Drive out a generation of young scientists (right now, in real surveys, three out of four U.S. researchers say they're considering leaving, with other countries waving half-billion-dollar welcome mats) and they don't trickle back when the mood changes. Teach the world that American grants can vanish for ideology, and every future partner prices that risk in forever, because they plan in decades and we would be promising them four years at a time.

The day nothing happens

Here is what the day this takes effect will look like to you: nothing. No buildings emptied, no sirens, no number on the news. That is the tell that it's working. The win is quiet, and it is Pyrrhic. Out where the public can see, nothing changes. Inside the laboratories, the corrosion starts that morning, and it runs deep.

It does not take many cancelled grants to change how everyone behaves. The moment scientists know that someone is now checking their work for political fit, they start trimming their own questions, softening their own findings, and steering around whole topics. Nobody had to order it. They saw what falling out of favor now costs, and they adjusted. That is the most efficient kind of control there is, because the people being controlled do the work themselves, for free.

And here is the part to be precise about, because it is worse than the word people reach for. This is not a purge. A purge has an end. You clear out the people you don't want, and then it's over. This never ends. It is a standing loyalty test, switched on for good. You can be the most respected scientist in your field, doing work nobody disputes, right up until the day your work brushes against the administration's goals, and then the very record that earned you respect becomes the file they use to end you. Everyone is on probation now, indefinitely, no matter how good they are.

We have run this experiment before

In 1933 another country decided its science had to pass an ideological test. About a quarter of its physicists were thrown out, including eleven past or future Nobel winners. The mathematics department at Göttingen, the best on Earth, built over more than a century, was hollowed out in under a decade. Asked later whether the math there had suffered, the old master David Hilbert said there was no mathematics in Göttingen anymore. It never came back. And the country that inherited all that fleeing genius, that became the world's science superpower on the strength of it? Us. America. We are now seriously discussing doing to ourselves the exact thing that handed us the crown, except this time the people standing downstream, arms open, are China and Europe.

The device we invented so that we'd never build it

In 1950, the physicist Leó Szilárd, one of the people who helped start the atomic age, went on the radio and described a new kind of bomb. Wrap an ordinary nuclear weapon in a jacket of cobalt, he said, and the blast would scatter radioactive dust so lethal and so long-lasting that it could, in principle, end human life everywhere. The "cobalt bomb."

Szilárd did not want anyone to build it. That was the opposite of his point. He described it precisely so that people would see, clearly, that we had become capable of building a thing that wins nothing and ends everything: a weapon with no possible victory in it, because the fallout comes for the side that pushed the button too, and it lingers for years after the war is "over." It was a thought experiment meant to be left on the shelf forever. No one has ever built one. That is the good reason no one does a thing like this. Not a rule against it, but the plain recognition that it has no winning move inside it.

This rule is Szilárd's cobalt bomb in administrative clothing. Not in scale, because a regulation is not a doomsday weapon, and anyone who says otherwise is exaggerating. But in logic, the shape is identical: a device that looks like power, that wins nothing, that cannot be aimed only at your enemies, and whose damage lingers long after the moment of "victory." It is the kind of thing we reason about precisely so that we understand why we must never do it. And someone has decided to take it down off the shelf and build it.

The part even a true believer should hear

Here is the test that should stop anyone, on any side. Say the mandated conclusion is correct. Take the strongest case: human-caused climate change. I'll be honest. I don't want it to be real. I spent a career in the field, and part of me would love to find the flaw. But it is real, and unlike an eighth-grader who is sure the world owes him what he wants, I don't get to have what I want. Thank God I don't. That is the whole point of the method.

The second you order scientists to treat even a true thing as beyond question, it stops being knowledge and becomes a loyalty oath. A result you're forbidden to test is just a correct-sounding sentence held for the wrong reason, and you've thrown away the only tool that could ever dislodge a wrong sentence later. Merit review isn't precious because it's always right. It is precious because it is the only machine we have for getting power to let go of an answer it likes but that happens to be false. Hand power the right to install the answers, and you've personally disarmed the thing that protects the truths you care about from the next person who hates them.

Why this is the opposite of a flex

So this isn't strength. It is the move of someone who cannot accept that reality will not hurry up and agree. The science won't bend fast enough to be satisfying, so there will be a few showy cancellations, and for a little while it will feel like winning. Then the bill comes due, and it is permanent, and it is ours alone, while the rest of the planet keeps walking right past us.

It is the same reasoning that stops a sane Senate from blowing up the filibuster even when it is maddening. Not because they're sure they'll win the next election, but because the move is irreversible and symmetric, and no adult bets the whole future on never losing. You don't build a doomsday device and trust that only your friends will ever stand near the button. Every generation tells itself the same warning, again and again. Prometheus and the stolen fire. Babel and the tower. Frankenstein and his creature. The machine built to win that no one can switch off. We are exactly the kind of creature that builds the thing that ends us, and the grown-up move is to step back from the brink while stepping back is still free.

It is still free. Barely.

What to do, before July 13

The comment window on OMB-2026-0034 closes July 13, 2026. Two things, and they each take ten minutes:

1. File a comment at regulations.gov on docket OMB-2026-0034. Say it in your own words: that you oppose giving political appointees the power to cancel expert-approved research, whoever holds office, because it is irreversible and it will end America's lead. Personal and specific beats a form letter every time.
2. Write your representatives, House and Senate, and say the same thing. Ask them to oppose the rule and to make expert merit review permanent in law.

One more thing, and then I'll let you go

You've read all the way down here, and you still can't tell me how I vote. That was not an accident. It is the whole point. Go back and look for the tell if you like. It isn't there, because the argument never needed it. Nothing in here asked you to love this administration or to hate it, to stand on the left or the right, to trust anyone or distrust anyone. A bad idea you can oppose without picking a side, without knowing or caring where the person next to you stands on anything else, is a rare and precious thing. It means you can throw your whole weight behind stopping it and never once wonder whether you've been recruited to a team.

So do that. Strenuously. Cleanly. Next to anyone who will stand with you.

And here is the reason, the one I promised you'd be able to say yourself by the end.

That is not a risk of the policy. That is the policy. It is why no one has ever done this before, in any country that wanted to stay a country worth being. Don't hand the next person the switch. Pull the wire out now, while pulling it out still costs us nothing.