Crossgrain No. 5: The United States of Strawman
In America today, the objection arrives before the claim is understood. The man who made the claim spends the rest of the conversation defending something he never said.
I have watched this since 2008, across tables and in the reviews my clients bring me. America teaches critical thinking with real conviction. Somewhere along the way it hardened into a reflex. The reflex looks for the point of attack before it looks at the claim. The man who asks a question, or asks for evidence, believes he has judged.
What was lost is the judgment that decides what to understand first, and then when to doubt, and what. Telling people to think more critically will not bring it back. I am not asking Americans to doubt less. They are already doing what they were told.
One day a professor of electrical engineering asked me how to invest in stocks. He said that whenever he bought a stock, it seemed to go down.
I told him he had to choose good water to fish in. By that I meant companies where a person’s study can bear fruit.
He asked how one finds such companies. I told him I had my own standard. I had used it for decades and had survived in the market by it. I look for companies around the world and keep only those that pass four screens, in this order.
First, the company has survived at least a hundred years.
Second, it has kept its original business through that time. What it did a hundred years ago, it still reasonably does.
Third, it has no record of doing serious harm to its minority shareholders like us.
Fourth, it has no record of serious punishment by the state.
The screen does not tell anyone to buy. It chooses what to study. A company that passes has generations of decisions on the record: what it did in good years and bad, how it treated the people who owned small pieces of it, how it behaved with the law watching. With that record, a human can study whether the company’s present trouble is temporary or structural, whether a boom is beginning or ending. Which of these companies will fail and which will keep doing well still has to be judged by us.
His answer came, instantly, “What, it didn’t go under in a hundred years, so it never will?”
I had not said that a hundred-year company would not fail. I had told him where to look. He answered a claim I had not made, and from that moment the conversation was about defending the number, 100.
I have thought about why. Nearly everything he had ever heard about stocks was about what to buy. I had told him what to study. He had never heard that kind of claim, so he heard the nearest one he had heard a thousand times, and answered that.
Consider what he could have said. He had asked because he did not know. He could have stayed first with what the number demands. A hundred years is a long time for any company to live. A hundred years without cheating its small shareholders, without serious punishment, without abandoning its business, is harder still. A professor at a leading university should have wondered how many such companies exist in the world, and found that interesting. He could have asked what the survivors have in common. Only then should he begin to doubt, if at all. He might then have asked: within that list, which companies are near their end, and which have another good century in them?
He did none of this. He did not judge whether my standard was right or wrong. He went straight in looking for a non-existent hole.
I told him, “You asked me for help. You’re the one who bought the meme bio stocks.”
A few weeks later, he apologized to me. He told me he had always done this. It came out before he knew it was coming.
He did not defend his objection but called it a reflex.
I wanted to know whether he was an exception. So I sought out people in America who are respected as top intellectuals: professors, physicians, people in policy, researchers at international organizations. I spent time and money doing it. Most American families own stocks, directly or through retirement accounts,1 so the subject opened doors.
A professor of political science heard the four screens and asked: Why a hundred years?
It was the same move in another shape. The number became the subject. If one hundred troubled him, he could have said what number would serve better, and why, or argued that a long record does not help a person study a company. What the number stands for is a record long enough to show generations of a business kept, shareholders not betrayed, the law not seriously broken. He asked me to defend the number and left the record alone.
A professor of medicine heard my account of the first two conversations. She defended the two professors. They had only done their job, she said, “A responsible intellectual must doubt.”
She knew that an intellectual must doubt, but not how. One had doubted a claim I never made. The other had doubted a number before asking what it stood for.
Later, a professor of physics told me I had not explained or proved why a hundred years mattered, so “How was the listener supposed to know? The author must prove it.”
Wrong. What it takes to survive a hundred years on those terms is not hidden. Anyone who stops for a second can see it. None of them spent even that second. They put the whole burden of understanding on the person who had offered help to them.
If professors read this way when they sit across from me with their names known, I can imagine how they read in private. I do not have to. Professors are among those I help, along with doctoral researchers and executives, and many come to me with a rejection. A paper or a proposal went out, and the reviews came back.
In those reviews I have seen the same reading I met across the table from these elite professors. The reviewer replaces the author’s claim with the nearest claim he has heard before, and attacks that. The newer the claim, the farther it is from anything he has heard, and the more surely it is replaced.
They ask why this number, why this case, where is more evidence, but state no judgment of their own about what evidence should have been presented. They do it anonymously, and the journals’ decision rests on what they wrote.
Once a work is rejected, those who come after inherit the conclusion without knowing whether the judgment behind it ever took place.
Over many years I have met many PhD students with the same attitude: go along with a rejection this once and get through; in a few years I will be the reviewer. The students do not accept the review because they believe it. They meet it because they must pass. In meeting it they learn where a review allows their ideas to be.
So the misreading is not easily exposed or corrected, and the authority of the top university professors passes to the next generation.
Writers adapt as well. They learn to make no claim that can be turned down.
Take a paper on the employment effects of the CHIPS Act by Erten, Stiglitz, and Verhoogen, in the Fall 2025 Brookings Papers on Economic Activity.2 “One key message of our study,” the authors write, “is that industrial policies can deliver measurable employment benefits in targeted strategic sectors, even in the short run” (p. 41). Their measured benefit is 15,000–16,000 direct and 30,000–45,000 total jobs (pp. 40–41). Yet between May 2021 and December 2022, the U.S. economy added 9.3 million.3 The act’s share is roughly 0.2 to 0.5 percent. A figure built to survive attack. Even this the authors fence: the national estimates “come with caveats” (p. 1), they “make no claim” that their scaling holds for government spending in general (p. 36), and they decline to argue that short-term employment is the best way to judge the act (p. 41). The reader cannot use a figure this small and this fenced in.
An intelligence that begins by looking for what is wrong in other people’s work lives on other people’s work.
Look at what each professor brought. Not one of them brought a judgment of their own on what I had said.
A real objection risks something. Had the political scientist said that eighty years would serve better, and why, he could have been wrong. Had the engineer said that old companies are harder to read than young ones, he could have been wrong. “Why a hundred years?” cannot be wrong because it says nothing.
This is why the habit spreads. A stated judgment can lose. The author answers it, and the two can be weighed. A question cannot lose. The author answers it, and the reviewer asks the next one. A reader learns quickly which of the two is safe, and the student watching him learns it too.
The cost falls on the author, and most of all on the author of something new. He can answer a judgment in a page. He cannot answer a question that asserts nothing, because it gives him nothing to answer. He can only go on defending, and keep adding evidence, since no one has said how much would do.
The trouble is that American doubt has been made free. An objection that risks nothing will outlast a claim that risks something, wherever no one keeps a record of either.
I write as someone who loves the United States of America, not the United States of Strawman. I want its next generation to read better than this one. I have one proposal.
When a research paper, a proposal, a strategy memo, or a policy plan is rejected in American academic and research review, keep it. Keep it exactly as submitted. Attach the reviews that rejected it exactly as written, with the date and the reviewer’s name. After ten years, publish the collection, and make it a text that the next generation of professors and reviewers must read.
The name is the center of it. A reviewer who rejects work should sign his reasons. A reviewer who rejects for “insufficient evidence” should say what evidence could have strengthened this argument and how much is missing. If he cannot say it, he has not judged the evidence. He has only declined to be persuaded. A question mark or a feeling of displeasure should not give anyone the power to demand endless defense from an author.
Take my four screens as a paper. Its claim has two parts. A long record of keeping the business, keeping faith with small shareholders, and keeping within the law lets a person’s study bear fruit. And a hundred years is long enough for that record to show. A reviewer who wanted to reject it would have to attack one of these parts, and argue something himself.
He could argue that the record misleads: that old companies carry some trait, inertia for instance, that hides whether their present trouble is temporary or structural, so that studying them bears less fruit than studying younger ones. Then he would have to show it.
Or he could argue that a hundred years throws away too much: that the same record shows itself often enough in companies of twenty or thirty years, and can be read as well there. Then he would have to show where.
Either of these would be a proper judgment on the part of the reviewer. “Why a hundred years?” is not a judgment. It is a misreading. A reader who asks in order to understand can ask freely. But asking does not give a reviewer the authority to reject and stop another person’s work. To exercise that authority, the reviewer must state his understanding of the claim and a judgment of his own.
Ten years, not at once.
A judgment needs time before it can be judged. In ten years the rejected work may have been carried out by someone else, or shown wrong, or quietly forgotten. The reviewer’s objection may then look exact, or it may look like an answer to a claim the author never made. Time shows which.
An immediate publication would make a review a performance. A reviewer afraid of being shown wrong might pass everything and praise everything to escape punishment. Publishing after a decade removes the immediate threat and keeps the record, to be read after both the reviewer and the field have moved on.
Governments seal their papers for a term and then open them,4 because some things cannot be said freely in their own time and should not be lost to the time after. Reviews are like that. What must be kept is what was said at the moment of judgment: the manuscript as submitted, the review as written, the name and the date. An explanation composed later by a reviewer who has learned how things turned out would teach nothing.
Most new work is weak, and most of it will have deserved rejection. That is part of why the reflex survives: a reviewer who refuses whatever is new will be right most of the time, without having read. A collection that kept only outcomes would teach the next generation to do the same. However, if the collection keeps the reasons, the reader ten years on can set the original work beside the actual objection and judge for himself.
Under this arrangement, a reviewer writes for two readers. One is the author in front of him. The other is a stranger ten years away who will read the manuscript first and the review second, and decide whether the reviewer understood what he refused.
The student changes too. A doctoral student today learns what his advisors concluded. But the student seldom sees how they read. A collection of rejected work with named reviews would let students study the reading itself. They could watch a reviewer turn an author’s claim into a different claim and attack that one, and see what it cost ten years later. They could also watch another reviewer name exactly what was missing in the rejected work and be proved right.
A rejected manuscript, the signed review that refused it, and ten years in which time can grade them both: that, students do not have. A few venues now publish their reviews, some for rejected work.5 But they publish them at once.
People may rot. A nation that keeps the record of its judgments will not rot through, because each generation can see where the last one went wrong. Thirty or forty years from now, someone will open what was refused in 2026 and read what was said against it, and build a better intelligence than ours.
For now, most judges of American thought leave no record that can be held against time. Where their reasons survive, their names are usually hidden.6
Make them leave one. Seal it for ten years, then open it, and let the next generation study it the way I study a company: from the first decision to the last, to learn what kind of judgment was exercised, and whether it can be trusted with what comes next.